How to spot an emerging category in search data

SEO rarely gives you a chance to get somewhere first. Emerging categories are one of the exceptions.

Before a market matures, search demand, keyword difficulty, and SERPs follow recognizable patterns. Identifying those signals early can help you build authority before competition catches up.

The key is knowing the difference between an emerging category and a short-lived trend.

The research project that exposed the pattern

Earlier this year, I did some market research for a prospective client, a small consultancy that helps business leaders with AI governance and privacy. Nothing unusual about the job. Pull the keyword data, size the demand, check the competition, and work out whether search is worth the investment.

What came back wasn’t a market. It was a market taking shape. The data had a distinct pattern, one I hadn’t seen so clearly since the early days of cloud computing.

Once you know the pattern, you can spot it anywhere. The original research covered the U.K., so I later ran the same keyword cluster through the U.S. database. The pattern was identical. In most cases, the U.S. numbers were larger and growing faster.

That matters because getting into a category early is one of the few genuinely unfair advantages left in SEO. Keyword difficulty is low, the SERPs are unsettled, the language is still up for grabs, and the eventual winners haven’t been decided. Twelve months later, none of that is true. The AI governance example illustrates the signals that separate an emerging category from a passing trend.

The questions people ask don’t exist yet

The consultancy came to me with a list of questions its clients actually ask. Things like “Is it safe to use ChatGPT?” “Is AI using my data to train?” and “Can AI read my company data?” Real questions from real buyers, word for word.

Almost none returned measurable search volume in either the U.K. or U.S. databases. Not low volume. No volume data at all.

My first instinct was that the tools were wrong. They weren’t. When a category is this new, people with the problem don’t yet know what the problem is called, so they can’t search for it consistently. 

The questions get asked out loud, in meetings, and increasingly inside AI chat tools rather than in a search box. The phrasing varies so much that no single query accumulates enough volume to register. Search demand has been drifting beyond countable keywords for a while now, and emerging categories are where that drift is most obvious.

This is the first signal, and it’s deeply counterintuitive: In an emerging category, the most authentic buyer language is invisible in keyword tools.

If you size the market based on the questions your client hears every day, you’ll conclude there is no market. There is. It just hasn’t settled on the language yet. That’s also why zero- and low-volume keywords are worth taking seriously rather than filtering them out by habit.

Instead, demand appears one level of abstraction higher, in the names of regulations, standards, and job titles. Which brings me to the second signal.

Regulations, standards, and job titles get named first

While the natural-language questions returned nothing, the formal vocabulary was growing at a rate you rarely see in keyword data.

In the U.S. database, searches for “AI governance framework” grew from around 40 a month last August to 3,600 by July. “AI regulation” went from 120 to 3,600 over the same 12 months. The ISO standard for AI management systems, ISO 42001, grew from 610 to 3,600 in the U.S. and from 180 to 1,900 in the U.K. 

The EU AI Act now generates 6,600 monthly searches in the U.S., comfortably ahead of its 5,400 in the U.K., despite being European legislation. Add the surrounding terms, and the core cluster exceeds 25,000 monthly searches in the U.S. and roughly 12,000 to 13,000 in the U.K. for a category that barely existed two years ago.

Twelve-month U.S. search volume growth for anchor terms (July 2026 snapshot)

My favorite example is the most local one. “Colorado AI Act” only started registering search volume in January and has already doubled to 760 monthly U.S. searches, with a keyword difficulty score of 19. 

A state legislature names a law, and a search market appears within months. That’s how categories emerge in search data. Not through buyer questions, but through proper nouns. 

A regulation gets passed, a standard gets published, a job title starts appearing on LinkedIn, and suddenly everyone with the problem converges on the same phrase. The regulation names the category before the market does.

If you want an early warning system for emerging categories in your industry, watch the formal vocabulary: new legislation, new standards, new certifications, and new job titles. They’re the first things with stable names, so they’re the first to accumulate search volume.

The language is still unstable

The third signal is messiness. In both databases, the same intent appeared under multiple phrasings with no clear winner. 

Governance, compliance, audit, and risk all describe overlapping concepts. Nobody, including the people selling these services, had settled on a name for the category.

In a mature market, you get one dominant head term and a neat pyramid of variations underneath it. In an emerging market, you get five competing labels with similar search volume and different difficulty scores. That fragmentation is annoying for reporting and valuable for strategy because it means the category’s vocabulary is still up for grabs. 

Brands that pick a label and use it consistently are often the ones whose language the market eventually adopts. It’s also a useful reminder that search volume alone is a poor basis for keyword decisions at best, and almost useless at the frontier.

Difficulty lags demand

Here’s the signal that makes all of this commercially interesting, not just intellectually interesting.

Difficulty scores are backward-looking. They measure the strength of the pages currently ranking, and in an emerging category, nobody with strong authority has bothered to rank yet. Demand has arrived before competition.

In the U.S. data, “data privacy consultant” gets 260 monthly searches with a keyword difficulty score of 7. “AI policy template” gets 320 searches with a difficulty score of 21. “AI governance consultant,” the most exact-fit buyer term in the cluster, gets 170 monthly searches with a difficulty score of 22.

And the Colorado AI Act term I mentioned earlier gets 760 monthly searches with a difficulty score of 19. These terms have real, growing search volume and clear buyer intent, with difficulty scores you normally only see for keywords nobody wants.

The category cluster: U.S. vs. U.K. (monthly search volume and keyword difficulty, July 2026)

The side-by-side comparison also shows the window closing in real time. “AI governance” has a keyword difficulty score of 25 in the U.K. and 68 in the U.S. Same term, same month. 

The U.S. market spotted the category first, and the head terms there are already well defended, while the buyer-intent long tail is still lightly contested. in both countries. 

That gap between volume growth and keyword difficulty is the clearest quantitative signature of an emerging category, and it doesn’t stay open. Twelve months of growth like this is enough to bring every content team into the category.

The SERPs are contested by mismatched players

The qualitative version of the same signal appears in the search results themselves. When I checked the top 10 results for the main consultant-intent terms, the mix was unusual. 

IBM, Accenture, and two of the Big Four were there, all targeting enterprise budgets. Yet the top positions were mostly held by boutiques, including a two-person consultancy outranking IBM for a major category term.

That doesn’t happen in a mature market. Authority has consolidated, the big players have built out their content, and the SERP has a stable pecking order. 

When billion-dollar brands and tiny specialists share the first page, with the specialists on top, you’re looking at a category where relevance still beats authority. That’s the moment a small, focused player can win, and it’s exactly the argument I took back to the client.

One more thing stood out: Every SERP I checked included an AI Overview. Given what AI Overviews are doing to click-through rates, that would normally be bad news. In an emerging category, it works the other way. 

The category is being defined within AI answers at the same time it’s being defined in organic results, which means early movers aren’t just winning rankings. They’re becoming the sources AI systems cite. That compounds quickly.

The category is usually visible somewhere else first

Everything so far has focused on reading the signals in keyword data. The uncomfortable truth is that keyword tools are lagging indicators. 

By the time a phrase registers in search volume, the language has already formed elsewhere: on social platforms, in video, in communities, and increasingly in AI chats. If you only watch keyword tools, you’re reading yesterday’s news slightly early.

The best-known example is a consumer one. Rise at Seven’s trend tracking identified “airport outfits” as an emerging search behavior, driven by TikTok, even as keyword tools showed very little demand. Instead of searching for joggers and hoodies, people were searching for the occasion. 

They built a dedicated airport outfits category for PrettyLittleThing and supported it with digital PR and links. The page reached the top position in both the U.K. and the U.S. as search volume climbed to 21,000 monthly searches, generating about 7,000 monthly organic visits across nearly 400 keywords. 

The category was born on TikTok. Google caught up later, and the brand that had already built the shelf captured the demand. I first heard Carrie Rose tell that story during a talk in Ibiza, which rather proves the point about where categories emerge before they appear in search.

The same listening posts work for any market, including boring B2B ones:

• TikTok and Instagram search: Type your seed terms and watch what autocomplete suggests and how creators phrase things. Hashtag growth is trackable through TikTok’s Creative Center. The language people use in captions today often becomes next year’s keyword data.
• YouTube: Watch autocomplete, new channels, video titles clustering around a topic, and the comments, where people describe their problems in their own words. For the AI governance cluster, YouTube explainers on the EU AI Act appeared well before the consultant-intent search terms.
• Pinterest and Pinterest Trends: For consumer and lifestyle topics, Pinterest publishes rising searches months before they appear on Google.
• Reddit and niche communities: Subreddit growth and recurring phrases are where many technical and B2B categories quietly name themselves. If three different threads independently use the same label for a problem, that label is a candidate keyword.
• Podcasts and conference talks: Practitioners say the words on stage months before anyone types them into Google. It’s harder to quantify, but it’s free trend research you can consume while walking the dog.
• Your own first-party data: Site search, sales call notes, and support tickets reveal intent with nowhere to hide, and they cost nothing to analyze.

The bridge back to SEO is simple: When the same phrase keeps appearing across these sources, add it to your keyword tools and check it monthly. 

The month a term starts showing search volume is your timing signal. Because you were watching the platforms instead of waiting for the tools, you’ll get there while keyword difficulty is still low.

How to tell a real category from noise

Everything above describes what the pattern looks like. Before you bet a strategy on it, rule out the ways keyword data can mislead you.

Check the cohort, not the total

The most common false positive is self-inflicted. If you or your tool added keywords to a tracking project during the period, your totals grew because the list grew, not because demand did.

Any growth claim needs to be checked against a fixed cohort of keywords measured at both points in time. That’s especially important for emerging categories, where everyone is adding new terms to projects every month.

Separate news spikes from structural demand

A regulation hitting the headlines produces a spike that fades. A regulation that comes into force creates demand that persists because every affected business has to deal with it on its own timeline.

Look at the trend over at least 12 months. Steady, compounding growth signals a category. One dramatic month signals a news story.

Look for commercial terms emerging behind informational ones

Curiosity produces informational queries. A category becomes a market when transactional language appears: consultant, agency, certification, cost, or template. 

In the AI governance data, the giveaway was the template and consultant terms in the tables above. They were small but growing, with meaningful search volume and low keyword difficulty. 

When people stop asking what something is and start looking for the document that helps them do it, budgets are being allocated.

Triangulate with data nobody can inflate

Finally, cross-check the story against first-party signals: what prospects ask on sales calls, what appears in support tickets, and what your client’s analytics show. 

In my case, the client’s inbound questions had shifted noticeably toward AI over the previous six months, matching the search trend. When the anecdotes and the data agree, trust the pattern.

What to do when you find one

If you or your client serves the space, the playbook follows directly from the signals. It’s also the playbook for an organic channel that has been fundamentally disrupted: Go where the competition hasn’t consolidated yet.

Move before keyword difficulty catches up with demand. Prioritize low-difficulty, buyer-intent terms over the headline regulation term every publisher will eventually chase.

Pick your label and commit to it. The vocabulary is still unstable, which means consistent naming across your site, LinkedIn presence, and PR can shape how the market talks about it. You’re not just ranking for the category. You’re helping define it.

Answer the invisible questions. Those zero-volume buyer questions haven’t disappeared. They’ve moved into AI tools. Publishing clear, direct answers positions you for AI citations now and for the search volume those questions will eventually accumulate as the language settles.

Build the practical assets early. The template, the checklist, the plain-language explainer of the standard. In a young category, the first person to publish a useful document often becomes the default authority at a cost that would embarrass a mature-market link-building campaign.

The window doesn’t stay open for long

Most of what looks like an emerging category isn’t one. It’s a news cycle, a vendor’s marketing push, or an artifact of your own keyword list growing. Genuine categories are rare, which is exactly why they’re worth looking for systematically.

But when the pattern is real, it’s one of the few situations in modern SEO where a small player with a modest budget can secure positions that will be out of reach two years later. Categories emerge quietly in keyword tables long before they become obvious on LinkedIn. Recognize the pattern, and you’ll get there first.

Note: Search volume and keyword difficulty figures are monthly metrics from SE Ranking’s U.S. and U.K. databases, pulled in July 2026.