Standard keyword research starts with a tool, a seed term and a volume column sorted descending — a list of things people search for, which is not a list of things worth writing about. You end up with a list of things people search for, which is not the same as a list of things worth writing about.
If you run Google Ads, you already own better data: the queries that produced actual customers, with the cost attached.
Where the keyword research data already is
The search terms report — the real queries people typed, not the keywords you bid on. Export the last 6–12 months with impressions, clicks, cost and conversions.
This is a list nobody else has. Your competitors can estimate your keywords; they cannot see which ones converted for you.
Sort it into four buckets
1. Converted, expensive
Queries that produce customers and cost you real money per click. This is your content plan, in priority order. Every one of these you rank for organically is budget you stop spending forever.
2. Converted, cheap
Working well in paid already. Lower SEO priority — you are winning these affordably.
3. High volume, no conversions
The terms a keyword tool would have put at the top of your list. They are traffic, not customers. Write for them only if you have a reason beyond volume — for example to build a retargeting pool.
4. Question queries
Anything starting with how, what, why, or containing cost, price, vs, near me. These are usually too research-stage to pay for and perfect to rank for. They also tend to be what people ask AI assistants, which makes them doubly worth owning.
The most valuable keyword list in your business is already sitting in your ad account, unread.
Turn it into briefs
For each term in bucket one and four:
- Read the actual query, not a normalised version. The phrasing tells you the intent — modular kitchen price gurgaon and best modular kitchen designer want different pages.
- Decide the page type. Commercial intent means a service or location page. Research intent means an article.
- Use your winning ad copy as the angle. You already know which message converts on this query; that is your H1 and opening line.
- Set the conversion goal per page. A research page should feed remarketing, not beg for a call.
What this avoids
- Writing for volume nobody buys on. The most common SEO failure, and the most expensive in wasted months.
- Guessing the angle. You have already paid to test which message converts. Reusing that is free.
- Arguing about priority. The list orders itself by money.
If you do not run ads
You have less, but not nothing. Search Console queries where you get impressions but few clicks show demand you already partly match. Your sales team's inbox and your WhatsApp history contain the exact phrasing customers use — and that language outperforms tool-generated keywords in both search and ad copy.
This is the same overlap we described in running SEO and paid together: each one makes the other cheaper, and most businesses run them as separate projects that never exchange data.
Want us to build the plan from your account? Start with a free audit.
We go through all of this on every account we take over — see how we run Google Ads.
Questions people ask
Why use ad data for keyword research?
Because it records what people typed and what happened next. A tool gives you volume and a difficulty score; the search terms report gives you the exact phrases that produced enquiries and the ones that burned budget.
Where is the search terms report?
In Google Ads, under Campaigns, then Insights and reports, then Search terms. Set the date range to at least 90 days and add the conversions column — a month of data on a small account is mostly noise.
What if I do not run ads?
Use Search Console instead. The queries report shows what you already appear for, including terms you never targeted, and positions four to fifteen are usually the cheapest wins available.
How much should I spend to get this data?
Any budget that produces a few hundred clicks on broad or phrase match will surface useful terms. Treat that spend as research with a side of enquiries rather than as a campaign you expect to be profitable from week one.