Try this. Open Google, or ChatGPT, or Google’s AI Mode, and search for the thing your business does, in the place you do it. “Emergency plumber in Brentwood.” “Accountant for small limited companies in Essex.” “Best dog groomer near Billericay.”
Now look at who comes up. In the ordinary results, a competitor sits above you. In the AI answer, a competitor is named and recommended, and you are not mentioned at all.
Here is the question that matters: do you know why?
Most business owners don’t. They have a feeling about it. The competitor is bigger, or has been around longer, or spends more on advertising, or has a flashier website. Sometimes those feelings are right. Often they are not, and the real reasons are specific, visible and fixable, if someone takes the trouble to look.
That looking has a name: competitor research. It is one of the most useful things a business can do for its online visibility, and one of the most routinely skipped. This article explains what it involves, what it usually turns up, and why the findings are so often the quickest route to improving your own website.
Why so many businesses skip it
Competitor research sounds obvious. Almost nobody does it properly.
A survey of 250 UK small and medium-sized business leaders at the end of 2025 found that only 53% regularly collect and use information about their competitors, and most of those who do keep it in a spreadsheet or a Word document. (Enterprise Times) In our experience of small service businesses the real figure is lower still, and “research” usually means an occasional look at a rival’s website and a strong opinion about their prices.
The reasons are understandable:
- It feels like a distraction. Owners are busy running the business. Studying someone else’s website looks like time spent on them rather than on you.
- It is uncomfortable. Nobody enjoys discovering that a competitor’s website is faster, clearer and better reviewed.
- The guesses feel like answers. “They spend more on ads”, “they’ve been going longer”, “they must be paying Google” all sound like explanations. They close the question without answering it.
- It looks like a technical job. Rankings, crawlers, structured data and AI citations sound like someone else’s department.
The cost of skipping it is that you end up improving your website blind. You rewrite the homepage, add a blog, maybe pay for some advertising, and never find out whether any of it addressed the actual reason the competitor was ahead. Competitor research replaces the guesses with a list.
What competitor research actually looks at
Proper competitor research is not a glance at a rival’s homepage. It is a structured comparison of your business against the two or three competitors that keep appearing where you want to appear, across five areas. Each area answers a different question about why they are winning.
| Area | What we compare | What it tells you |
|---|---|---|
| Content | Which customer questions each site answers, how directly, and with how much first-hand detail (prices, timescales, examples, photos, named people) | Whether they are simply answering questions you have never written down |
| Structure and writing | Headings, page layout, whether each section stands on its own and leads with the answer, how services and places are named | Whether their pages are easier for a reader, and an AI, to lift a clear answer from |
| Technical health | Page speed, mobile layout, crawl errors, broken links, indexing, structured data, and whether hosting or security settings block search and AI crawlers | Whether search engines and AI tools can reach and read their site more easily than yours |
| Authority and reputation | Reviews and ratings, Google Business Profile completeness, directory listings, mentions in local press and trade bodies, links from other sites, consistency of name, address and phone number everywhere | Whether the wider web vouches for them more than it vouches for you |
| Search and AI presence | The keywords they rank for that you do not, the questions where AI tools name them, and which pages the AI is citing when it does | Exactly where the gap is, and which of their pages is doing the work |
The last row is where most of the surprises come from. Business owners tend to assume a competitor is winning everywhere. The data usually says they are winning on a handful of specific questions, through a handful of specific pages, and that is a much more manageable problem.
What it looks like in practice
Here is a typical case, with the names changed. Two firms of accountants in the same Essex town, similar size, similar fees, both with decent websites. Ask an AI assistant “who is a good accountant for a small limited company in Brentwood?” and one is named every time. The other never is. The owner of the second firm is convinced the first “must be paying for it”.
They were not. The comparison showed four things:
- The first firm answered the question. It had a page called “Accountancy for limited companies in Brentwood: what we do and what it costs”, with a from-price, a list of what is included and a named accountant who looks after those clients. The second firm had a page called “Services” that described everything it did in general terms and gave no prices.
- Other people talked about the first firm. It had more than sixty Google reviews, listings in the local chamber of commerce and two trade directories, and had been quoted in the local paper about tax changes. The second firm had nine reviews and a Google Business Profile that still listed its old address.
- The first firm’s website was technically clean. Pages loaded in under two seconds on a phone, every service page was indexed, and nothing blocked AI crawlers. The second firm’s site had a security setting that was quietly turning away the crawlers behind ChatGPT and Perplexity, so those tools had never read it at all.
- The first firm was not better at everything. On “payroll services Brentwood” the second firm actually ranked higher, because it had a detailed payroll page and the first firm did not. Nobody had noticed, because nobody had looked.
None of this needed a bigger budget to fix. It needed a clear list of what the competitor was doing that the second firm was not, in order of impact, which is exactly what competitor research produces.
Techniques worth borrowing, and ones to leave alone
The point of studying a competitor is not to copy them. It is to work out which of the things they do actually produce their results, and then do those things better, in your own voice and with your own evidence.
The techniques that usually turn out to matter:
- Answering the specific question. Winning pages tend to be built around one question a customer actually asks, with the answer near the top and the detail underneath. If a competitor has a page for each service in each town and you have one page called “Services”, that structural difference alone can explain a great deal.
- Showing proof. Prices, timescales, named staff, photographs of real work, case studies and reviews. These are the things an AI tool cannot invent and a nervous customer wants to see.
- Being talked about elsewhere. AI visibility in Google’s AI Overviews correlates more strongly with how often a brand is mentioned across the web than with how many links point to its site, according to an analysis of 75,000 brands. (Ahrefs) Reviews, directory listings, local press, trade bodies and partner websites all count. A competitor who is active in the local business community often has an online advantage they did not deliberately build.
- Keeping the website technically tidy. Fast pages, a clean mobile layout, no broken links, everything indexed, crawlers allowed in. Dull, and decisive.
- Ranking well in ordinary search. It still matters, but less than it did. In early 2026 only around 38% of the pages Google’s AI Overviews cited also ranked in the top ten for the same search, down from 76% the previous summer. (Search Engine Journal) A competitor can be cited by AI from page three, and so can you, if the page deserves it.
The techniques to leave alone:
- Copying their content. It will not outrank the original, and it strips out the one thing that makes a page worth citing, which is that it comes from you.
- Chasing every keyword they rank for. Some of their rankings are irrelevant to your business or not worth having. Research tells you which gaps matter.
- Shortcuts. Google is explicit that it ignores so-called AI optimisation hacks and that mass-produced, search-engine-first content is treated as spam. (Google Search Central) If a competitor appears to be winning through tricks, it will not last, and following them down that road is a poor use of your time.
Good competitor research separates the first list from the second. Without it, businesses copy the visible things, usually the design and the slogans, and miss the structural ones that are doing the work.
Why an outside view matters
In principle you could do all of this yourself. In practice, three things get in the way.
The first is tooling. Finding out which keywords a competitor ranks for, which of their pages AI tools cite, how fast their site loads, whether it blocks AI crawlers and how many places mention them takes a set of specialist tools and the experience to read what they say. That is a lot to buy and learn for a job you will do a few times a year.
The second is objectivity. It is very hard to look at your own website and a rival’s with the same eyes. Owners forgive their own vague “Services” page and judge the competitor’s by its typos. An outsider sees both as a customer, or an AI, would.
The third is knowing what to do with the findings. A list of forty differences is not a plan. The value is in knowing which three of them explain most of the gap, and in which order to fix them.
How DotScott runs a competitor review
DotScott works with small and medium-sized businesses to join the dots between what their customers search for, what their competitors are doing about it, and what their own website needs to change. A competitor review usually runs like this.
- Agree the searches that matter. The ten to twenty things your customers actually type or ask, in the places you serve. Not what you wish they searched for, what they do.
- See who wins them, and where. We record who ranks and who is named in AI answers for each one, and note which of their pages is doing it. This is where the two or three competitors worth studying reveal themselves, and they are not always the ones you expected.
- Compare across the five areas. Content, structure, technical health, authority and search and AI presence, for your site and theirs, with the evidence for each.
- Explain the gap in plain English. Why they are ahead, where you are ahead, and which differences actually matter. If a competitor’s advantage is something you cannot or should not copy, we say so.
- Turn it into a plan. A short, ordered list of changes to your own site and your wider online presence, with the reasoning behind each, that we can either hand over or help you deliver.
The outcome is not a report about your competitors. It is a clearer, better-structured, better-supported website of your own, built on evidence of what works in your market rather than guesses.
Where to start
Go back to the search you did at the top of this article. Pick the one competitor who keeps appearing where you want to be. Open their page and yours side by side and ask three questions. Which customer question does their page answer that yours does not? What proof do they show that you do not? What do other people say about them online that nobody says about you?
If you can answer all three honestly, you already have the beginnings of a plan. If you would rather someone did it properly, across every search that matters to your business, with the technical checks and the evidence, get in touch and we will talk it through.
Sources
- ZenRows uncovers UK SME competitor blindspot, Enterprise Times, December 2025, reporting a ZenRows survey of 250 UK SME leaders
- An analysis of AI Overview brand visibility factors (75K brands studied), Ahrefs, May 2025
- Google AI Overview citations from top-ranking pages drop sharply, Search Engine Journal, March 2026
- Optimizing your website for generative AI features on Google Search, Google Search Central