Every keyword research guide explains what search volume is. Almost none of them show you the moment where you look at a list of numbers and decide which page to build, which one to skip, and which one is a trap. That moment is the whole job. So instead of definitions, here is the actual work, on our own site, with the real numbers, for two decisions we made this summer: a page about SEO for franchises, and four pages about Colorado towns.
The three numbers, and what each one is actually telling you
When we were deciding whether to build a Franchise SEO page, the keyword tool gave us this:
| Keyword | Searches / month | Difficulty | Cost per click |
|---|---|---|---|
| franchise seo services | 1,300 | 6 | $23.16 |
| franchise seo company | 1,000 | 4 | $25.52 |
| local seo for franchises | 880 | 11 | $18.85 |
| seo for franchises | 880 | 26 | $24.91 |
| franchise seo agency | 720 | 5 | $22.53 |
| franchise seo expert | 480 | 10 | $24.91 |
Volume is the one everyone looks at first, and it is the least useful on its own. 1,300 searches a month sounds like a market. It is only a market if the people searching want what you sell, which volume cannot tell you.
Difficulty is an estimate of how strong the sites already ranking are. Anything under about 15 means the current results are beatable by a well-made page from a modest site. This whole cluster sits between 4 and 26. That is unusually open for a commercial term, and it was the first thing that made us look harder.
Cost per click is the number most people skip, and it is the one that answers the question volume cannot. Advertisers are paying $18 to $25 every time someone clicks an ad on these terms. Nobody pays $25 a click for curiosity. A high CPC is other companies telling you, with their own money, that the people typing this phrase turn into customers. Read together: real buyers, weak competition, roughly 8,000 searches a month across the cluster. We built the page.
The trap that looked like a keyword
Sitting right next to those terms in the export was seo franchise opportunities, 210 searches a month, similar difficulty, obviously about SEO and franchises. It would have been easy to fold it into the page. We did not, and the reason is the single most useful habit in keyword research: before you target a phrase, search it yourself and look at what ranks.
The results for that phrase are about buying and running an SEO business, not hiring one. The person typing it is not a franchise owner who wants help ranking. They are someone who wants to buy an SEO franchise. Same words, opposite customer. Putting that phrase on our page would have earned us traffic that could never convert and, worse, told Google our page was about something it was not. It stays off.
Keyword tools group by words. Google groups by intent. Every keyword list has at least one of these in it, and the tool will never flag it for you.
When a cluster earns a second page, and when it has to wait
Right beside the franchise terms is a second, related cluster: seo for multiple locations (1,300 a month, difficulty 21) and multi location local seo (880, difficulty 19). Overlapping audience, a business with several locations, but not the same search, and not the same page. Someone searching “franchise SEO” has a franchise. Someone searching “SEO for multiple locations” might have three dental offices and no franchise agreement at all.
That is a second page, not a section on the first one. But we deliberately have not built it yet. The franchise page went live in late August, and we want to see how it ranks before adding a sibling. Two thin pages on neighbouring topics do worse than one strong page, and a site that suddenly sprouts a set of similar pages is a pattern Google has learned to discount. The order of operations matters: build one, watch it, then build the next.
When two towns earn two pages
The same question came up on the local side, and the answer went the other way. We wanted local SEO pages for Northern Colorado, and the obvious efficient move was one page for Fort Collins and Loveland together, they are ten minutes apart and the region behaves as one market.
The numbers said otherwise. Searches for SEO and marketing services in Fort Collins run to roughly 2,700 a month; Loveland is about 770, at similar difficulty. That is not one market with two names, it is a large market and a smaller one, and a combined page would be a weaker match for both. There was also an argument only one of the two pages could make: for a Loveland business, the competition that matters is fifteen minutes north. You cannot say that to Fort Collins readers on the same page. So: two pages, each making its own case, cross-linked.
Aspen taught a different lesson. Fewer than 6,500 people live there, yet the search volume for local marketing help is real, because the person searching is usually not in Aspen. They are researching from out of state, weeks ahead, for a business that serves visitors. The keyword tells you who is searching, not just how many. A page written for a resident would have missed the actual customer entirely.
One more thing the tool showed us for these towns: difficulty scores of 2 to 7, and a results page made almost entirely of templated agency location pages, the same paragraph with a different city name pasted in. Nobody had written a page that understood how business actually works in the town. That is not a keyword insight; it is a competitive one. But you only find it by looking at the results, which brings us back to the habit above.
What we actually do, in order
- Pull the cluster, not the keyword. Every phrasing of the same need, with volume, difficulty and CPC side by side.
- Read CPC as intent. High CPC and low difficulty is the combination worth building for. High volume and near-zero CPC is usually curiosity.
- Search every candidate yourself. If the results are not selling what you sell, the keyword is not yours, whatever the words say.
- Decide page boundaries by intent, not by topic. Different customer, different page. Same customer, same page, even when the phrasing differs.
- Build one page at a time when the pages are siblings. Watch it rank before adding the next.
- Write down why. Six months later, when a page is or is not ranking, you will want to know what you expected and why.
What happened next
Honestly, it is too early to say. The franchise page and the four town pages went live at the end of August 2026, and as we found out the hard way, Google took its time even crawling them. We will report how they rank when there is something real to report, whether or not it makes us look good. That is the same standard we hold on our own case study.
If you would rather have someone do this work for your business than learn to read a CPC column, that is what our SEO and AI search service is for, and the keyword research is the first thing we show you, with the reasoning attached.
