LEAN ANALYTICS 
LEANUX FEST 
#LEANUXFEST – SEP 16, 2014 
YVES FERKET - NASCOM 
Image Source: istrategy2009.com
I AM YVES FERKET 
HI! AND I WORK AT NASCOMIN BELGIUM.
I USED TO 
DO UX
But switched to...
The dark side...
Sales 
Analytics 
Analytics 
Excel 
Excel 
Timesheets 
coffee
The planet internet case
The planet internet case
InT 5 fihgurees: Planet Internet Portal Case 
• 1999 
• 2th biggest Belgian site 
• +60.000 visitors/day 
• +30.000 pages changed 
• 600 comments received
AbTout thhe ceomm ePnts:lanet Internet Portal Case 
• 6 likes… 
• Friends/Family/Colleagues 
• All the other mails & comments: 
• I can’t find my e-mail? 
• Where’s the SMS button? 
• You misplaced my favorite section? 
• Where is X? 
• I can’t find Y? 
• (multiply 594 times)
so: 
Lessons Learned (1999) 
• People don’t care about design 
• Don’t mess with top tasks 
• Measure before you change 
• 1999 isn’t always a party
So I stopped designing & started measuring users
Picture source: Daniel Dare https://www.flickr.com/photos/danieldare/6193235979/
Analytics 
some basics
<Google Analytics code inserted in HTML> 
Button 
Button 
website website with analytics 
<Javascript code 
on buttons to 
track events>
Done!
Building a Site 
@Nascom
A 
So where do we start Analytics? 
Design Develo 
p 
Discove 
r 
B 
Goals 
User Journeys 
Wireframes 
…. 
Art Direction 
Page Models 
Copy 
…. 
CMS 
HTML 
CSS 
….
A 
So where do we start Analytics? 
Design Develo 
p 
Discove 
r 
B 
Goals 
User Journeys 
Wireframes 
…. 
Art Direction 
Page Models 
Copy 
…. 
CMS 
HTML 
CSS 
….
A 
Analytics starts here 
Design Develo 
p 
Discove 
r 
B 
Goals 
User Journeys 
Wireframes 
…. 
Art Direction 
Page Models 
Copy 
…. 
CMS 
HTML 
CSS 
….
A 
Design Develo 
p 
Discove 
r 
B 
Goals 
User Journeys 
Wireframes 
…. 
Art Direction 
Page Models 
Copy 
…. 
CMS 
HTML 
CSS 
…. 
Not here
Image source: http://kustomculture.skynetblogs.be/archive/2007/06/27/cad-attack-1949-custom-cadillac.html 
Just slappin’ on Google Analytics code 
is putting a screwdriver on the hood of 
a car and expect it to fix the engine.
So, start with a set of Goals 
& Objectives
A project’s Goals & Objectives 
1. Define your goals 
2. Translate your Goals in “Human questions” 
3. Kpi’ify your human questions 
4. Create your data/triggers 
5. Create an actionable dashboard
What do you 
expect as an 
outcome? 
What would 
be a success 
factor? 
Why are we 
doing this?
This is something 
we do in every 
project. 
It’s clear, simple & 
agreed on. 
Distribute 100 
points to your goals 
By everybody
A project’s Goals & Objectives 
1. Define your goals 
2. Translate your Goals in “Human questions” 
3. Kpi’ify your human questions 
4. Create your data/triggers 
5. Create an actionable dashboard
Translate Goals 
in Human 
understood by 
everybody. 
Marketing 
Sales 
Management 
CEO 
Which human question would 
lead to your goal. 
E.g. Do people do x on our site?
Human 
Questions are a 
bit like a 
hypothesis 
"We believe [TYPE OF USER] has a 
problem [DOING THING]. We can help 
them with [OUR SOLUTION]. We'll know 
we're right if [CHANGE IN METRIC]."
A project’s Goals & Objectives 
1. Define your goals 
2. Translate your Goals in “Human questions” 
3. Kpi’ify your human questions 
4. Create your data/triggers 
5. Create an actionable dashboard
You know what 
would be a cool 
new feature? 
Shut up and 
explain why KPI 2 
is down by 12% If a big change 
in a KPI doesn’t 
cause 
somebody to 
pick up a phone 
or run to 
someone’s desk 
it’s not a good 
KPI
What 
figure , indicator or 
ratio would give 
an answer to the 
Human Questions 
?
A project’s Goals & Objectives 
1. Define your goals 
2. Translate your Goals in “Human questions” 
3. Kpi’ify your human questions 
4. Create your data/triggers 
5. Create an actionable dashboard
A project’s data/triggers 
Where is your data coming from? 
• online metrics? 
• exports from tools? 
• External partners (travel, ticket sales offices,...) 
• ... 
In what format is it available 
• document format: excel, MySQL, CSV, PDF, … 
• timing: live, hourly, daily, weekly, monthly, random, ...
A project’s data/triggers 
Triggers? 
Data triggers are events that generate data on the answer of the human 
questions 
• a click on a website 
• a money transfer in a sale 
• a scan of a bracelet 
• a click on a like button 
• a click on a buy button 
• result of a survey 
• ….
A project’s Goals & Objectives 
1. Define your goals 
2. Translate your Goals in “Human questions” 
3. Kpi’ify your human questions 
4. Create your data/triggers 
5. Create an actionable dashboard
THE DEFINITION: 
“AN INFORMATION RADIATOR IS A DISPLAY 
POSTED IN A PLACE WHERE PEOPLE CAN SEE IT 
AS THEY WORK OR WALK BY. IT SHOWS READERS 
INFORMATION THEY CARE ABOUT WITHOUT 
HAVING TO ASK ANYONE A QUESTION. THIS MEANS 
MORE COMMUNICATION WITH FEWER 
INTERRUPTIONS.
Dashboards
Image source: http://www.hotbutterstudio.com/
Sales in 
excel 
Google 
Reports 
Facebook Your data 
Insights 
Web 
Analytics 
Insights 
LinkedIn 
Stats 
YouTube 
Stats 
Pinterest 
Stats 
Twitter 
Stats 
Google+ 
Stats 
A calendar in 
Google Docs 
A survey 
result in a 
dropbox
Dashboard basics 
2 3 
Create (or connect to) 
a data source(s) with 
the data you need 
choose a graphic 
style that suits your 
data 
enrich data with 
indicators and info to 
tell a story 
Create “queries” on 
your data to fill your 
graph 
1 
3
Dashboard basics 
1. Is it quantifiable? 
Your metric should be measurable. 
2. Is it visible, understandable and clearly communicated? 
Your metric should serve as a reference point in conversations. 
3. Is it comparative? 
You should be able to compare the metric to benchmarks or other points in time. 
4. Is it a rate or ratio? 
Showing a rate of change is more illustrative than an absolute number. 
5. Does it change your behavior? 
If the metric you choose doesn’t enact change, you need a different one 
(Is it attributed to revenue/cost? ) 
Attributing your efforts to value for the business. 
It’s not a dashboard basic but will get you the budget to create one + will get CEO attention ;) 
Adapted Sources of : Kissmetrics, Ben Yoskovitz and Alistair Croll, Author of Lean Analytics
Dashboard example: 
human question: do they read our new content 
People who spend 
+180 secs grow from 
12%... 
… to 22% by changing 
content in 6 weeks 
1. Is it quantifiable? 
Your metric should be measurable. We created extra “time on page” 
segments In low (-30secs), medium (-180secs) and High (+180 secs)
Dashboard example: 
human question: how’s Facebook doing 
1: these Facebook Posts 
4: also in % 
change… 
2: created this 
effect 3: which is compared to the same 
previous period with indicators 
5: …and in in 
human terms 6: and explained. 
2: Is it visible and clearly communicated? 
Your metric should serve as a reference point in conversations.
Dashboard example: 
human question: how’s Facebook growth doing 
3: Is it comparative? 
You should be able to compare the metric to benchmarks or other points in 
time.
Dashboard example: 
human question: what social channel 
drives conversion? 
4: Is it a rate or ratio? 
Showing a rate of change is more illustrative than an absolute number.
Dashboard example: 
human question: which channels works best? 
Which affiliate 
partner or channel 
generated most 
conversion and 
should we invest 
in? 
5: Does it change your behavior? 
If the metric you choose doesn’t enact change, you need a 
different one
Dashboard example: 
human question: are we on campaign budget 
Daily budget automatically 
updated through API 
Budget goal from Campaign budget 
Google Doc 
(Is it attributed to revenue/cost? ) 
Attributing your efforts to value or cost for the business. 
We do get calls if we are in the red here ;)
“Understand the Customer 
Experience so that you can 
influence the optimal 
customer behavior that will 
lead to win-win outcomes” 
The Holy Analytics 
Trinity Outcome 
Orders/Leads/Revenue 
How, Why 
Conversion Rates 
Problem resolution 
Behavior 
Clickstream 
Click Density Analysis 
Segmentation 
Key Metrics, 
Search Intent 
Experience 
Research 
Customer Satisfaction 
A/B Testing 
User Tests 
Voice of customer 
Source: avinash kaushik
“Understand the Customer 
Experience so that you can 
influence the optimal 
customer behavior that will 
lead to win-win outcomes” 
The Holy Analytics 
Trinity Outcome 
Orders/Leads/Revenue 
How, Why 
Conversion Rates 
Problem resolution 
Behavior 
Clickstream 
Click Density Analysis 
Segmentation 
Key Metrics, 
Search Intent 
Experience 
Research 
Customer Satisfaction 
A/B Testing 
User Tests 
Voice of customer 
Analytics 
User 
research & 
tests, 
surveys 
Business 
data 
Source: avinash kaushik
“Understand the Customer 
Experience so that you can 
influence the optimal 
customer behavior that will 
lead to win-win outcomes” 
The Holy Analytics 
Trinity Outcome 
Orders/Leads/Revenue 
How, Why 
Conversion Rates 
Problem resolution 
Behavior 
Clickstream 
Click Density Analysis 
Segmentation 
Key Metrics, 
Search Intent 
Experience 
Research 
Customer Satisfaction 
A/B Testing 
User Tests 
Voice of customer 
Analytics 
User 
research & 
tests, 
surveys 
Business 
data 
DASHBOARD 
Source: avinash kaushik
That’s a lot of Analytics 
where’s the “lean” part?
A 
Analytics starts here 
Design Develo 
p 
Discove 
r 
B 
Goals 
User Journeys 
Wireframes 
…. 
Art Direction 
Page Models 
Copy 
…. 
CMS 
HTML 
CSS 
…. 
I re-use the 
wireframes of the 
UX-people and 
annotate them 
how the frontend 
people should 
place the triggers 
(because some 
companies need 
a metrics report )
Dashboards 
• This expensive looking dashboard cost 
about 20$/user/month 
• To create: 
• is more work up front 
• …but a lot less work then a monthly 
report 
• Plays really well with 
• ….
Dashboard data sources
Dashboard data sources
Remember 
these? 
"We believe [TYPE OF USER] has a 
problem [DOING THING]. We can help 
them with [OUR SOLUTION]. We'll know 
we're right if [CHANGE IN METRIC]."
"We believe [Ann] has a problem 
[using our product selector]. We can 
help them with [starting from moods vs 
products]. We'll know we're right if 
[she spends more time in the 
collection pages]." 
Can we test Ann? 
• 36 years old 
• Living in a city 
• Married ?
Facebook Ad 
Homepage 
V1 
Homepage 
V2 
1000 Clicks 
1000 Clicks 
Google 
Analytics 
Result 
validation
Facebook Ad
Homepage 
V1
Homepage 
V2
“Ann” who spend +180 
secs can grow from 
12%... 
… to 27% by a new 
homepage in 4 weeks 
Not real data from this customer, for explanatory purpose only 
Google 
Analytics
Test your results in http://isvalid.org/ 
Result 
validation
Create a PLANNING 
We don’t use expensive planning tools, 
but Google sheets 
• To make a planning for each 
channel, we can show in the 
dashboard 
• To pick up the changes, posts, 
events, e-mail lists… we created 
and pushed out in the dashboard 
• …and correlate with visitor data
Create a PLANNING 
• Pick up events from the Google 
Sheets Calendar…and correlate 
with visitor data
Consumer Satisfaction 
• We don’t use expensive customer 
satisfaction tools 
• We use Google Customer Survey’s 
• …or a simple Google Form that 
sends data to a Google sheet… 
• Which we pick up in our Dashboards
We don’t build Sites 
@Nascom
We do service design 
@Nascom
“We need a site” 
As seen by developers: 
The customer needs to go from A to B. 
Let’s build him a car
“We need a site” 
As seen by an agency: 
The customer needs to go from A to B. 
Let’s build him a fancy good looking car
“We need a site” 
As seen by an Nascom: 
The customer needs to go from A to B. 
Let’s see what his clients need. And give them a map. Learn 
them to drive. Provide a dashboard… 
And improve so we can get to point C
So where do we start Analytics? 
Start here 
Do not stop here. 
Keep measuring/improving
Measure to improve Slide 
1. Is it comparative? 
You should be able to compare the 
metric to benchmarks or other points in 
time. 
Source: Shana Rusonis, Content Marketeer, KissMetrics
We pick one KPI 
and improve on 
that. 
Go on to the next. 
Source: avinash kaushik
TWITTER ME: 
@NODESKTOPHERO 
LINKEDIN: 
WWW.LINKEDIN.COM/IN/YVESFERKET
Questions? 
NASCOM 
INFO@NASCOM.BE 
@NASCOM ON TWITTER

Lean Analytics & Analytics Dashboards

  • 1.
    LEAN ANALYTICS LEANUXFEST #LEANUXFEST – SEP 16, 2014 YVES FERKET - NASCOM Image Source: istrategy2009.com
  • 2.
    I AM YVESFERKET HI! AND I WORK AT NASCOMIN BELGIUM.
  • 4.
    I USED TO DO UX
  • 5.
  • 6.
  • 7.
    Sales Analytics Analytics Excel Excel Timesheets coffee
  • 8.
  • 9.
  • 10.
    InT 5 fihgurees:Planet Internet Portal Case • 1999 • 2th biggest Belgian site • +60.000 visitors/day • +30.000 pages changed • 600 comments received
  • 11.
    AbTout thhe ceommePnts:lanet Internet Portal Case • 6 likes… • Friends/Family/Colleagues • All the other mails & comments: • I can’t find my e-mail? • Where’s the SMS button? • You misplaced my favorite section? • Where is X? • I can’t find Y? • (multiply 594 times)
  • 12.
    so: Lessons Learned(1999) • People don’t care about design • Don’t mess with top tasks • Measure before you change • 1999 isn’t always a party
  • 13.
    So I stoppeddesigning & started measuring users
  • 14.
    Picture source: DanielDare https://www.flickr.com/photos/danieldare/6193235979/
  • 16.
  • 17.
    <Google Analytics codeinserted in HTML> Button Button website website with analytics <Javascript code on buttons to track events>
  • 18.
  • 19.
  • 20.
    A So wheredo we start Analytics? Design Develo p Discove r B Goals User Journeys Wireframes …. Art Direction Page Models Copy …. CMS HTML CSS ….
  • 21.
    A So wheredo we start Analytics? Design Develo p Discove r B Goals User Journeys Wireframes …. Art Direction Page Models Copy …. CMS HTML CSS ….
  • 22.
    A Analytics startshere Design Develo p Discove r B Goals User Journeys Wireframes …. Art Direction Page Models Copy …. CMS HTML CSS ….
  • 23.
    A Design Develo p Discove r B Goals User Journeys Wireframes …. Art Direction Page Models Copy …. CMS HTML CSS …. Not here
  • 24.
    Image source: http://kustomculture.skynetblogs.be/archive/2007/06/27/cad-attack-1949-custom-cadillac.html Just slappin’ on Google Analytics code is putting a screwdriver on the hood of a car and expect it to fix the engine.
  • 25.
    So, start witha set of Goals & Objectives
  • 26.
    A project’s Goals& Objectives 1. Define your goals 2. Translate your Goals in “Human questions” 3. Kpi’ify your human questions 4. Create your data/triggers 5. Create an actionable dashboard
  • 27.
    What do you expect as an outcome? What would be a success factor? Why are we doing this?
  • 28.
    This is something we do in every project. It’s clear, simple & agreed on. Distribute 100 points to your goals By everybody
  • 29.
    A project’s Goals& Objectives 1. Define your goals 2. Translate your Goals in “Human questions” 3. Kpi’ify your human questions 4. Create your data/triggers 5. Create an actionable dashboard
  • 30.
    Translate Goals inHuman understood by everybody. Marketing Sales Management CEO Which human question would lead to your goal. E.g. Do people do x on our site?
  • 31.
    Human Questions area bit like a hypothesis "We believe [TYPE OF USER] has a problem [DOING THING]. We can help them with [OUR SOLUTION]. We'll know we're right if [CHANGE IN METRIC]."
  • 32.
    A project’s Goals& Objectives 1. Define your goals 2. Translate your Goals in “Human questions” 3. Kpi’ify your human questions 4. Create your data/triggers 5. Create an actionable dashboard
  • 33.
    You know what would be a cool new feature? Shut up and explain why KPI 2 is down by 12% If a big change in a KPI doesn’t cause somebody to pick up a phone or run to someone’s desk it’s not a good KPI
  • 34.
    What figure ,indicator or ratio would give an answer to the Human Questions ?
  • 35.
    A project’s Goals& Objectives 1. Define your goals 2. Translate your Goals in “Human questions” 3. Kpi’ify your human questions 4. Create your data/triggers 5. Create an actionable dashboard
  • 36.
    A project’s data/triggers Where is your data coming from? • online metrics? • exports from tools? • External partners (travel, ticket sales offices,...) • ... In what format is it available • document format: excel, MySQL, CSV, PDF, … • timing: live, hourly, daily, weekly, monthly, random, ...
  • 37.
    A project’s data/triggers Triggers? Data triggers are events that generate data on the answer of the human questions • a click on a website • a money transfer in a sale • a scan of a bracelet • a click on a like button • a click on a buy button • result of a survey • ….
  • 38.
    A project’s Goals& Objectives 1. Define your goals 2. Translate your Goals in “Human questions” 3. Kpi’ify your human questions 4. Create your data/triggers 5. Create an actionable dashboard
  • 39.
    THE DEFINITION: “ANINFORMATION RADIATOR IS A DISPLAY POSTED IN A PLACE WHERE PEOPLE CAN SEE IT AS THEY WORK OR WALK BY. IT SHOWS READERS INFORMATION THEY CARE ABOUT WITHOUT HAVING TO ASK ANYONE A QUESTION. THIS MEANS MORE COMMUNICATION WITH FEWER INTERRUPTIONS.
  • 40.
  • 41.
  • 42.
    Sales in excel Google Reports Facebook Your data Insights Web Analytics Insights LinkedIn Stats YouTube Stats Pinterest Stats Twitter Stats Google+ Stats A calendar in Google Docs A survey result in a dropbox
  • 43.
    Dashboard basics 23 Create (or connect to) a data source(s) with the data you need choose a graphic style that suits your data enrich data with indicators and info to tell a story Create “queries” on your data to fill your graph 1 3
  • 44.
    Dashboard basics 1.Is it quantifiable? Your metric should be measurable. 2. Is it visible, understandable and clearly communicated? Your metric should serve as a reference point in conversations. 3. Is it comparative? You should be able to compare the metric to benchmarks or other points in time. 4. Is it a rate or ratio? Showing a rate of change is more illustrative than an absolute number. 5. Does it change your behavior? If the metric you choose doesn’t enact change, you need a different one (Is it attributed to revenue/cost? ) Attributing your efforts to value for the business. It’s not a dashboard basic but will get you the budget to create one + will get CEO attention ;) Adapted Sources of : Kissmetrics, Ben Yoskovitz and Alistair Croll, Author of Lean Analytics
  • 45.
    Dashboard example: humanquestion: do they read our new content People who spend +180 secs grow from 12%... … to 22% by changing content in 6 weeks 1. Is it quantifiable? Your metric should be measurable. We created extra “time on page” segments In low (-30secs), medium (-180secs) and High (+180 secs)
  • 46.
    Dashboard example: humanquestion: how’s Facebook doing 1: these Facebook Posts 4: also in % change… 2: created this effect 3: which is compared to the same previous period with indicators 5: …and in in human terms 6: and explained. 2: Is it visible and clearly communicated? Your metric should serve as a reference point in conversations.
  • 47.
    Dashboard example: humanquestion: how’s Facebook growth doing 3: Is it comparative? You should be able to compare the metric to benchmarks or other points in time.
  • 48.
    Dashboard example: humanquestion: what social channel drives conversion? 4: Is it a rate or ratio? Showing a rate of change is more illustrative than an absolute number.
  • 49.
    Dashboard example: humanquestion: which channels works best? Which affiliate partner or channel generated most conversion and should we invest in? 5: Does it change your behavior? If the metric you choose doesn’t enact change, you need a different one
  • 50.
    Dashboard example: humanquestion: are we on campaign budget Daily budget automatically updated through API Budget goal from Campaign budget Google Doc (Is it attributed to revenue/cost? ) Attributing your efforts to value or cost for the business. We do get calls if we are in the red here ;)
  • 51.
    “Understand the Customer Experience so that you can influence the optimal customer behavior that will lead to win-win outcomes” The Holy Analytics Trinity Outcome Orders/Leads/Revenue How, Why Conversion Rates Problem resolution Behavior Clickstream Click Density Analysis Segmentation Key Metrics, Search Intent Experience Research Customer Satisfaction A/B Testing User Tests Voice of customer Source: avinash kaushik
  • 52.
    “Understand the Customer Experience so that you can influence the optimal customer behavior that will lead to win-win outcomes” The Holy Analytics Trinity Outcome Orders/Leads/Revenue How, Why Conversion Rates Problem resolution Behavior Clickstream Click Density Analysis Segmentation Key Metrics, Search Intent Experience Research Customer Satisfaction A/B Testing User Tests Voice of customer Analytics User research & tests, surveys Business data Source: avinash kaushik
  • 53.
    “Understand the Customer Experience so that you can influence the optimal customer behavior that will lead to win-win outcomes” The Holy Analytics Trinity Outcome Orders/Leads/Revenue How, Why Conversion Rates Problem resolution Behavior Clickstream Click Density Analysis Segmentation Key Metrics, Search Intent Experience Research Customer Satisfaction A/B Testing User Tests Voice of customer Analytics User research & tests, surveys Business data DASHBOARD Source: avinash kaushik
  • 54.
    That’s a lotof Analytics where’s the “lean” part?
  • 55.
    A Analytics startshere Design Develo p Discove r B Goals User Journeys Wireframes …. Art Direction Page Models Copy …. CMS HTML CSS …. I re-use the wireframes of the UX-people and annotate them how the frontend people should place the triggers (because some companies need a metrics report )
  • 56.
    Dashboards • Thisexpensive looking dashboard cost about 20$/user/month • To create: • is more work up front • …but a lot less work then a monthly report • Plays really well with • ….
  • 57.
  • 58.
  • 59.
    Remember these? "Webelieve [TYPE OF USER] has a problem [DOING THING]. We can help them with [OUR SOLUTION]. We'll know we're right if [CHANGE IN METRIC]."
  • 60.
    "We believe [Ann]has a problem [using our product selector]. We can help them with [starting from moods vs products]. We'll know we're right if [she spends more time in the collection pages]." Can we test Ann? • 36 years old • Living in a city • Married ?
  • 62.
    Facebook Ad Homepage V1 Homepage V2 1000 Clicks 1000 Clicks Google Analytics Result validation
  • 63.
  • 64.
  • 65.
  • 66.
    “Ann” who spend+180 secs can grow from 12%... … to 27% by a new homepage in 4 weeks Not real data from this customer, for explanatory purpose only Google Analytics
  • 67.
    Test your resultsin http://isvalid.org/ Result validation
  • 68.
    Create a PLANNING We don’t use expensive planning tools, but Google sheets • To make a planning for each channel, we can show in the dashboard • To pick up the changes, posts, events, e-mail lists… we created and pushed out in the dashboard • …and correlate with visitor data
  • 69.
    Create a PLANNING • Pick up events from the Google Sheets Calendar…and correlate with visitor data
  • 70.
    Consumer Satisfaction •We don’t use expensive customer satisfaction tools • We use Google Customer Survey’s • …or a simple Google Form that sends data to a Google sheet… • Which we pick up in our Dashboards
  • 71.
    We don’t buildSites @Nascom
  • 72.
    We do servicedesign @Nascom
  • 73.
    “We need asite” As seen by developers: The customer needs to go from A to B. Let’s build him a car
  • 74.
    “We need asite” As seen by an agency: The customer needs to go from A to B. Let’s build him a fancy good looking car
  • 75.
    “We need asite” As seen by an Nascom: The customer needs to go from A to B. Let’s see what his clients need. And give them a map. Learn them to drive. Provide a dashboard… And improve so we can get to point C
  • 76.
    So where dowe start Analytics? Start here Do not stop here. Keep measuring/improving
  • 77.
    Measure to improveSlide 1. Is it comparative? You should be able to compare the metric to benchmarks or other points in time. Source: Shana Rusonis, Content Marketeer, KissMetrics
  • 78.
    We pick oneKPI and improve on that. Go on to the next. Source: avinash kaushik
  • 79.
    TWITTER ME: @NODESKTOPHERO LINKEDIN: WWW.LINKEDIN.COM/IN/YVESFERKET
  • 80.

Editor's Notes

  • #3 Freelancer @Nascom
  • #4 I’m a productdesiger…& started designing sites around 95’ish
  • #5 Design sites, do wireframes…
  • #8 No more design…So why this sudden switch?
  • #9 Because of this site: An ugly Belgian Portal site in 1999
  • #10 Which we redesigned to this (still 1999 …don’t expect to much, almost no CSS or anything. We used HTML tables)
  • #11 This is what we did with the UX team. . I expected cheering in the streets and spontaneous parties breaking out with people dancing & padding each other on the back.
  • #12 This was the result.
  • #14 Eyetracking. Pré google analytics era
  • #15 And overnight my live changed from the rockstar live of a designer
  • #16 To a web analytics nerd. I admit : Ilove excel.
  • #18 You measure pages viewed, you measure events like clicks
  • #19 And you get reports. Who looks at any analytics on a regular basis?
  • #20 So this is more or less a general site building approach
  • #22 In the beginning? In the end? Who says after the code goes live?
  • #28 Some of Us did this yesterday. Get Marketing Sales & CEO in the room for this
  • #29 And here, sales & marketing start fighting…. Example: a lingerie site
  • #31 Put the collection in the picture Do they find their way to the collection What’s most popular… Some of them are easy to answer…some of them don’t.
  • #32 It’s a bit like this….
  • #33 KEEPEEai-IFy
  • #40 Alistair Cockburn in his book “crystal clear: a human-powered methodology for small teams”
  • #41 So because Google analytics alone will not do…we create dashboards
  • #46  ----- Meeting Notes (16/09/14 12:57) ----- Content groups can help to determine content vs functional pages
  • #48 Not a big fan of Facebookfan counting, but it serves the purpose
  • #49 Traffic over conversion ratio = cTR
  • #50 We shifted budget from some big tourist sites to niche players.
  • #52 So what should you strive for? Understand the Customer Experience so that you can influence the optimal customer behavior that will lead to win-win outcomes”
  • #53 Understand the Customer Experience so that you can influence the optimal customer behavior that will lead to win-win outcomes”
  • #54 The dashboard doesn’t give you god powers…but at least you know what you created is doing..
  • #56 Some developers demand a metrics guide. The more developers working on it, the thicker they need the guide. It is useful for tagging or front enders
  • #59 Measure Social, Site en campaigns through the reporting API
  • #62 Stakeholders Pains en Gains Onderscheidende doelgroepen krijgen een gezicht
  • #63 We can run a facebook ad for 2K clicks (you may use less) to an A/B test (for Ann’s only). Measure in GA, validate afterwards
  • #64 Age 33-39 Women Married Living in Antwerp-Hasselt-Gent-Brussels,… Even can put in education
  • #65 A version (control)
  • #66 So would have used a bank case study if I had to speak in the morning… I circled the thing you have to look at… This is the B version
  • #67 We track the time on pages for Ann’s coming in through facebook segment
  • #69 Interne en external stakeholders, partners can access it. Garbage in is garbage out
  • #70 Team pic didnt
  • #78 Improve: teest + measure