Neal Cole

GEO Bullshit Jargon Translator - episode 7 - "Personalisation" 🔵 What GEO 'experts' say: "AI search is different to traditional search because next-gen AI search personalises answers to each user." 🔵 What GEO 'experts' mean: "You and I could type the same thing, and the machine will serve us different answers based on what we've done before." 🔵 What SEOs have been saying for years: "User intent matters. Previous behaviour affects search results. Things such as a searcher's location, browser, how often they visit a site, and more, can affect search results. The same keyword can mean ten different things depending on the person. That's why you don't write one-size-fits-all content." 🔎 SEO, innit?

54

6 Comments

GEO Optimisation Lesson 2: Optimise for Performance Attributes Yesterday, I explained how GEO was based around the optimisation of a website for its business/ brand positioning. Today, it's all about the performance attributes of products or services. This is part of a positioning strategy again. Take this response to my query in the image: "This Deansgate steakhouse is consistently praised for its luxury Sunday roast, featuring slow-roasted rump with beef dripping potatoes, Yorkshire puddings, and rich bone marrow gravy". Or this: "Blacklock Manchester Tucked away in a subterranean former textile warehouse on Peter Street, this chophouse offers a celebrated roast with your choice of beef rump, lamb leg, or pork loin cooked over a charcoal grill" And yes, this is copywriting. For the SEO or GEO a large part of the optimisation process is understanding what a business's performance attributes are. And coming up with ways to not only display them on a site but also to create 'mutual information' around them. Take wheelchair access. Does your business have a dedicated page or a clear section on your website that focuses on this? Does that page discuss vehicle access, parking options, and other relevant details? Or what about allergy information? Some restaurants use the same fryers for various items, which can cause issues for individuals with allergies. Is this information displayed clearly? Or do you go out of your way to cater for those concerns? This type of optimisation might seem obvious to some and less to others...but it's absolutely part of GEO, not SEO. Why? As AI search usage increases (and people become more familiar with it), the power shifts from those with the largest SEO or paid search budgets to those who are the best matches. Now, this is how SEO should always have been done. And it's actually what Google have always told SEOs to do. But the industrialisation of paid search and SEO has meant that SEOs were focused on matching sites for keyword rankings that earn business. AI-powered search changes this. It's the equivalent of fishing with massive nets and then sorting the catch, whereas keyword/traditional search is akin to fishing with rods. GEO and AI mode take marketing back to what matters. It doesn't actually matter who wins the AI search race. Because search is changing. However, make no mistake, this is going to change revenue models, and businesses must prepare for this. AI search is huge, and it's coming fast.

42

13 Comments

Genuine request. If you're a company who likes to think 'long term', particularly around something as nebulous and fast-moving as LLMs/AI, please please please understand that KPIs or targets for the work proposed may not exist yet, or may not yet be more than 0.01%...because we're preparing for a different market environment.... ...and that doesn't mean the work is any less valuable.

34

2 Comments

SEO isn’t about ticking boxes. It’s about sequencing. You can build 100 links, publish 50 blogs, or optimise every title tag… But if the fundamentals aren’t in place, growth will stall. SEO has never been one-size-fits-all. One single tactic alone doesn't maximise visibility. It's all about a compounding strategy, layers working together in the right order. Yes, there are common priorities across industries… but there’s no magic lever you pull once and watch rankings explode. Even the most experienced SEOs don't always get it right the first time, because it’s a process of testing, refining, and compounding. So if your marketing feels busy but not effective, the issue probably isn’t a lack of effort; it’s sequencing. The best agencies don’t just do more, they do what matters most, in the right order. That’s the difference between a trickle of progress and consistent growth.

22

9 Comments

Fab to join the State of Play panel last night on the future of SEO Worried about how AI might affect SEO? Key things we're thinking about right now at eight&four: 1) AI hasn't and will not 'kill' Google The actual big story going on in Search isn't LLM stealing market share, it's expansion & fragmentation of search surfaces... Datos’ Q2 report showed YouTube has tripled YoY, Pinterest search doubled, Reddit search +30%. Meanwhile, LLM prompt growth is modest (and early evidence seems to indicate the more people use the LLM's, the MORE they then use Google...!) So - don't get too focused on AI panic, instead stand back and think about how to build distributed influence across as many search surfaces as possible.. this means SEO teams need to be partnering with social teams, PR teams, paid teams - breaking down these baked in digital siloes internally (it's why we integrated a PR & comms agency into eight&four this year - to drive that integrated thinking for our clients) 2) If an agency is trying to panic you, walk away ESPECIALLY if they're trying to sell you a "GEO Ranking dashboard" 🙄. As always with any AI hype, step back, think critically, trust your professional judgement. Replace GEO theatre with clear roadmaps to assist with LLM inclusion, adapt owned content entities to support answer generation (in LLM's, or new AI features being implemented in Google - AI mode, Overview etc.). And yes, take those traditional SEO techniques (technical, schema, IA) into Reddit, YouTube, TikTok! 3) This is a huge opportunity for "SEO" Any hype cycle brings one clear benefit to marketeers - ATTENTION. Use this window of attention SEO is currently getting to make powerful changes to the way your organisation values and prioritises world class content strategy. Reclaim that consultancy seat. Authoritatively push for more effective content investment. 2026? Let's break down silo'ed digital channels, and work effectively together to drive DISCOVERABILITY for our brands. One integrated team, different specialists, working together to show up for our customers, with the right information, in the right places. Fortune favours the brave - go get it!! 🔥🔥🔥 Big thanks to Jessica R. for the invite, and fellow panellists Andrew Fox Helen Pollitt Jessica Atkinson João Filipe Pereira Mark Evans for a great chat 🙏

36

6 Comments

Signals, Tracking and Attribution... what’s the difference? 🤔 In B2B marketing, these three concepts are often used interchangeably. But they’re not the same, and understanding the difference is quite important. 👉 Signals: the raw interactions Signals are the traces buyers leave behind. A website visit, a webinar sign-up, a LinkedIn ad click, or an email open, all of these are signals. They show intent, interest, or activity. But on their own, they don’t tell anything besides something has happened. 👉 Tracking: how we pick up on those traces Tracking is how we capture and organize those signals. With the right setup (whether in HubSpot, Dreamdata, or another stack) we can tie signals to a person, an account, or an activity. Good tracking ensures we don’t just see isolated actions but a more connected journey. This is also when signals become much more interesting and why an increasingly amount of companies are investing in tech that can help sales & marketing pick up relevant signals from ICP fit prospects to hit more great timings and close more deals. 👉 Attribution: making sense of impact (and likely the most misunderstood and complex of the three) Attribution is about assigning credit to the right channels and activities. Did the webinar influence the deal? Was it the LinkedIn ad? Or the sales call? Attribution models help you answer these questions and decide where to invest more resources. Multi-touch attribution is one of the more popular ones and can allow you to attribute more accurately across many different sources, people and activities, and over a longer period of time. So, the next time this is a topic of discussion, make sure to divide the three as they are different from one another. Happy data hunting peeps 🤺 #CertifiedMarketingExpert #b2b #b2bmarketing #b2bsales #data

38

2 Comments

Hey CROs, we need to talk about website teardowns. Everyone loves a good website teardown, but the truth is 'most' of them are just confidently delivered guesswork wrapped in nice frameworks and a lot of missing context. Most of these teardowns are done without access to behavioral data, test results, or any real customer research. When we, as humans, operate with limited context, we become vulnerable to guesswork, pattern-matching, and confirmation bias. And we often disguise it as expert opinion: “This looks wrong because I’ve seen something similar fail before”. But, conversion problems are not just about copy or design. They’re often rooted in deeper issues like navigation structure, pricing strategy, audience targeting, or funnel logic. And if you’re not looking at GA4 event flows or intra-site funnel breakdowns, it’s very hard to speak confidently about what’s really not working. #cro #experimentation #conversionoptimization

28

13 Comments

If GA4 says your TikTok ads aren’t converting, don’t blame GA. Blame the dredded in-app browser! When someone clicks a TikTok ad, it opens in TikTok’s own in-app browser. If they later buy in Chrome or Safari, GA4 sees it as coming from a completely new user. Whereas TikTok happily takes the credit in their own reports. This isn’t unique to TikTok. Most social apps have their own in-app browsers, breaking the link GA relies on to join a user's attribution journey. GA isn’t broken - it’s just working with what it can see. Treat GA and your ad platform numbers as different viewpoints, not competing truths. And when the stakes are high, consider incrementality testing or even MMM if feasible. Perfect attribution is a myth. Knowing where the cracks in each methodology is the advantage.

90

25 Comments