Devstars
Short on time? I’ve recorded a 2-minute audio version covering the key points from this post. Listen to the audio TL;DR on SoundCloud and come back for the detail when you’re ready.
Every quarter I pull together what we’re seeing across our client base, our research, and the wider industry. Q1 of 2026 was one of those quarters where AI search optimisation stopped being a nice-to-have and became the central question for any business that depends on being found online.
The ground shifted. Not gradually. Quickly. And most businesses haven’t caught up yet.
Here’s what happened, what it means for your generative engine optimisation strategy, and what I’d be doing about it right now if I were running marketing for a mid-sized business.
The first was geopolitical. The conflict in the Middle East sent fuel costs surging, with diesel hitting its highest price in three years. UK growth forecasts got slashed. Two-thirds of businesses told Barclays they’re already feeling the squeeze on energy and fuel costs.
The second was technological. While businesses were firefighting energy bills, AI search quietly crossed a tipping point.
Google now shows AI-generated answers in roughly 60% of searches. That number was closer to 30% this time last year. And here’s the stat that should stop every business owner in their tracks: 93% of AI Mode queries end without anyone clicking through to a website.
We’ve been tracking what we call the Jaw Effect across our clients for over a year now. Impressions going up. Clicks going down. The gap widening. It used to be an early signal. Now it’s the default.
Before you write that off as bad news, there’s a flip side that makes this genuinely exciting.
Visitors who do arrive from AI search results convert at 4.4 times the rate of traditional organic traffic. Read that again. Not 10% better. Not double. Four and a half times.
What’s happening is straightforward. AI tools are doing the filtering that humans used to do manually. By the time someone clicks through from an AI recommendation, they’ve already done their research, compared options, and decided you might be right for them. They’re arriving ready to buy.
Less traffic, dramatically better quality. It’s a fundamental shift from volume to value.
The thing is, AI tools don’t just randomly decide who to recommend. They look for specific signals, and most business websites aren’t sending them.
We ran our GEO Readiness Assessment across a sample of 30 businesses in the first quarter. The average score was pretty sobering. The most common issues were the same ones we see everywhere.
Web pages that open with “Welcome to [Company Name], we are a leading provider of…” instead of directly answering the question someone actually asked. AI tools cite content that answers questions clearly in the first 50 to 70 words. If your homepage reads like a brochure, you’re invisible.
No structured data. Schema markup, the technical code that helps AI tools understand what your pages are about, was missing entirely from over two thirds of the sites we assessed. It’s not glamorous work, but it’s the foundation that everything else builds on. We’ve seen what happens when you get this right. South Coast Powersports went from zero to over 100 high-intent keyword rankings in six months, largely through structured data mastery.
Single-platform dependency. Most of the businesses we looked at were getting 80% or more of their digital visibility from Google alone. YouTube processes over 3 billion searches daily. LinkedIn drives B2B decisions. Instagram handles 6.5 billion searches. If you’re only optimising for one platform, you’re leaving a tonne of opportunity on the table.
Our weekly intelligence briefing tracks how AI tools decide what to cite. A few numbers from the latest research that are worth paying attention to.
Deep pages receive 82% of AI citations versus homepages. Your services pages, your detailed guides, your FAQ content. That’s what gets recommended. Not your homepage. Not your about page.
Content with clear question-and-answer structures dramatically improves citation rates. This isn’t complicated. If someone asks “what does a fractional CMO do?” and your page starts with a clear, direct answer, AI tools will find it and use it.
Citation volatility is high. Roughly 70% of cited pages change over a two to three month period. This matters because it means the window to establish yourself is open, but it won’t stay open indefinitely. The businesses building authority now will have a compounding advantage.
One more shift worth flagging from Q1. AI agents, software that actually does tasks rather than just answers questions, hit 40% adoption in enterprise applications. That’s not a forecast. That’s now.
Amazon laid off 16,000 people this quarter and cited a shift toward AI-driven automation. Shopify is building agentic storefronts where people can browse and buy entirely through conversation with ChatGPT. OpenAI launched an ads manager for ChatGPT. It’s enterprise-only for now, but the direction is clear. AI-powered advertising is coming for businesses of all sizes.
We’ve been building our own AI agent team internally for content distribution, research, and client reporting, and we’re starting to help clients do the same. The businesses that figure out how to use AI to scale their expertise, not replace their people, will have a serious head start.
If I were a marketing director reading this, here’s where I’d focus for Q2.
Run an AI visibility check. Ask ChatGPT, Claude, or Perplexity a question your ideal customer would ask. See if your business gets mentioned. If it doesn’t, that’s your starting point. We’ve built a free GEO Readiness Test at devstars.com/check/geo that gives you a quick score and tells you what to fix first.
Restructure your content to lead with answers. Not mission statements, not company history, not “we are delighted to…” Direct answers to the questions your customers are actually asking. Every page. Starting with the ones that get the most impressions.
Get your structured data sorted. Schema markup for your team, your company, your articles, your FAQs. This is the technical foundation that compounds everything else. It’s not the exciting part, but it’s what separates the businesses that show up in AI recommendations from the ones that don’t.
Diversify your traffic sources. No single platform should account for more than 30% of your visibility. If you haven’t started on YouTube or LinkedIn yet, Q2 is the time.
We’re based in Jersey, which gives us an interesting perspective on all of this. The island’s financial services sector, worth hundreds of billions in assets under management, is sitting on a massive digital opportunity. Virtually none of the trust companies, fund administrators, or private banks we’ve assessed are optimised for AI search. Their London and Zurich competitors will get there eventually. Right now, the window is wide open.
But this isn’t just a Jersey story. It’s the same across every sector we work in. Whether it’s Headmasters managing visibility across 56 salon locations, Investors in Property competing for international buyers, or the scale-up businesses we work with across the UK, the pattern is identical. The businesses adapting now will own the next 12 months. The ones waiting will spend those 12 months trying to catch up.
Forewarned is forearmed. And if you’ve read this far, you’re already ahead of most.
Stuart Watkins is the founder of Devstars, a Jersey and UK-based digital growth consultancy specialising in bespoke web software and Generative Engine Optimisation. He provides fractional CMO services for ambitious businesses and has spent 35 years in marketing, from A&M Records to enterprise digital transformation.
Take the free GEO Readiness Test at devstars.com/check/geo or get in touch at hello@devstars.com.
Tell me what you’re trying to fix. Half an hour, no pitch, no slide deck.
If we’re the right fit we’ll talk about what’s next. If we’re not, I’ll point you to someone who is.
