Handling and communicating intelligence information: A conceptual,

historical and information design analysis

Dr David J. Lonsdale, PhD, is a Lecturer in War and Security Studies and the Director of the
Centre for Security Studies at the University of Hull, UK. He served on the Foreign and
Commonwealth Office’s Academic Network on Future Conflict and Cyber Security. He has
taught Intelligence Studies for 13 years. His publications include ‘Understanding Contemporary
Strategy’ and 'Intelligence Reform: Adapting to the Changing Security Environment’.

Dr Maria dos Santos Lonsdale, PhD, is an Associate Professor in Graphic and Communication
Design at the School of Design, University of Leeds, UK. She received her PhD in 2006 from the
Department of Typography and Graphic Communication, University of Reading, UK. Her main
areas of research are Information, Instructional, Typographic and Graphic Design. Specific
interests include Design for Security and the role Information Design plays in making security
and intelligence information clearer and more accessible in situations of time and performance
pressure. Her research is notable in the field of Design, as it involves user-centred research
methods and experimental studies to test design solutions to particular problems encountered
in real-life contexts (further supported by qualitative methods).

Funding details
This work was supported by the Office of the Chief Scientific Advisor (UK) under Grant 113900.

Copyright
This is an Accepted Manuscript of an article published by Taylor & Francis Group in Intelligence
and National Security journal. This article is protected by copyright, all rights reserved

Version: Accepted Version

1
Abstract

Effective communication of information is essential to intelligence work. This paper

identifies the main obstacles to good communication: policy-related challenges;
cognitive impediments; resource limitations; cultural and structural issues within
intelligence communities; and technical information. To illustrate, it examines four cases
when poor communication contributed to intelligence shortcomings. Via questionnaire
and document survey, the study identifies the current state of practice in UK intelligence
communities. The survey of visualization documents currently in use revealed errors
against established principles of Information Design. Thus, to ensure better handling
and dissemination of intelligence, there is a distinct need to apply Information Design
principles.

Keywords: intelligence information; intelligence cycle, information design, information

visualization, visual communication, dissemination

2
1. Introduction

‘The process of relaying intelligence can distort its meaning.’ 1

It seems self-evident to write that intelligence is about information. Nonetheless, it is important

to emphasize that the latter is integral to every stage of the Intelligence Cycle. The process of
national security intelligence is about knowing what information you need, how to collect it,
how to analyse it, and finally how to communicate it to those who use it in support of decision-
making. In the words of James S. Major, a forty-year veteran of intelligence, ‘In intelligence
work, information is our life’s blood.’ 2

And yet, the quest to better exploit national security information is hindered by a number of
factors. Some of the main impediments are: limited resources, time pressure, scale and
complexity of threat, organizational structure and culture, cognitive impediments (including
confirmation bias, groupthink, mirror imaging etc.), and decision-makers lack of engagement
with intelligence. These myriad challenges call for various responses. One area were advances
can be made is improving the communication of information, since poor communication can
exacerbate difficulties in the collation, analysis and dissemination stages of the intelligence
cycle.

In this way, communication is defined as ‘the imparting or exchange of information’, 3

operating on the ‘sender, message, receiver’ schema. 4 This includes how information is shared,
but also the form of presentation used for the message. The form of presentation, especially in
its ‘visual communication’ guise, is the focus of this paper. According to the IDA (Information
Design Association), for information to have impact, it must be easy to find, simple to use, and
instantly understandable. 5 That is, information needs to be designed. With this in mind, this
study looks at how information is shared in intelligence communities, with a particular focus on
the visual organization and display of information.

There is a small, but growing literature dedicated to the task of communicating national
security intelligence. 6 This interdisciplinary study, which brings together researchers from
Intelligence Studies and Information Design, makes an original contribution to this literature. It
was tasked specifically to assess how visualization of intelligence information can enhance the
effectiveness of communication. In this study, information visualization approaches in
intelligence communication include: infographics, charts, graphs, tables, pictograms, etc.;
where color, graphic and typographic features are of high relevance. Visualization contributes
to good communication by making information clear, concise, persuasive, actionable, and able
to be processed quickly and accurately. If done properly, good communication of information
should help mitigate the effects of mind-set and information overload; can improve receptivity
in policy makers, and help maximize the allocation of resources. The benefits of good
communication to intelligence work are described in the main body of the paper.

A broader aim of this study is to pioneer research that brings together the fields of information
design and intelligence studies, both in academia and the public sector. More specifically, this
study aims to answer the following research questions:
• How important is the effective communication of information in intelligence work?
• What are the obstacles to effective communication of information in intelligence work?

3
• Which principles of information design relate to the handling and communication of
intelligence information? (Principles are here interpreted as a set of guidelines/rules on
how to best use and organize graphics, typography, color, contrast, hierarchy, etc. to
make information clear and accessible. Further breakdown of some of these principles can
be seen in Table 1 in Section 5.2).
• How effectively is intelligence and security information handled and communicated in
departments within government that conduct intelligence analysis (‘intelligence
communities’).

To answer these research questions, this paper begins by identifying the significance of good
information communication in the work of intelligence. Then, on the basis of theoretical and
historical analysis, it identifies the main obstacles to good communication: policy-related
challenges; cognitive impediments; resource limitations (especially in relation to information
overload and time pressure); cultural and structural issues within intelligence communities; and
technical and specialist information. To provide an historical basis for the analysis, the paper
discusses four cases in which poor communication of information contributed to intelligence
shortcomings: Pearl Harbor, 9/11, the UK’s 45 minute claim on the Iraqi WMD programme, and
recent terror incidents in the UK. These represent cases in which information was not clearly
communicated, and thereby the message was not clearly received and understood by the
receiver. It is important to note that the objective of this paper is not to explain intelligence
failure per se. Rather, the focus of the paper is on the role of information communication and its
contribution to shortcomings and failures in intelligence. Moreover, because the literature on
intelligence information communication is so scarce, the case studies cited include
shortcomings in communication that relate to both sharing and forms of visual presentation.
The cases also cover communication issues at various stages of the intelligence cycle.

The work then discusses current research that was conducted with departments within
government that conduct intelligence analysis. Specifically, the researchers conducted a
questionnaire of existing staff and a document survey of existing forms of visual
communication within the UK intelligence communities. 7 Responses from the questionnaire
reveal the challenges of handling and visually communicating masses of information. The
document survey reveals significant room for improvement in the visualization of information.
The work concludes with an analysis of the contribution information design and visualization
can make to the communication of intelligence information.

2. Why Good Information Communication is Important

Despite the requirement for clear and effective visual communication of intelligence, there is
often a considerable gap between the intelligence that is communicated and the needs of the
user. 8 This is the case in relation to both current intelligence and long-range estimates. 9 This is
problematic, because as United States Intelligence Community Directive (ICD) 208 states,
analysts should provide consumers with intelligence in a form they can easily share and use. 10
The fact that this often does not happen is explained by Information-gap theory. This describes
a situation in which there exists a severe and consequential disparity between what one knows
and what one needs to know to make effective decisions. 11 Good visual communication can
help close the information-gap. There is evidence that receptivity improves when decision
makers are presented with information that is clear and actionable. 12 The obvious conclusion
to draw is that much commendable intelligence work is being wasted due to poor
communication. This is certainly one of the conclusions of the 9/11 Commission Report, which

4
identified communication problems within the intelligence community, but also
acknowledged the difficulty of making decision makers aware of the nature of the threat prior
to the attack. 13

The answer to an information-gap is not simply more information. Indeed, an increase in raw
quantity can exacerbate uncertainty through information overload. Rather, and bearing in
mind resource and time constraints, information needs to be communicated in a clear, concise
and understandable form to those who likely do not have extensive knowledge of intelligence
matters. 14 If analysts wish to provide effective decision aid to policy makers, then clear
communication is essential. 15 This is especially important in an increasingly complex security
environment, when policy makers and/or analysts have to make sense of, and prioritize
amongst, vast arrays of security challenges. 16 The history of the President’s Daily Brief (PDB) is
evidence of the importance of clear visual communication. From President Truman onwards,
the occupants of the White House have requested a clear format for the PDB. For example,
President Ford liked short sentences and President Carter requested clear visual hierarchy in
text. 17 These, and other communication techniques that are well established in information
design, should be applied to all intelligence products.

There is more, however, to good intelligence communication than simply making the
information clear and understandable. To avoid costly mistakes, intelligence reports must
include important information regarding the veracity of the intelligence. For example, products
should include reference to the limits of intelligence and the levels of certainty contained
within. This was a key recommendation of the Butler Review. 18 There are various methods to
communicate uncertainty. These include numeric and linguistic probabilities, but could also
take the form of visual cues, including the use of color, size, placement, etc. 19 Underlying
assumptions must also be evident in any report. 20 The troubled intelligence surrounding Iraq’s
Weapons of Mass Destruction programme was built upon a shared assumption (shared by all
major intelligence communities and United Nations weapons inspectors) that Iraq continued to
hold WMD ambitions and was reconstituting its programmes. In the case of Iraq, this
assumption was taken for granted and therefore not clearly identified and emphasized in the
intelligence products. As a result, the reliability of certain sources, including the HUMINT source
Curveball, was not questioned sufficiently.

The validity of sources is another area where effective visual communication is key. 21 Both
analysts and decision makers need an indication of source validity. As one intelligence
community report describes, intelligence agencies must apply the ‘show your work’ principle if
they wish to enhance decision maker engagement with their outputs. 22 From the analyst’s
perspective, details on sources are also important. In the case of Iraqi WMD, working from
IMINT sources, analysts assumed there was increased Iraqi activity around suspected WMD
facilities. In fact, the perceived increase was merely reflecting the fact that the satellites had
been reprogrammed to image the sites more regularly. This important source information was
not communicated to analysts. 23

When assessing how intelligence information is visually communicated, we are left with a
quandary. On the one hand, information must be concise, clear and accurate. 24 At the same
time, the information must contain a degree of background information, especially that
relating to levels of certainty, source validity, and underlying assumptions. These various
demands appear somewhat contradictory. How does one engage a busy decision maker or
analyst, without overwhelming them with qualifying information? The answer to this problem,
as detailed in this paper, is the use of effective information design, including a legible

5
typographic layout and strong visualization. The importance of this is evident when one
considers the myriad obstacles to communicating national security information.

3. Obstacles to Effective Information Communication

Amongst the many obstacles impinging upon the efficacy of information communication, five
are especially prominent in the discourse: policy-related challenges; cognitive impediments;
resource limitations (especially in relation to information overload and time pressure); cultural
and structural issues within intelligence communities; and technical and specialist information.
The ultimate test of the intelligence process is whether it makes a positive contribution to
national security decision-making. It is of significant concern, then, that engaging with policy
makers is fraught with difficulties. Since the scale of modern governance has increased
exponentially, policy makers are notoriously busy, with little time to dedicate to the many
issues they face on a day-to-day basis. 25 The 1991 National Intelligence Estimate (NIE) on the
collapse of the Former Yugoslavia evidences the challenge this poses to intelligence
communities. The 1991 NIE is noted for presenting an accurate and detailed prediction of
coming events in Yugoslavia. Nonetheless, policy makers at the time did not act upon the
information, partly because they were distracted with other issues and agendas. 26 One can
perhaps conclude that the intelligence contained in the NIE was not compelling enough in
what was a complex and demanding policy environment.

Additionally, intelligence products are but one source of policy-relevant information. Policy
makers have access to many sources of information beyond those provided by the intelligence
services. 27 These include, but are not limited to: increasingly prevalent open source material
(including social media), as well as information from diplomatic channels. It is, therefore,
increasingly difficult for the intelligence services to have their voice heard sufficiently to aid
decision making.

Even if policy makers take note of intelligence, there may be political reasons to ignore or
minimize the impact of intelligence outputs. Taking policy action always comes with political
costs, and may upset established agendas or political sensibilities. 28 In 1916, the British
government had general and detailed intelligence on the forthcoming Irish Uprising, but failed
to act upon it due to political concerns related to the implementation of Home Rule. 29 Likewise,
in relation to Iraq’s WMD programmes, the regime change agenda was so dominant in the
White House and Pentagon that it was extremely difficult for contrary intelligence to get a
hearing. As the Commission on the Intelligence Capabilities of the United States Regarding
Weapons of Mass Destruction concludes, ‘it is hard to deny the conclusion that intelligence
analysts worked in an environment that did not encourage scepticism about the conventional
wisdom.’ 30 In the face of such stubborn political obstacles, intelligence reporting must be
compelling (including visual impact) if it is to break through and enable a change in policy
momentum.

Information is analyzed by, and communicated to, human beings. As a consequence of this
truism, the processing of national security information is subject to various encumbering
cognitive, emotional and social obstacles. 31 These include so-called ‘mind-set’ issues such as
confirmation bias, mirror imaging, groupthink, etc. Although mind-set plays an important part
in analysis, by alerting analysts to patterns of behaviour, it also has played a negative role in the
history of intelligence receptivity. For example, the US failed to anticipate the 1958 Iraq
revolution because it relied too heavily on Iraqi government HUMINT sources, was too focused

6
on the communist threat in the country, and thus was unreceptive to information that went
against this narrative. 32 The fall of the Berlin Wall and the Yom Kippur War offer further
examples of foreign and security policy not reacting to information that challenged established
assumptions. 33 As veteran CIA analyst and writer on intelligence matters, Jack Davis, notes:
‘confirmation bias represents the inherent human mental condition of analysts to see more
vividly information that supports their mind-set.’ 34 Effective visualization of information can
help increase the vividness of divergent data, and thus can help reduce the effects of mind-set
bias.

Perhaps the greatest challenge facing contemporary intelligence communities is that of limited
resources in the face of growing and diverse security threats. When considered alongside the
requirement to process vast amount of data, often under time pressure, the resource issue
becomes a major obstacle to effective communication and effective analysis. Although the
Intelligence and Security Committee of Parliament (ISC) Report on the killing of Fusilier Lee
Rigby concludes that the attack could not have been reasonably prevented from an
intelligence perspective, it does identify the impact resource issues have on the processing and
communication of intelligence. Specifically, the report notes that important information
concerning the relationship between Michael Adebowale and an Al Qaeda Subject of Interest
(SoI) was not passed by GCHQ to MI5. This was a communication failure. In its response to the
ISC, GCHQ noted that ‘all three Agencies had been…working at very high tempo, long hours…
So I think at the bottom of this, we have got something that felt very low profile and you have
got an analyst who was unfamiliar with a lot of the targets and the detail of it that he was newly
tasked on…’ 35

Faced with limited resources and copious information to process, intelligence services
inevitably have to prioritize certain threats or SoIs. As evidenced by the GCHQ example above,
the value of certain pieces of information may not be fully appreciated, and the scale of
developing threats may be missed. In its comprehensive report, The 9/11 Commission also
identified the severe challenges that result from this state of affairs; ‘We sympathize with the
working level officers, drowning in information and trying to decide what is important or what
needs to be done.’ 36 Of course, part of the answer to the resource obstacle is an increase in
resources. However, working on the reasonable assumption that resources will never match
needs, we can look to better communication and visualization of information. As will be
discussed later, improved visual communication can help staff process information more
effectively.

Intelligence communities and policy-making bureaucracies are social institutions.

Consequently, they exhibit unique cultures that interact within complex institutional
structures. This can affect the methodology behind how national security information is
processed, communicated and used. For example, it is reported that the CIA favours collection
over analysis, thereby producing volume of information rather than deep interpretation of
data. 37 When interpretation is done, it tends to be deductive rather than inductive in nature.
Richard Heuer, one of the CIA’s foremost analytic methodologists, notes that ‘analysts typically
form a picture first and then select the pieces to fit.’ 38 As a result, the information that is
communicated within intelligence communities, as well as to decision makers, will often be
large in volume but bereft of the subtleties that reflect complex security realities. Again,
effective visualization can help manage volume of information and communicate complex
realities in a digestible form.

7
Problems with information communication are also the result of structural issues within
intelligence communities. 39 Information flow around communities was a significant problem
prior to 9/11. This is evident between the agencies and even within them. This was especially
problematic for the FBI, which had cultural problems with how it handled its intelligence
responsibilities. 40 In particular, it is said that the FBI is more an amalgam of 56 field offices than
it is a coherent, unified federal institution. 41 The attack on Pearl Harbor also highlights the
impact of structure on information communication. On the morning of the attack, the US Army
and Navy both spotted unusual Japanese military activity around the island. However, ‘little
communication between them meant that no one received both pieces of information
together prior to the attack.’ 42 Had these two pieces of information been considered together,
it is possible that the island would have gone on a heightened state of alert. More effective
communication of information, including visualization, can facilitate inter and intra-
institutional coordination and awareness.

The technical and specialized nature of some information creates a further obstacle to the
effective communication of intelligence. As noted by Brian Holmes and May Greenlee,
‘Scientific and technical intelligence analysts thus face the great challenge of quickly,
effectively, and clearly conveying information to policymakers.’ 43 As with all information,
technical intelligence must be communicated in a way that is accessible and actionable, but
also maintains accuracy. 44 The dangers associated with this area are well illustrated by the
example of aluminium tubes in the Iraqi WMD case. The UK intelligence community had
information that the tubes Iraq was attempting to procure would require substantial re-
engineering to make them suitable for a nuclear weapons programme. However, according to
the Butler Review, in the publicly released Iraq Dossier, ‘the JIC omitted the important
information about the need for substantial re-engineering of the aluminium tubes to make
them suitable for use as gas centrifuge rotors. This omission had the effect of materially
strengthening the impression that they may have been intended for a gas centrifuge and
hence for a nuclear programme.’ 45 A recent study, conducted by the Center for Strategic and
International Studies, found that when digesting technical information policy makers require
simple explanations, limited jargon and obvious relevance to policy. 46 The significance of good
communication and presentation is obvious. Specifically, well designed infographics can
communicate detailed, copious, complex and specialist information in an accessible and
compelling manner.

4. Examples of Communication Failure

The obstacles identified above, in conjunction with poor handling and communication of
information, can produce dire consequences. Too often, decision makers or analysts do not
have access to the information they require, or do not comprehend the significance of the
information available to them.

Pearl Harbor is a textbook example of this problem, and reveals a number of obstacles to
effective utilization of intelligence. Firstly, the senior American commanders in Hawaii –
Admiral Kimmel and US Army General Walter Short – suffered from severe cognitive rigidity
and mirror imaging. Their mind-set was such that they completely dismissed the possibility of
an attack on Hawaii. Thus, any information suggesting Japanese intent to attack the naval base
was dismissed under confirmation bias. This failing was exacerbated by the fact that decision
makers (military and political) in Washington distrusted Communications Intelligence
(COMINT), largely because it was a relatively new discipline and not properly understood. As a

8
consequence, indications of hostile Japanese intent, especially intercepted communications
between Tokyo and its Washington embassy, were ignored. 47

The intelligence agencies in Washington also had a role to play in the unfolding disaster about
to engulf Pearl Harbor. In October and November 1941, US COMINT had intercepted Japanese
communications requesting detailed knowledge on the placement of ships at Pearl Harbor.
This intelligence was never passed to commanders on Hawaii. 48 As noted, the confirmation
bias of Kimmel and Short suggests that such information may not have altered their
perspective on the vulnerability of Pearl Harbor. Nonetheless, had the information been
detailed, overwhelming, and presented clearly, it perhaps could have made a difference in
terms of alert preparedness. As Dahl notes, ‘receptivity is largely influenced by the level of
precision of the intelligence provided.’ 49 This brings us to the final problem with intelligence
communication prior to Pearl Harbor, and underlines the interplay of the different obstacles
discussed above. With limited resources available, US intelligence assets were focused on
Japanese and German diplomatic intercepts, rather than on the more detailed Japanese naval
codes that could have provided more detailed information on the attack plans.

9/11 provides another sobering example of how reasonably convincing intelligence

information was not able to adequately inform decision making, or indeed elements within the
intelligence community, of the developing Al Qaeda threat. This is best summed-up by the
9/11 Commission Report, when it describes how counterterrorism coordinator, Richard Clarke,
spent 12 years unsuccessfully trying to persuade Washington of the direct and growing danger
posed by Al Qaeda. 50 A number of examples relating to 9/11 illustrate the challenge of
communicating the Al Qaeda threat in the years prior to the attack. On Friday, December 4th,
1998, the President’s Daily Brief contained a reasonably lengthy piece detailing Al Qaeda plans
to highjack US airliners. Despite appearing in what is regarded as the most important
intelligence product of the US intelligence community, this intelligence warning had little
impact on President Clinton or Vice-President Gore. Less than a year later, Clinton stated: ‘I am
not aware of any specific threats against American airlines.’ Similarly, Vice-President Gore does
not recall any such warning being made. 51

The Bush administration, likewise, failed to recognize or act upon intelligence regarding the
threat posed by Al Qaeda. Most famously, the PDB of 6th August 2001 contained a section
titled ‘Bin Ladin Determined to Strike in the US’. Although the intention of the piece was to
highlight the current threat from Al Qaeda, President Bush regarded it as more historical in
nature. 52 Finally, in July 2001, the FBI and Counterterrorism Security Group (CSG) issued
warnings to federal agencies (including the Federal Aviation Authority) concerning the
growing risk of terrorist attack against US interests. Unfortunately, the warnings were not well
communicated. There was confusion over whether the threats were directed against targets in
the US, and also how much information could be disseminated and to whom. 53 Taken
together, these instances reveal that although there was ample intelligence regarding the
growing Al Qaeda threat, and indeed, the ‘system was blinking red’ in the summer of 2001, the
intelligence community had little impact on US counterterrorism policy. The intelligence was
there, it just was not having the desired effect on decision makers. Whilst there are various
reasons why decision makers may not register or act upon intelligence products, insufficiently
compelling or actionable communication of information is one of them.

Turning our attention to intelligence matters in the UK, we find similar problems. One of the
more infamous cases of poor intelligence communication is the 45 minute claim relating to
Iraq’s WMD. This piece of information, which concerns the time it would take Iraq to deploy

9
chemical and biological weapons to military units, appeared in a 9 September 2002 Joint
Intelligence Committee (JIC) Assessment and the 24 September publicly released government
dossier on Iraq. The importance of the 45 minute claim is that it strengthened the case for
action against Iraq, by giving the impression of an imminent threat to the UK from Iraq’s WMD
programme. Simply put, it strengthened the case for war. Moreover, in retrospect it damaged
the reputation of the intelligence services. As the author of the Iraq Dossier, the JIC became
embroiled in accusations of ‘sexing-up’ intelligence to support government policy on Iraq.

On closer inspection, the problem stems from poor communication of national security
information. The Assessments Staff of the JIC, who wrote the 9 September Assessment, based
their report on intelligence from MI6 and the Defense Intelligence Service (DIS). Crucially,
although this intelligence noted a time of 20-45 minutes for deployment of WMD, it did not
indicate the means of delivery this referred to. In giving evidence to the ISC, the Assessments
Staff admitted that in their judgement the deployment time most likely referred to battlefield
weapons, rather than Iraq’s longer-range ballistic missile programme. 54 This uncertainty on
delivery means, and the Assessment Staff’s subsequent judgement, were not reported in the 9
September Assessment, and subsequently failed to appear in the Iraq Dossier. Had this
limitation to the intelligence been made clear, the 45 minute claim may not have been so
prominent, and would not have caused the intelligence community such a post-war headache.
In its report on the matter, the ISC concludes: ‘Whilst the 9 September JIC Assessment was a
balanced assessment of scenarios, it did not highlight in the key judgements the uncertainties
and gaps in the UK’s knowledge about the Iraqi biological and chemical weapons.’ 55 Moreover,
the Butler Review notes that post-war investigations by MI6 have raised doubts concerning the
validity of the sources upon which the 45 minute claim was based. 56

The recent terror attacks in London and Manchester provide further examples of important
information not being effectively handled or communicated, with the result that its
significance was not fully appreciated. Referring to Salman Abedi, who perpetrated the
Manchester Arena attack, the Anderson Report concludes that ‘On two separate occasions in
the months prior to the attack, intelligence was received by MI5 whose significance was not
fully appreciated at the time … In retrospect, the intelligence can be seen to have been highly
relevant to the planned attack.’ 57 Although the report concludes that disruption of the attack
would have been unlikely, a better appreciation of the counterterrorism relevance of the
information would have opened a new investigation on Abedi, and perhaps subjected him to
questioning upon his return from Libya prior to the attack. 58

The Anderson Report also addresses the case of Youssef Zaghba, one of the three men who
perpetrated the attack on London Bridge. In this case, two information
handling/communication issues stand out. The Italian authorities had information that Zaghba
wanted to travel to Syria to join ISIS. In response, they placed Zaghba on the Schengen
Information System, an EU-wide system designed to share and coordinate information relating
to law enforcement and border control matters. Unfortunately, Zaghba was incorrectly given a
marker relating to serious crime, rather than one indicating a terrorist threat. In this sense,
information concerning the subject was incorrectly handled and communicated. Furthermore,
MI5 received a communication from Italian authorities requesting terror-related information on
Zaghba. Anderson reports that there is no evidence that MI5 responded to the information
request, nor that the Italian authorities followed-up the unanswered request. Again, we are left
with the impression that the significance of information relating to national security was not
fully appreciated, and was not well handled. Although the Anderson report acknowledges that
‘even if the intelligence had been actioned, it would have resulted in a nil return’. 59

10
Finally, we return to the killing of Fusilier Lee Rigby. As background, we have already noted the
significant resource constraints that impacted on this case. In relation to handling and
communicating national security information, a number of issues are raised by this case. Prior
to the attack, one of the perpetrators, Michael Adebolajo, had appeared in five different
intelligence operations. Although he was never identified as a key individual engaged in
planning an attack, the ISC raised concerns about how the intelligence services handle
information on recurring Subjects of Interest. 60 The ISC report also raises concerns about how
information was handled and recorded, especially surrounding Adebolajo’s trip to Kenya,
where he was arrested on the suspicion of attempting to travel to Somalia to engage in
terrorist activity. The same report identifies a ‘missed opportunity’ concerning important
information regarding the online activities of the second attacker, Michael Adebowale.
Specifically, analysts did not see information about online contacts between Adebowale and
another SoI. 61 Finally, the ISC report identifies that the way MI5 handled information did not
adequately reflect the developing narrative of a case. 62 Various measures have already been
instituted to address these concerns. Nonetheless, in this particular situation, effective
visualization of information could have been used to further improve how such information is
handled, presented, understood and utilized. As Adrian Wolfberg notes, complicated and/or
ambiguous intelligence information is more likely to be rejected or ignored. 63

5. Current Practice in the UK Intelligence Community

Building from this theoretical and historical basis, the researchers conducted a questionnaire of
staff and a visual survey of documents in the UK intelligence communities. A visual survey is a
research method that involves a comprehensive and systematic analysis of the graphic and
typographic features of a document on the basis of well-established guidelines, in this case
from the field of Information Design. This visual survey was concerned specifically with
documents communicating the allocation of resource decisions in intelligence communities.
The findings from both methodologies support the narrative that intelligence communities
deal with masses of information on complex security threats, experience information overload
and that current practices are not aligned with Information Design guidelines, meaning
documents do not always communicate intelligence information in the optimum way.

5.1. Findings of the Questionnaire

The questionnaire was initially devised online via Google Forms. However, due to internal
security, it was then printed and distributed internally to 30 people within intelligence
communities who deal with management-level resource allocation tasks. It contained 23
questions, as mentioned further below. Ten responses were received. 64 The respondents’
details were as follows. Gender: female 60%, male 40%; Age: 30-39 - 30%, 40-49 – 40%, 50-59,
30%; Education: undergraduate 60%, postgraduate (Masters) 10%, A-Level 10%, other 20%.

The objective of the survey was to ascertain how much information the respondents deal with,
how it is presented to them, whether or not it is presented in a clear and effective manner, and
how easy is it to make use of the information. The questionnaire is divided into four sections:
‘About You’, ‘Receiving Information Documents’, ‘Use of Information Documents’, and
‘Perceiving Information Documents’. In total, the respondents were asked 23 questions.

11
The responses to the survey clearly indicate that the respondents deal with large amounts of
information on a regular basis. Indeed, 100% of the respondents receive and process more than
20 information documents per week. All 100% of the survey population ‘Always’ (30%) or
‘Often’ (70%) process information under time pressure. This results in 40% of the respondents
indicating that they feel ‘Always’ or ‘Often’ overwhelmed by the amount of information they
receive (Figure 1). This fits with the findings from the literature and reports on intelligence
communities. As previously noted, there is a general consensus that intelligence communities
have to handle vast amounts of data, with limited resources. At times, this leads to staff being
overworked and potentially making errors in the handling of information.

100%
Always under time pressure 30%

Often under time pressure 70%

Rarely under time pressure 0%

Always overwhelmed 20%

Often overwhelmed 30%

Rarely overwhelmed 50%

Figure 1. Responses for (per 3 rows from top to bottom): 1) Do you read the information documents under some time
pressure? | 2) Do you ever feel overwhelmed by the amount of information in the documents you receive?

Of particular interest to this study, is the fact that 90% of respondents wanted the information
to contain some form of visualization. More specifically, 80% of respondents prefer the
information to be communicated in a combination of text and visuals (Figure 2). The
information design literature confirms that such a combined approach reduces cognitive load,
thereby maximising the potential for understanding, retention and recall. 65 As visualization is
becoming more prevalent and required within intelligence communities, it is essential that
infographics (a combination of text with visuals/graphics) play a major role, and that they are
constructed according to design principles.

12
 100%
Always opport. seek clarification 30%

Often opport. seek clarification 70%

Rarely opport. seek clarification 0%

Hard copy 20%

Digital 20%

Both 60%

Text only 10%

Visualization only 10%

A combination 80%

Figure 2. Responses for: 1) Do you have the opportunity to ask questions and/or seek clarification on the information
documents you receive? | 2) Do you prefer a hard copy of the document, a digital version or both? | 3) Would you prefer the
information to be presented as text only, visualization only, or a combination of both?

The information received by the respondents is used in a number of ways. As many as 90% of
the survey population specified that the information they receive both directly supports
decision-making and/or provides background data. In terms of what information they need,
60% require both detailed analysis and raw data; with 30% stipulating a more pressing need for
detailed analysis. All respondents specified that they ‘Often’ need to recall the information at a
later date. Finally, 60% of the population make notes on the documents they receive (Figure 3).

100%
Support decision making 10%

Provide background data 0%

Both 90%

Detailed analysis 30%

Both 60%

Other 10%

Always have to recall 0%

Often have to recall 100%

Rarely have to recall 0%

Always make notes 10%

Often make notes 50%

Rarely make notes 40%

Figure 3. Responses for: 1) Do the information documents you receive support decision making, provide background data, or
both? | 2) In your role, do you need raw data, detailed analysis, or both? | 3) Do you need to recall the information in the
documents at a later date? | 4) Do you make notes on the document?

13
These responses speak to different forms of information communication, and provide
important cues for creating appropriate infographics. For example, information used directly in
support of decision-making needs to be presented in a form that is appropriate for iconic
memory and working memory. Information usually remains in the iconic memory for less than
a second before being forwarded to the working memory. Therefore, this reinforces the
importance of grabbing the user’s attention at a glance with well designed information.
Working memory is maintained for about 5 to 15 seconds; a very limited time during which the
user is making sense of the information and either keeps the information or decides it is
irrelevant and ignores it. Therefore, well designed information that takes less time to make
sense of is also more likely to be kept and forwarded to long-term memory. 66 All in all, amongst
other things, this requires accessible, clear and legible information that can help the user make
sense of the information quickly.

Furthermore, the fact that the respondents make notes on the documents suggests a need for
generous margins and ample white space for printed information, and an appropriate note-
making facility in digital media. Historically, wide margins and ample white space have been
notable features of the President’s Daily Brief 67 and strongly supported by the Information
Design and Typography literature. 68

The questionnaire also sought the respondents’ perception of the information documents they
receive and use. For the most part, the survey population has a positive perception of the
documents. 80% consider the documents to be ‘Rarely’ confusing and unclear; 100% consider
them ‘Often’ to be clearly organized; 90% thought the relative importance of the information
was ‘Often’ clear; 80% that the relationships amongst the information was ‘Often’ clear; 80%
that the information was ‘Often’ presented in a consistent manner; and 80% thought the
documents had the right amount of information (Figure 4 and 5). These results appear to
contradict the findings of the visual survey of sample documents conducted for the study. As is
detailed below, nine documents used by the questionnaire population were surveyed, and all
were found to significantly breach a number of information design guidelines. That being the
case, the perception of the users should have been more negative.

100%
Always confusing 0%

Often confusing 20%

Rarely confusing 80%

Always clearly organized 0%

Often clearly organized 100%

Rarely clearly organized 0%

Always importance clear 0%

Often importance clear 90%

Rarely importance clear 10%

Figure 4. Responses for: 1) Overall, do you ever find the way information is presented in the documents confusing or unclear? |
2) Overall, is the information organized clearly on the document page?

14
 100%
Always relationships clear 0%

Often relationships clear 80%

Rarely relationships clear 20%

Always consistent 0%

Often consistent 80%

Rarely consistent 20%

Too much information 20%

Right amount of information 80%

Not enough information 0%

Figure 5. Responses for: 1) Overall, are the relationships amongst the information in the documents clear? | 2) Overall, is similar
information presented in a consistent manner in the documents? | 3) Overall, would you say documents have too much
information, the right amount, or not enough information?

How do we explain this discrepancy? Since the information design guidelines are so well
established, based on both research and practice, we can reject the notion that the guidelines
are in error. It is possible, therefore, that the survey population has low expectations of
information design, and is unaware of how much more effective information communication
can be when it follows well established information design principles. It is also worth noting
that 20% of the survey population have negative responses to the documents they receive. In
this sense, these users think the documents lack clarity, consistency and effective organization
of information. Moreover, the positive responses, which are ‘Often’ rather than ‘Always’, still
leave room for improvement. Methodologically, it is also possible that had a ‘Sometimes’
response been available, the frequency of the ‘Often’ response may have been reduced. This
would have reduced the positive nature of the responses.

Finally, the users were asked if the presentation of the information they received had ever
radically changed their perception of an issue. This speaks to the ‘impact’ of intelligence
information. The responses to this question were less positive than the other questions in the
Perception section. 40% indicated that the presentation of information ‘Often’ changed their
perception, whereas 60% responded ‘Rarely’ (Figure 6). This suggests that improved
visualization could increase the impact of information communication.

15
 100%
Always indication of reliability 40%

Often indication of reliability 50%

Rarely indication of reliability 10%

Always changed perception 0%

Often changed perception 40%

Rarely changed perception 60%

Figure 6. Responses for: 1) Does the information in the documents include an indication of its reliability? 2) Overall, has the
way information is presented in a document ever radically changed your perception of an issue?

In terms of population groups (age, gender, and level of education), a few notable outcomes
stand out, and are supported by research findings from the literature and from practice.
Perhaps the most striking result concerns level of education. All of the respondents who felt
overwhelmed (at times) by the information they receive, have a qualification below university
undergraduate. This group was also more likely to perceive the information as less clear and
poorly organized. This somewhat reflects research findings that non-professionals rate clarity of
presentation more highly than professionals. The former also prefer standard visualizations
over abstract visualization. 69 One possible explanation for these results is that those with a
university education have had more exposure to many different forms of data. This speaks to
the need to communicate information in a manner tailored to the user. Although half of the
women felt overwhelmed (at times) with the information, as a group they were completely
happy with the clarity and organization of the information documents. This contrasts
somewhat with men, who were much less happy with the clarity and consistency of the
information. This again supports research findings, which specify that women are more
comfortable with many different forms of information presentation, including non-graphic
data. 70

Finally, some differences were notable in age groups. Respondents in their 30s seemed most at
ease with the information they receive. None of this age group felt overwhelmed with the
information, and almost unanimously perceived the documents to be clear and well organized.
In contrast, all of those in the 40s age group felt overwhelmed by the information (at times).
50% of the 40s age group also had issues with the clarity and presentation of the information.
One third of the 50s age group felt overwhelmed by the information, and perceived the
documents to be unclear at times. The two thirds of the over 50s who were not overwhelmed
have a university level of education, either at undergraduate or postgraduate level.

5.2. Visual Survey of Documents

A total of nine documents were collected from intelligence communities, seven of which can
be considered an infographic (i.e. they combine text and visuals/graphics), and two a written
document as only text is used. All these documents are used for decision making and resource
allocation. The frequency with which they are sent to management staff in intelligence
communities varies between “as needed”, “every two weeks” and “every month”. One
document is produced around ten times a week. For security reasons, and since the focus was
only on visual features, the original classified content was changed to “Lorem Ipsum” text. 71

16
Due to the lack of research and literature on this subject matter (information design for security
material), the authors referred to research in the area of infographics (including text legibility,
color, visual elements, layout and structure), data visualization (as for infographics, plus the
different chart types) and risk information (including visualization of risk information and
intelligence reports). Five databases were used: Science Direct, EBSCO, Google Scholar, Sage
Journals, Taylor and Francis Online; and three design journals: Information Design Journal,
Visible Language, Infodesign. The selection of the databases and journals was based on their
relevancy for research focusing on information visualization. The search used specific keywords
such as: “infographics”, “data visualization”, “information visualization” “graphic risk
information”, and only focused on articles that were written in English and Portuguese (the
latter because Brazil is very strong in information design research). The articles gathered went
through two stages of examination and filtering process. In the first stage, the articles were
checked and filtered based on the titles and abstracts, and then organized in individual folders
according to sub-areas. In the second stage, the selected articles were reviewed in their entirety
through detailed reading and highlighting of relevant information (as well as making notes).
Approximately 230 papers were reviewed in stage 2. Due to the little research available in the
area of information design and visualization, approximately 40 books were reviewed, which
mainly offered practice-based tacit knowledge and recommendations.

From this survey of the literature, the researchers identified 486 information design guidelines.
These guidelines were then reduced to 75, i.e. to those guidelines that were applicable to the
nine documents surveyed. These guidelines therefore serve as a framework to conduct the
survey on the visual features of the nine documents currently used in intelligence
communities– see Table 1. The nine documents were selected to be representative of those
currently in use in the intelligence communities. For security reasons, the great majority of the
text was converted to ‘dummy text’, i.e. Lorem Ipsum text.

17
VISUAL FEATURES SURVEYED

General Text Graphics Color

• Uncluttered • Maximum 2 clearly different typefaces • Graphics used effectively • 3 colors for infographics
• Simple and easily accessible • Bold used sparingly for emphasis • Graphics adequately arranged • 3-5 colors for coding w/ equal strength
• Consistency across infographics • No underlining used for emphasis • Arrows provide direction effectively • 2-3 colors for charts
• Clear contrast between elements • No Italic used for descriptive text • Icons used to make data clearer • Contrast between text and background
• Information clearly organized • Title is dominant and clear • Color groups pieces of information
• Clear visual hierarchy • Clear contrast between text elements • Color helps to scan information
• Generous margins • Good interword space (and no rivers) • Color signals relationships
• Aesthetically pleasant • Left aligned text (not right, nor centered) • Color used sparingly for emphasis
• Design principles applied • All-capitals only for short headings • Color used harmoniously
• Target users considered • 9-12pt font size • No highly saturated (primary) colors
• Tested with target users • 60-70 characters per text line • Only colors that can be named are used
• 1-4 extra points interlinear space • Color used consistently
• Sanserif bold type on color background • Color contrast used for coding
• Sans serif for digital and small size type • Soft colors used for charts
• Effective direct labelling
• Text not inclined on labels
• Concise labels

Layout Tables Pie/donut charts Bubble charts

• Layout in zigzag form • Tables fade to background • Chart accompanied by text • Effective at giving general sense
• Grid used to organize information • White space used instead of borders • Effective direct labelling • Effective at comparing different values
• Information well organized • Light rules (and borders if used) • Multiple donut charts avoided • Visualization proportional to real data
• Clear hierarchy • Light filling colors used • Sum of parts equals 100% • Labels inside bubbles are effective
• Good spread of white space overall • Background does not alternate per row • No more than five segments • Outside labels clear and close to bubble
• Good space around graphic elements • Grid lines not used for every entry • No further segmenting within a slice • Colors for each category clearly distinct
• Good space around headings and text • Segments displayed clockwise • Semi-transparent overlapped bubbles
• Few and effective alignment of elements • Donut chart used to show total value
• Headings aligned in the same axes

Table 1. Framework with 75 information design guidelines used for the survey on the visual features of nine documents
currently used in intelligence communities.

The results of the survey can be seen next in the form of a heat map where: dark grey is used for
those features/guidelines that are not present in the respective document; medium grey for
when those features/guidelines are present; and light grey for when some features are partially
present, i.e. they might be in parts of the document but not in other parts. In simple terms, dark
grey is bad and medium grey is good. A section of the heat map (surveying general guidelines
only) is given as an example in Figure 7.

18
 MATERIALS SURVEYED

Pack 1 Pack 2 Stand alone docs

GENERAL GUIDELINES Doc 1 Doc 2 Doc 3 Doc 4 Doc 5 Doc 6 Doc 7 Doc 8 Doc 9

Uncluttered

Simple and easily accessible

Consistency accross document

Clear contrast between elements

Information clearly organized

Clear visual hierarchy

Generous margins

Aesthetically pleasant

Design principles applied

Target users considered

Tested with target users

Guidelines not followed Guidelines followed partially in the whole document

Guidelines followed

Figure 7. Heat map showing the general guidelines disregarded, followed and partially followed in the 9 documents surveyed.

The results show that overall the documents fail to communicate and visualize information in
an effective manner. As shown in Figure 8, only 26% of document features were in agreement
with the guidelines from Information Design. As many as 64% were not in agreement, and a
further 10% were only partially in agreement.

19
 100%

64%

26%

10%

Guidelines not followed

Guidelines followed

Guidelines followed partially

Figure 8. Overall document analysis

A detailed analysis was then conducted in order to identify the extent to which each document
fails to follow good principles of information communication and visualization. The findings are
very straightforward: the higher the amount of visualized information, the more problems
there are in terms of good principles of visualization. The same is true for those documents
containing mainly text, but where attempts were made to display the text in a table or inside
boxes (Figure 9, 10 and 11). Strikingly, some documents were as high as 98% in breach of
information design guidelines (Figure 10).

100%

76%
68%
60%

39%

31%
24.5%

1% 0.5% 1%

Guidelines not followed Guidelines not followed Guidelines not followed

Guidelines followed Guidelines followed Guidelines followed

Guidelines followed partially Guidelines followed partially Guidelines followed partially

Doc 1 Doc 2 Doc 3

Text only Text + visualization Mostly text + some visualization

Figure 9. Individual document analysis (information pack 1 containing 3 documents)

20
 100%
98%

54%

44%

1% 1% 2%

Guidelines not followed Guidelines not followed

Guidelines followed Guidelines followed

Guidelines followed partially Guidelines followed partially

Doc 4 Doc 5
Visualization only Mostly text + some visualization

Figure 10. Individual document analysis (information pack 2)

100%

97.5%

74%

60%
53%
47%

24% 25%

16%

2% 0.5% 1% 0%

Guidelines not followed Guidelines not followed Guidelines not followed Guidelines not followed

Guidelines followed Guidelines followed Guidelines followed Guidelines followed

Guidelines followed partially Guidelines followed partially Guidelines followed partially Guidelines followed partially

Doc 6 Doc 7 Doc 8 Doc 9

Tables (text only) + a bit of visualization Tables (text only) + visualization Table (mostly text and some visualization) Text only

Figure 11. Individual document analysis (stand-alone documents)

21
Although all pages are cluttered with information and visual elements, which in itself can result
in a disengaged user, there are more serious concerns than that. These documents are of
compulsory use, and staff read them under a certain amount of time pressure as shown in the
results of the questionnaire. Therefore, reading strategies such as scanning and skimming are
likely to be used as a starting point to have a general sense and then identify relevant
information. However, as shown by Lonsdale, lack of legibility in a document impairs speed and
accuracy of finding information in such circumstances. 72

Moreover, because these documents are used for decision making, the information not only
needs to be accessed and understood, but also recalled. Yet, humans can only store three or
four chunks of information at a time in our working memory, which is then forwarded into the
long-term memory. 73 However, if the cognitive load is high, as it will be with most documents
here surveyed, then recall is very likely to be impaired.

In conclusion, and based on the guidelines from Information Design, the questionnaire and
detailed visual survey strongly identifies the need to redesign the documents currently used in
intelligence communities. Failing to do so will mean continuing with a form of communication
that is clearly unsuitable within the context of intelligence information. This could lead to poor
allocation of resources and decision making, which history suggests can have serious
consequences.

6. The Contribution of Information Design and Visualization

This study has established that effective visual communication of information is important; that
there are many obstacles that stand in the way; that serious consequences can result from poor
communication; and that current practice in UK intelligence communities seems to breach well
established information design guidelines. The paper will now identify the many ways in which
good visualization can substantially enhance the communication and use of national security
intelligence. Specifically, it will be shown that visualization can highlight important
information; improve decision maker engagement with intelligence; provide more compelling
evidence; identify and clarify complex relationships amongst data; improve the accuracy of
intelligence communication; improve risk indicators in complex security threats; and facilitate
better understanding of complex technical information.

The Operational Improvement Review (OIR) was commissioned by the Commissioner of the
Metropolitan Police Service and the Director General of MI5 after the Westminster and
Manchester attacks. Its purpose was to identify and take forward improvements to processes
and capabilities for managing terrorism investigations. 74 It identified the need for better
handling and exploitation of data. As noted in the Anderson Report, due to a combination of
limited resources and information overload, staff in intelligence communities are constantly
having to make tough choices regarding what threats to prioritize. In these difficult
circumstances, ‘even marginal improvements are capable of paying dividends.’ 75 Strong and
effective visualization of data would facilitate better information management and
exploitation. Information, however compelling, does not speak convincingly for itself. As
Richard Betts notes, warnings are evidence filtered through perception. 76 Alongside the
cognitive impediments discussed above, perception is shaped, at least in part, by culture. The
latter speaks to us through stories and symbolism. 77 Visualization helps to limit the negative
effects of mind-set and culture, whilst at the same time making use of cultural symbolism to
engage the user of intelligence through familiar visual references. Arguably, the need for

22
visualization is increasing. As consumers of intelligence become more digital-savvy, they have
an expectation of enhanced visualization. 78 For example, a recent survey found that generals
serving in Iraq and Afghanistan wanted visuals that communicated complex and meaningful
information on the enemy in a digestible form. Simple wire-diagrams of enemy command and
control are no longer sufficient. 79

As noted in relation to the break-up of Yugoslavia, even accurate and detailed estimates may
not engage decision makers distracted by a range of other priorities. Moreover, and to
reiterate, Stephen Marrin reports that policy-makers often ignore intelligence that contradicts
their established policy agendas or assumptions. 80 In such challenging circumstances,
visualization provides possibilities for a more compelling and effective relationship between
decision makers and intelligence communities. 81 Visualization can more readily engage with
busy policy makers, and emphasize the importance of key information that might otherwise be
lost in a sea of data. In this sense, it can provide better decision aid. The history of the
President’s Daily Brief reveals how the visual presentation of information can improve
executive engagement with intelligence products. In his ground-breaking study, David Priess
notes how improvements in layout and visualization substantially increased the utility of the
PDB for Presidents Johnson and Carter. Indeed, the latter insisted on the inclusion of
visualization to improve his understanding of complex security matters. 82

The Second World War provides a compelling example of this argument. Following the attack
on Pearl Harbor, the US needed to know the location of the next major Japanese assault. With
limited resources, especially after the losses taken on Hawaii, the US fleet could not defend
every vulnerable point. The codebreakers in Naval Intelligence were convinced that Midway
was the next target. However, with cryptanalysis being such a new discipline, it was difficult to
persuade Nimitz, commander of US forces in the Pacific, and Washington of the veracity of the
intelligence emanating from Japanese decrypts. Exasperated by the lack of headway,
Lieutenant Commander Edwin Layton, an intelligence officer in the Naval Intelligence Unit
(Hypo), persuaded Nimitz to send his war plans officer, Captain McCormick, for a briefing.
There, in what we might call an early form of visualization, Layton laid out on a table all of the
key intelligence from decrypts in a manner that displayed the developing picture of Japanese
war plans. The effect on McCormick was overwhelming: ‘McCormick was fascinated. In the end,
he spent not two but three and a half hours poking around, flipping the material, asking a
thousand tough, show-me questions. Ultimately: McCormick came away completely
convinced, and to sell McCormick was to sell Nimitz.’ 83 As a result of this compelling
intelligence visual presentation, the US fleet was deployed to Midway. And, on 4-5 June 1942,
the US fleet inflicted a crushing defeat on the Japanese, including the loss of four precious
aircraft carriers.

Recent terror attacks in the UK have underlined the nature of the increasingly complex threat
environment. For much of their existence, the intelligence agencies dealt with state-based
threats, or reasonably well-defined and structured terror and insurgent threats (such as the IRA,
Al Qaeda and communist insurgencies in the colonies). In the contemporary setting, UK
national security is most immediately threatened by lone-wolf terrorists or very small
groupings of individuals. As is evident from the ISC and Anderson reports, this creates a
massive intelligence challenge, in which any one of thousands of Subjects of Interest (SoI)
could move rapidly to attack planning. In the face of this challenge, intelligence communities
need more effective risk indicators and means to better identify the relationships amongst data
on recurring SoI. 84 Part of the answer is more responsive databases. Visualization can also help
by more clearly communicating complex relationships and thresholds for deeper investigation.

23
Visualization can also help to improve the accuracy and level of detail of an intelligence
product. Various typographic and design techniques can be used to include background
information on sources, differentiate facts from opinions, and make the limits and uncertainties
of intelligence known to consumers. Some examples of these techniques are: color coding (to
help the user connect and rank information sets), hierarchy (ordering of graphic and
typographic information sets to help the user understand their relative importance), contrast
(small versus large, light versus bold/dark), legibility (clear, clutter free and well organized
graphics and typography), etc. 85 All of these factors are crucial when trying to build user trust
in intelligence. At the same time, busy and time-pressured decision makers should not be
overwhelmed by masses of background information and qualifying statements. Visualization
offers an effective shortcut to communicate such detail without masses of extra text.

Finally, visualization can help in those instances when technical information must be
communicated to non-specialists. Most executive consumers of intelligence fall into this
category, and intelligence must be produced with the end-user in mind. As Brian Nussbaum
notes in relation to the heavily specialized realm of cyber security, there is ‘a broader problem
of tailoring the products and deliveries to the customer’s level of technical understanding. To
overstate the importance of this tailoring and translation process is difficult.’ 86 One telling
example of when visualization made an important contribution is when Director of National
Intelligence, Mike McConnell, was looking to induce policy makers to support an amendment
to FISA to enable the NSA to intercept foreign communications passing through the US. In
doing so, McConnell used Verisign’s map of the world showing patterns and densities of
network bandwidth, 80% of which ran through the US. As Fred Kaplan notes, everyone who
heard and saw McConnell’s pitch supported the idea: ‘President Bush found a rationale for
action in the Verisign map’. 87 The result was the Protect America Act, which gave McConnell
exactly what he needed.

7. Conclusion

The challenges to more effective intelligence are many, and require several possible remedies:
Structured Analytic Techniques (SAT), more resources, restructured communities, more
oversight, etc. All of these initiatives, and more, have a role to play in improving the efficacy of
intelligence aid to decision makers. Additionally, as evidenced in this paper and in the OIR
commissioned by the Metropolitan Police Service and MI5, effective handling and exploitation
of information is an increasingly essential component of modern intelligence work. Since the
intelligence cycle is concerned with the flow and use of information, effective visual
communication is essential. However, this paper identified five primary obstacles to effective
information communication: policy-related challenges; cognitive impediments; resource
limitations (especially in relation to information overload and time pressure); cultural and
structural issues within the intelligence communities; and technical and specialist information.
The possible consequences of these difficulties are evidenced in examples of intelligence
shortcomings, including Pearl Harbor, 9/11, Iraq’s WMD programme, and recent terror attacks
in the UK. These examples all highlight the consequences of mishandling the communication
of information and the need to communicate information in an effective manner.

This study identified the challenges faced by staff in contemporary intelligence communities. In
particular, there is evidence of staff feeling overwhelmed by the amount of complex
information they need to process, and a desire for increased use of visualization to act as an aid.

24
As a result, it is evident that Information Design, and increasingly, effective visualization, are
indispensable to help manage large amounts of complex information. As noted, visualization
can highlight important information; improve decision maker engagement with intelligence;
provide more compelling evidence; identify and clarify complex relationships amongst data;
improve the accuracy of intelligence communication; improve risk indicators in complex
security threats; and facilitate better understanding of complex technical information

Yet, visualization can be done poorly. Indeed, as evidenced in the document survey, poorly
executed visualization is counterproductive. The survey of extant visualization documents
revealed errors against well-established principles of Information Design. In some cases, 98% of
document features breached Information Design guidelines. Thus, in the quest for better
handling and dissemination of intelligence, there is a distinct need to apply effective
Information Design principles to the world of intelligence. A second stage of this research will
go on to do exactly that, creating and testing new infographics and forms of visualization for
use within intelligence communities. In this way, by bringing together the disciplines of
Intelligence Studies and Information Design, the challenges to effective communication of
information can be met.

Bibliography

Aldrich, R. J. ‘Strategic culture as a constraint: intelligence analysis, memory and organizational

learning in the social sciences and history’. Intelligence and National Security 32:5 (2017): 625-
635.

Analyst-IC Associate Teams Program, ‘Probing the implications’. Studies in Intelligence 56:1
(2011).

Anderson, D. Attacks in London and Manchester March-June 2017: independent assessment of MI5
and police internal reviews (2017) Available online:
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/664682/Attac
ks_in_London_and_Manchester_Open_Report.pdf

Arslan, D. and Toy, E. ‘The visual problems of infographics’. Global Journal on Humanites & Social
Sciences (2015): 409-414, Available online: http://www.world-education-
center.org/index.php/pntsbs,

Bar-Joesph, U. and McDermott, R. ‘The intelligence analysis crisis’ in Johnson, L. K. (ed). The
Oxford handbook of national security intelligence. Oxford: Oxford University Press, 2010.

Barnes, A. ‘Making Intelligence Analysis More Intelligent: Using Numeric Probabilities’.

Intelligence and National Security 31:3 (2016): 327-344.

Ben-Haim, Y. ‘Policy neutrality and uncertainty: an info-gap perspective’. Intelligence and

National Security 31:7 (2016): 978-992.

Betts, R. K. ‘Surprise despite warning: why sudden attacks succeed.’, Political Science Quarterly
95:4 (1980-81).

Bringhurst, R. The elements of typographic style. Vancouver: Hartley & Marks, 1992.

25
Butler, R. Review of intelligence on weapons of mass destruction. London: The Stationary Office,
2004.

Carter, R. Maxa, S. Sanders, M. Meggs, P. B. and Day, B. Typographic design: Form and
communication (7th ed). New Jersey: John Wiley & Sons, 2018.

Cline, R. S. ‘Policy without intelligence’. Foreign Policy 17 (1974-5): 121-135.

Cowan, N. ‘The magical mystery four: How is working memory capacity limited, and why?’.
Current Directions in Psychological Science 19.1 (2010): 51–57.

Coyle, C. L. Malek, M. Mayse, C. Patil, V. and Shell, S. ‘Data can be beautiful: Crafting a
compelling story with SAS® Visual Analytics’. SAS Global Forum (2017).

Dahl, E. J. ‘Why won't they listen? Comparing receptivity toward intelligence at Pearl Harbor
and Midway’. Intelligence and National Security 28:1 (2013): 68-90.

Dahl, E. J. Intelligence and surprise attack: Failure and success from Pearl Harbor to 9/11 and
beyond. Washington, DC: Georgetown University Press, 2013.

Davis, J. ‘Why bad things happen to good analysts’. Studies in Intelligence 60:3 (2014): 16-17.

Dhami, M. K. ‘Towards an evidence-based approach to communicating uncertainty in

intelligence analysis’. Intelligence and National Security 33:2 (2018): 257-272.

Dhami, M. K. Mandel, D. R. Mellers B. A. and Tetlock, P. E. ‘Improving Intelligence Analysis with

Decision Science’. Perspectives on Psychological Science 10:6 (2015): 753-757.

Dikson, K. ‘7 mistakes to avoid when you create an infographic’ Venngage (2016), Available
online: https://venngage.com/blog/create-an-infographic/

Few, S. ‘Common mistakes in data presentation’. Perceptual Edge September 18 (2004).

Few, S. Show me the numbers. Burlingane, CA: Analytics Press, 2012.

Friedman J. A. and Zeckhauser, R. ‘Assessing uncertainty in intelligence’. Intelligence and

National Security 27:6 (2012): 824-847.

Friedman J. A. and Zeckhauser, R. ‘Handling and mishandling estimative probability: likelihood,

confidence, and the search for Bin Laden’. Intelligence and National Security 30:1 (2015): 77-99.

Hartley, J. ‘Designing instructional and informational text’ in Jonassen D. H. (ed) Handbook of

research on educational communications and technology (2nd Ed). Mahwah, New Jersey:
Lawrence Erlbaum Associates, 2004: 917-947.

Hollister Hedley, J. ‘Analysis for strategic intelligence’ in Johnson, L. K. (ed) Handbook of

intelligence studies. London: Routledge, 2009.

26
Intelligence and Security Committee. Iraqi weapons of mass destruction – intelligence and
assessments. London: HMSO, 2003.

Intelligence and Security Committee of Parliament. Report on the intelligence relating to the
murder of Fusilier Lee Rigby. London: HMSO, 2014).

Jackson, P. ‘On Uncertainty and the limits of intelligence’ in Johnson L. K. (ed). The Oxford
handbook of national security intelligence. Oxford: Oxford University Press, 2010.

Jervis, R. ‘Reports, politics, and intelligence failures: the case of Iraq’. The Journal of Strategic
Studies 29:1 (2006): 3-52.

Kaplan, F. Dark territory: the secret history of cyber war. New York: Simon & Schuster, 2016.

Karam, J. G. ‘Missing revolution: the American intelligence failure in Iraq, 1958’. Intelligence and
National Security 32:6 (2017): 693-709.

Lazard A. and Atkinson, L. ‘Putting environmental infographics center stage: The role of visuals
at the elaboration likelihood model’s critical point of persuasion’. Science Communication 37:1
(2015): 6-33.

Le, T. Reeder, B. Thompson, H. and Demiris, G. ‘Health providers’ perceptions of novel

approaches to visualizing integrated health information’. Methods of Information in Medicine 52
(2013): 250-258.

Lonsdale, M. D. S. ‘Does typographic design of examination materials affect performance?’.

Information Design Journal 15.2 (2007): 114-138.

Lonsdale, M. D. S. ‘Typographic features of text and their contribution to the legibility of

academic reading materials: An empirical study’. Visible Language 50:1 (2016): 79-111.

Lonsdale, M. D. S. ‘Typographic features of text: outcomes from research and practice’. Visible
Language 48:3 (2014): 29-67.

Lupton, E. Thinking with Type. New York: Princeton Architectural Press, 2010.

Lyra, K. T. Isotani, S. Reis, R. C. D. Marques, L. B. Pedro, L. Z. Jaques, P. A. and Bitencourt, I. I.

‘Infographics or graphics+text: Which material is best for robust learning?’. Proceedings of the
IEEE International Conference on Advanced Learning Technologies (ICALT) (2016).

Majooni, A. Masood, M. and Akhavan, A. ‘An eye-tracking study on the effect of infographic
structures on viewer’s comprehension and cognitive load’. Information Visualisation (2017): 1-
10.

Marrin, S. ‘Adding value to the intelligence product’ in Johnson, L. K. (ed). Handbook of

intelligence studies. London: Routledge, 2009.

Marrin, S. ‘Rethinking analytic politicization’. Intelligence and National Security 28:1 (2013): 32-
54.

27
Major, J. Communicating with intelligence: writing and briefing for national security (2nd Edition).
Lanham: Rowman and Littlefield, 2014.

McLean, R. The Thames and Hudson manual of typography. London: Thames and Hudson, 1980.

McDermott, R. and Bar-Joseph, U. ‘Pearl Harbor and Midway: the decisive influence of two men
on the outcomes’. Intelligence and National Security 31:7 (2016): 949-962.
Meador, C. L. & Cerf, V. G. ‘Rethinking the President’s Daily Intelligence Brief’. Studies in
Intelligence, 57:4 (2013): 1-14.

Murray, I. R. Murray, A. D. Wordie, S. J. Oliver, C. W. Murray, A. W. and Simpson, A. H. R. W.

‘Maximising the impact of your work using infographics’. Bone and Joint Research 6 (2017): 619–
620.

National Commission on Terrorist Attacks Upon the United States. Final report of the national
commission on terrorist attacks upon the United States (2004) Available online: https://www.9-
11commission.gov/report/911Report.pdf

Niebaum, K. Cunningham-Sabo, L. and Carroll, J. ‘Infographics: An innovative tool to capture

consumers’ attention’. Extension Journal 53:6 (2015): 1-6.

Nolte, W. M. ‘Intelligence analysis in an uncertain environment’ in L. K. Johnson (ed), The Oxford

handbook of national security intelligence. Oxford: Oxford University Press, 2010.

Nussbaum, B. H. ‘Communicating cyber intelligence to non-technical customers’. International

Journal of Intelligence and CounterIntelligence 30:4 (2017): 743-764.

Oxford Living Dictionaries, https://en.oxforddictionaries.com/definition/communication

Patterson, R. E. Blaha, L. M. Grinstein, G. G. Liggett, K. K. Kaveney, D. E. Sheldon, K. C. Havig. P. R.

and Moore, J. A. ‘A human cognition framework for information visualisation’. Computers &
Graphics 42 (2014): 42–58.

Petersen, M. ‘What I learned in 40 years of doing intelligence analysis for US foreign

policymakers’. Studies in Intelligence 55:1 (2011): 13-20.

Peterson, B. K. ‘Tables and Graphs Improve Reader Performance and Reader Reaction’.
International Journal of Business Communication 20:2 (1983): 47-55.

Pissierssens, ‘Revealing the scientific basis of graphical representation design’. Master of Science
in Business Engineering, Universiteit Gent (2017).

Priess, D. The President’s book of secrets: the untold story of intelligence briefings to America’s
presidents. New York: Public Affairs, 2016.

Quispel A. and Maes, A. ‘Would you prefer pie or cupcakes? Preferences for data visualisation
designs of professionals and laypeople in graphic design’. Journal of Visual Languages and
Computing 25 (2014): 107–116.

28
Silberman, L. H. and Robb, C. S. The commission on the intelligence capabilities of the United States
regarding weapons of mass destruction (2005) Available online:
http://govinfo.library.unt.edu/wmd/report/wmd_report.pdf p.11.

Simon, O. Introduction to typography. London: Faber & Faber, 1945.

Sless, D. Learning and Visual Communication. London: Croom Helm, 1981.

Sloan, G. ‘The British state and the Irish rebellion of 1916: an intelligence failure or a failure of
response?’. Intelligence and National Security 28:4 (2013): 453-494.

Spencer, H. The visible word. London: Lund Humphries, 1969.

Spiegelhalter, D. Pearson, M. and Short, I. ‘Visualizing uncertainty about the future’. Science 333
(2011): 1393-1400,

Stones, C. and Gent, M. 7 G.R.A.P.H.I.C. Principles of Public Health Infographic Design. Leeds:
University of Leeds, Public Health England, 2015.

Tetlan, L. and Marschalek, D. ‘How humans process visual information: A focused primer for
designing information’. Visible Language 50:3 (2016): 65-88,

Treverton G. F. and Miles, R. Unheeded warning of war: why policymakers ignored the 1990
Yugoslavia estimate. Washington, DC: Center for the Study of Intelligence Central Intelligence
Agency, 2015.

Tromblay, D. E. ‘Information Technology (IT) woes and intelligence agency failures: the Federal
Bureau of Investigation’s troubled IT evolution as a microcosm of a dysfunctional corporate
culture’. Intelligence and National Security 32:6 (2017): 817-832.

Tschichold, J. Asymmetric typography (trans. R. McLean). London: Faber & Faber Limited, 1967.

Wilder, D. C. ‘An educated consumer is our best customer’. Studies in Intelligence 55:2 (2011): 23-
31.

Wolfberg, A. ‘When generals consume intelligence: the problems that arise and how they solve
them’. Intelligence and National Security 32:4 (2017): 461-462.

Wohlstetter, R. Pearl Harbor: warning and decision. Stanford: Stanford University Press, 1962.

Zegart, A. ‘9/11 and the FBI: The organisational roots of failure’. Intelligence and National Security
22:2 (2007): 165-184.

29
1
Betts, ‘Surprise despite warning’, 555.

2
Major, Communicating with intelligence, xvii.

3
Oxford Living Dictionaries.

4
Sless, Learning and Visual Communication, 24.

5
Infodesign.org.uk

6
Major, Communicating with intelligence, 1-14.

7
The document survey and questionnaire were conducted within intelligence communities, but with no access to any

information that is not in the public domain.

8
Marrin, ‘Adding value’, 199.

9
Analyst-IC Associate Teams Program, ‘Probing the implications’, 2.

10
ibid, 3.

11
Ben-Haim, ‘Policy neutrality and uncertainty’, 982.

12
Dahl, ‘Why won't they listen?’, 68-90; Cline, ‘Policy without intelligence’, 121-135; Ben-Haim, ‘Policy neutrality’; Marrin,

‘Rethinking analytic politicization’, 32-54; Friedman and Zeckhauser, ‘Handling and mishandling’, 77-99.

13
National Commission on Terrorist Attacks Upon the United States, Final report, xvi.

14
Wilder, ‘An educated consumer’, 25; Hollister Hedley, ‘Analysis for strategic intelligence’, 216.

15
Marrin, ‘Rethinking analytic politicization’, p.4.

16
Meador & Cerf, ‘Rethinking the President’s Daily’, 5.

17
Priess, The President’s book..

18
Butler, Review of intelligence, 145.

19
Dhami, ‘Towards an evidence-based approach’, 257-272; Barnes, ‘Making Intelligence Analysis’, 327-344; Dhami, et al.

‘Improving Intelligence Analysis’, 753-757; Lupton, Thinking with Type; Stones and Gent, 7 G.R.A.P.H.I.C. Principles.

20
Jervis, ‘Reports, politics, and intelligence’, 24.

21
Friedman and Zeckhauser, ‘Assessing uncertainty’, 824-847.

22
Analyst-IC, ‘Probing the implications’, 5.

23
Jervis, ‘Reports, politics, and intelligence’.

24
Nussbaum, ‘Communicating cyber intelligence’, 751.

25
Petersen, ‘What I learned’.

26
Treverton and Miles, Unheeded warning.

27
Priess, The President’s book.

28
Betts, ‘Surprise despite warning’, p.91.

29
Sloan, ‘The British state’, 453-494.

30
Silberman and Robb, The commission, 11.

31
Wolfberg, ‘When generals consume intelligence’, 461-462.

30
32
Karam, ‘Missing revolution’, 693-709.

33
Davis, ‘Why bad things happen’, 16-17.

34
ibid, p.16.

35
Intelligence and Security Committee of Parliament, Report, 121-122.

36
National Commission on Terrorist Attacks Upon the United States, Final report, 355.

37
Davis, ‘Why bad things happen’, Bar-Joesph and McDermott, ‘The intelligence analysis crisis’, 363; Nolte, ‘Intelligence

analysis’, 413.

38
Marrin, ‘Rethinking analytic politicization’, p.729.

39
Jackson, ‘On Uncertainty’, 463.

40
Zegart, ‘9/11 and the FBI’.

41
Tromblay, ‘Information Technology (IT) woes’, 819.

42
McDermott and Bar-Joseph, ‘Pearl Harbor and Midway’, 954.

43
Nussbaum, ‘Communicating cyber intelligence’, p.746.

44
ibid, 750-751.

45
Butler, Review of intelligence, pp.133-134.

46
Nussbaum, ‘Communicating cyber intelligence’, 749.

47
Dahl, Intelligence and surprise attack.

48
Wohlstetter, Pearl Harbor; Dahl, Intelligence and surprise attack, McDermott & Bar-Joseph, ‘Pearl Harbor and Midway’.

49
Dahl, Intelligence and surprise attack, 70.

50
National Commission on Terrorist Attacks Upon the United States, Final report, 213.

51
Priess, The President’s book, 219.

52
National Commission on Terrorist Attacks Upon the United States, Final report, 260.

53
ibid, p.258.

54
Butler, Review of intelligence, p.126.

55
Intelligence and Security Committee. Iraqi weapons of mass destruction, 22.

56
Butler, Review of intelligence, p.127.

57
Anderson, Attacks in London and Manchester, 16.

58
ibid, 27.

59
ibid, 20-21.

60
Intelligence and Security Committee of Parliament. Report, 52.

61
ibid, p.95.

62
ibid, pp.66-67.

63
Wolfberg, ‘When generals consume intelligence’, 461.

64
The results of only ten respondents are used to present these findings. Although the responses cannot be construed

as a complete representative sample of all staff within intelligence communities, it gives a good indication of those

areas where information design can be used to enhance the communication of intelligence information.
31
65
Spiegelhalter, Pearson, and Short, ‘Visualizing uncertainty’, 1393-1400; Le, et al., ‘Health providers’ perceptions’, 250-

258; Lazard and Atkinson, ‘Putting environmental infographics’, 6-33; Niebaum, Cunningham-Sabo, and Carroll,

‘Infographics: An innovative’, 1-6; Lyra, et al., ‘Infographics or graphics+text’; Murray et al., ‘Maximising the impact’,

619–620.

66
Few, ‘Common mistakes’; Few, Show me the numbers; Le et al., ‘Health providers’ perceptions’; Patterson, et al., ‘A

human cognition framework’; Lyra et al., ‘Infographics or graphics+text’; Tetlan and Marschalek, ‘How humans

process’; Coyle, et al., ‘Data can be beautiful’; Majooni, Masood, and Akhavan, ‘An eye-tracking study’.

67
Priess, The President’s book.

68
Simon, Introduction to typography; Spencer, The visible word; Tschichold, Asymmetric typography; McLean, The Thames

and Hudson manual; Bringhurst, The elements; Hartley, ‘Designing instructional’; Lonsdale, ‘Typographic features of

text’; Arslan and Toy, ‘The visual problems’; Dikson, ‘7 mistakes to avoid’; Carter, et al., Typographic design.

69
Quispel and Maes, ‘Would you prefer pie’.

70
Peterson, ‘Tables and Graphs’.

71
Because context can be important in understanding the relationship between text and visuals, the general nature of

the documents was explained to the researchers, and hence the relationship between the text and visuals was

understood for the purposes of the study.

72
Lonsdale, ‘Does typographic design’; Lonsdale, ‘Typographic features of text: outcomes’; Lonsdale, ‘Typographic

features of text’.

73
Cowan, ‘The magical mystery four’; Patterson et al., ‘A human cognition framework’; Tetlan and Marschalek, ‘How

humans process’; Coyle et al., ‘Data can be beautiful’; Majooni et al., ‘An eye-tracking study’; Pissierssens, ‘Revealing

the scientific basis’.

74
Anderson, ‘Attacks in London’, 30.

75
ibid, pp.46-48.

76
Betts, ‘Surprise despite warning’.

77
Aldrich, ‘Strategic culture’, 631.

78
Meador & Cerf, ‘Rethinking the President’s Daily’, 3.

79
Wolfberg, ‘When Generals consume intelligence’, 467.

80
Marrin, ‘Rethinking analytic politicization’, 3.

81
Meador & Cerf, ‘Rethinking the President’s Daily’, 1.

82
Priess, The President’s Book, 47 & 127.

83
Dahl, ‘Why won’t they listen?’, 84.

84
Intelligence and Security Committee of Parliament, Report, 81.

85
Priess, The President’s Book; Butler, Review of Intelligence.

86
Nussbaum, ‘Communicating Cyber Intelligence’, p.746.

87
Kaplan, Dark territory, 193.

32