Adobe has a market capitalisation north of $160 billion. Descript has roughly 200 employees. In traditional search, that gap in domain authority, content volume, and backlink profile would be almost impossible to close. In AI search, Descript is reportedly punching well above its weight - appearing in LLM recommendations alongside, and sometimes ahead of, much larger competitors including Adobe and CapCut.
This is not an anomaly. It is a signal. And if you manage marketing for a brand that isn't the category leader, it deserves your full attention.
Why AI Search Doesn't Reward Size the Way Traditional Search Did
In Google's classic ranking system, size conferred compounding advantages. More pages, more backlinks, more historical authority - it all fed the machine. A company like Adobe, with decades of web presence and millions of inbound links, was structurally difficult to displace for competitive keywords.
AI search engines - whether that's Google's AI Overviews, ChatGPT, Perplexity, or Gemini - are drawing on a different set of signals. These systems are trying to answer a question, not rank a list of documents. The question they're asking is: which brand or tool genuinely fits what this user needs? That shifts the advantage from whoever has the most content to whoever has the most relevant, credible, and well-structured answer.
Descript's visibility in AI search, as tracked using Semrush's AI Visibility score, appears to stem from exactly this dynamic. It's a focused product with a specific use case - podcasters, video creators, and content teams who need fast, AI-assisted editing. That specificity translates well into the kinds of queries AI systems receive.
Specificity Beats Scale - When the Topic Is Right
Adobe is genuinely a better answer to the question 'what software should a professional designer use?' But it's not necessarily the best answer to 'what's the easiest way to edit a podcast recording?' or 'which tool lets me remove filler words automatically?' Descript is built for those use cases. Its product documentation, its content, and its positioning all reinforce exactly that.
This is the underlying mechanic that the Descript case study exposes. AI models learn from what exists on the web, but they also learn from structure and coherence. A brand that has a clear, consistent answer to a specific class of questions will outperform a broader brand that never addresses those questions directly - even if the broader brand has ten times the domain authority.
For UK businesses operating in competitive categories, this is genuinely actionable. The question isn't 'how do we produce more content?' It's 'which specific queries do we want to own, and does our content actually answer them in a way an AI system would find useful and citable?'
How to Measure Where You Actually Stand
Most brands still have no formal view of their AI search visibility. They might check Google Search Console for organic performance, run the occasional Ahrefs audit, and call it done. That workflow tells you nothing about how often your brand is cited when someone asks ChatGPT or Perplexity for a recommendation in your category.
Semrush's AI Visibility score is one of the tools being used to quantify this - it's what the Backlinko analysis used to compare Descript against its larger competitors. There are other approaches too: manually prompting AI tools with your target queries and logging where your brand appears, how it's described, and which competitors are cited instead. This is manual, but it's illuminating. You'll often find that AI systems are confidently recommending a competitor for a use case your product handles better.
Benchmarking your AI visibility against competitors should now be standard practice. Not monthly - at minimum weekly, given how frequently these models are updated and how quickly citation patterns can shift.
What Descript Is Likely Doing Right
Without access to Descript's full content strategy, there are patterns consistent with strong LLM visibility that are worth considering. Their content almost certainly maps closely to real user questions - not broad SEO topics, but specific problems: 'how do I remove background noise from a video', 'what's the best tool for transcribing an interview'. These are exactly the query types AI systems field constantly.
Their product positioning is also unusually tight for a SaaS company. Descript doesn't try to be everything. That coherence - being clearly, demonstrably good at a defined set of tasks - makes it easier for AI systems to build an accurate, confident representation of what the product does. Adobe, by contrast, covers so much ground that an AI system has to work much harder to return it as a precise recommendation for a narrow query.
The practical implication: if your brand is trying to be cited by AI systems, clarity of positioning matters as much as content volume. An AI model recommending a tool is effectively making an editorial judgement. Make it easy to get that judgement right.
The Opportunity for Challenger Brands in the UK
UK businesses in competitive sectors - professional services, SaaS, e-commerce, financial products - are largely still fighting yesterday's battle. They're optimising for Google's ten blue links while a growing share of their potential customers are getting answers from AI systems that don't return ten blue links at all.
The Descript case study shows that the window to establish AI search presence before category leaders wake up to this is real - and finite. Descript isn't beating Adobe because Adobe is weak. It's doing so because it moved with purpose into a space where Adobe hasn't yet competed effectively. That gap exists in most UK sectors right now.
Challenger brands have a structural advantage here that they have rarely had in search: a more manageable scope. You don't need to rank for everything. You need to be the definitive answer for the queries that matter most to your buyers. Get that right, and the size of your marketing budget becomes less relevant than the quality of your positioning.
What to Do This Week
Start with an audit. Open ChatGPT, Perplexity, and Google's AI Overviews. Ask the five to ten questions your ideal customer is most likely to search before buying in your category. Note which brands appear, how they're described, and whether your brand features at all. This takes an hour and will tell you more about your AI visibility than most formal reports.
Then look at your existing content against those queries. Is there a clear, well-structured piece of content on your site that directly answers each one? If not, that's your gap - not domain authority, not backlinks, but the absence of a credible, findable answer. That's a gap you can close, regardless of whether you're 200 people or 200,000.