High Intent Keywords: A Revenue-First Method For Ordering Targets By Money, Not Volume

Order high intent keywords by the revenue each query tends to bring, not by search volume. Score every candidate on buyer stage and customer value, drop anything your product cannot serve, and use volume and difficulty only to break ties. A 1,600-search bottom-funnel query will usually beat a 9,000-search definition query.

This holds when you can build a fixed number of pages this quarter and your product serves a defined buyer.

Informational terms still earn a place when a documented path leads from them to a buying page, and the weights in this method are a starting point to tune against your own conversion data.

Why A Volume-Ranked Keyword List Misleads You

A keyword list sorted by volume is a list of what the most people type. That is the wrong thing to sort by. Volume tells you how many people search a term. It says nothing about how many of them will ever pay you.

Volume measures how many people search a term, and intent measures how likely those people are to buy. A low-volume keyword with strong buyer intent can therefore produce more leads than a high-volume keyword with weak intent.

The gap can be wide. In Emulent's published example, an informational query with 10,000+ monthly searches converts at 0.5 to 1 percent, while a transactional query with 500 to 2,000 monthly searches converts at 10 to 20 percent. Emulent presents those figures as an illustration, not a measured benchmark, so read them as the shape of the problem and not as a number to plan around. The shape is the point. The query everyone searches is rarely the query that fills your pipeline.

Volume still earns a place. It sizes the ceiling on a keyword, and a term with no searches at all is not worth a page. What volume cannot do is set the order. Volume tools rank terms by how many people search them, so a narrow buying query with a few hundred searches sits near the bottom of the export even when it sits near the top of your pipeline. The tool is answering a different question from the one you are paying to answer.

The real question is which keyword deserves a page first, given that you can only build so many this quarter. In our read of the top ten results for high intent keywords, none gave a per-keyword scoring formula or a worked numeric example. They define the term and stop. This article gives you the rule and the arithmetic, with every input yours to replace.

What Makes A Keyword High Intent In B2B SaaS

A high intent keyword is a search whose wording shows the searcher is choosing, comparing or buying a solution. The searcher has stopped asking what something is and started asking which one to pick.

Searchers move through three stages. At the learning stage, they want to understand a problem. At the comparing stage, they weigh options and read alternatives. At the buying stage, they know the category they need and are choosing a vendor, a plan or a price. Only the last two are high intent, and the buying stage is where the scoring rule in this article puts the most weight. Our guide to lifecycle content mapped to funnel stages covers how each stage gets its own page type.

Four signals mark a bottom-funnel query, each with an example from a hypothetical billing product:

• Commercial Language: The query names a price, a plan, an alternative or a comparison, such as "subscription billing software pricing".
• Specificity: The query names a category, a use case or a competitor, such as "invoicing software for usage-based billing".
• SERP Signals: The results page is full of vendor pages, comparison pages and ads and short on explainers, as it is for "best subscription billing software".
• Conversion Behaviour: The visitors who land from the query book demos or start trials more often than visitors from your other queries, which your analytics will show you for any query you already rank for.

Long-tail terms often signal intent, because a searcher who types seven words usually knows what they want. They do not guarantee it. "How to explain subscription billing to a new hire" is long and has no buying intent at all. Read the wording and the live results page, and ignore the word count.

Intent also depends on who is asking. A query that signals a buyer for one product signals a student for another. That is why the scoring rule below starts with a gate on product fit before it looks at anything else.

Where To Find High Intent Keyword Candidates

Six sources surface candidates that volume tools miss, and they are listed here from closest to a buying conversation to furthest. Sales calls and customer language come first, then Search Console queries, paid search data, competitor and alternatives pages, autosuggest and forum threads, and keyword tools last. The order matters. The closer a source sits to a real buyer, the more likely its phrases are queries that convert.

Sales calls and customer language: Read call recordings and the first message in demo requests. The phrases prospects use to describe their problem are close to the queries they typed before they found you. A prospect who says they are outgrowing a spreadsheet for invoicing has just handed you "invoicing software to replace spreadsheets".

Search Console queries: Filter the Performance report to queries containing words like pricing, alternative, vs, best and for. Low impressions are fine here. A query with twenty impressions that already sends a visitor to your site tells you a buyer used that wording. Look for the ones that show up on pages not built for them.

Paid search data: Open the search terms report from campaigns that already produce demos. A term that pays for itself in ads is a strong organic candidate, such as "subscription billing software for agencies". Terms that spend money and never convert are a useful list of what to disqualify.

Competitor and alternatives pages: Read how competitors name their comparison and "alternatives to" pages. Each one points at a query shape, such as "[competitor] alternatives" or "[competitor] vs [you]". Searchers type those when they are one step from a decision.

Autosuggest and forum threads: Type your category into the search box followed by words like for, vs and pricing, and read the suggestions. Then read the forum threads where buyers ask which tool to pick. The wording there is raw, and raw wording is what you want. Our guide to keyword research best practices for SaaS covers how to turn that wording into a clean list.

Keyword tools: Use them last, and use them to size candidates you already have, not to find them. Pasting your list into a tool shows volume and difficulty for each line. Our SaaS keyword research workflow shows how the tool step fits after the sources above.

Disqualify Before You Score

Scoring takes judgement, and judgement is expensive. Cut the list first so you spend it only on keywords that could actually get built. Four tests do the cutting, and a keyword that fails any one never reaches the scoring step.

1. If the product cannot serve the buyer behind the query, cut the keyword.
2. If the live results page shows a different intent from the one you assumed, such as articles ranking for a query you thought was a buying query, park the keyword or cut it.
3. If the results page already answers the query without a click, such as a snippet, a calculator or a direct answer panel, park the keyword.
4. If another page of yours would compete for the same query, merge the keyword into that page.

The first test removes the most. A product that serves mid-market finance teams has no use for a query from a freelancer, however many people type it. The second and third tests save you from pages that rank and never get clicked. The fourth keeps you from splitting one ranking across two URLs.

Park means park, not delete. A parked keyword goes into a backlog with the reason written next to it, so the next review can ask whether the reason still holds. Score only what survives.

The Revenue-First Scoring Rule

The rule is a fit gate first, then buyer stage multiplied by customer value, with the highest score built first.

Run it in five steps:

1. Set product fit to 0 or 1 for the keyword.
2. Score buyer stage from 1 to 5 using the wording of the query and the live results page.
3. Score customer value from 1 to 5 using the size of customer the query tends to bring.
4. Multiply the three numbers to get the score.
5. Sort the list from the highest score down, and settle equal scores with the tie-break in the next section.

Because stage and value multiply, a buying-stage query for your largest customers scores 25 and a definition query for your smallest scores 1. That gap is deliberate. Commercial and transactional intent is commonly weighted two to three times higher than informational intent unless a documented nurture path exists. Multiplying gets you there without a special rule. A stage-5 query scores 2.5 times a stage-2 query at the same customer value.

Each score is shorthand for expected pipeline. A high score means a query that tends to bring bigger customers who are closer to buying. The worked example below checks that shorthand against a pipeline calculation, so you can see the score and the money agree.

Teams that want more inputs have a reference point. DerivateX's published six-factor model rates buyer stage, product fit, conversion likelihood, deal value, ranking feasibility and AI-search overlap from 0 to 5. Buyer stage, product fit and conversion likelihood carry a weight of 3, and the other three carry a weight of 2, for a maximum of 75 and a recommendation to build above 52. DerivateX describes its example as modelled, not measured, so use it as a template for extra inputs and not as evidence of results.

Three inputs are enough for a first pass. Each one can be scored from the query and your customer list, and extra inputs mostly add arguments in the planning meeting. Add a fourth only when you can name a decision it would change.

Where Volume And Difficulty Fit In

Volume and difficulty settle ties; they never set the order. When two keywords carry the same score, build the one that is easier to rank first, and if difficulty is level, build the one with more volume. That is the whole job. The two numbers decide between equals and nothing else.

Difficulty has a second use, which is a feasibility check. A high-scoring keyword with a very high difficulty does not drop off the list. It moves to a different place on a value-versus-feasibility grid, which has four outcomes:

• High Value And Rankable: Keywords with high value and a realistic path to page one get built first.
• High Value But Hard To Rank: Keywords with high value and a hard path to page one become long-term bets, built after the first group ships.
• Low Value But Easy: Keywords with low value and an easy path become traffic plays, built only when spare capacity exists.
• Low On Both: Keywords with low value and a hard path get dropped.

Once the scoring is done, pull monthly volume and keyword difficulty for the shortlist from Semrush, so those two numbers only settle ties and never set the order. If you are still choosing tools, our ROI-first SEO tool stack covers where volume and difficulty figures come from and what each source costs you in time.

One warning. Teams that open the keyword tool first end up scoring with volume whether they mean to or not. Score before you look at volume, write the scores down, and then add volume to the sheet.

Worked Example: A 1,600-Volume Query Beats A 9,000-Volume Query

Here is the rule run on two hypothetical queries for a subscription billing product. Every input below is an assumption. Replace each one with your own figures before you act on the result.

• Query A: "subscription billing software alternatives", a bottom-funnel query with 1,600 searches a month and KD 25.
• Query B: "what is subscription billing", an informational definition query with 9,000 searches a month and KD 40.

Both pass the fit gate, because the product serves both searchers. Query A scores 5 for buyer stage, since a searcher looking for alternatives is ready to switch, and 4 for customer value, since teams replacing a billing system tend to be established accounts. Query B scores 2 for buyer stage, since the searcher is learning, and 3 for customer value, since a definition query brings a mixed crowd of small and large companies.

The 1,600-search, KD 25 query scores 20 and ranks first, and the 9,000-search, KD 40 query scores 6 and ranks second, so the lower-volume query gets built first. A volume-ranked list would have reversed that order.

Now cross-check the score against money. Annual pipeline for a keyword can be estimated as monthly searches x expected click-through rate x visit-to-demo rate x demo-to-opportunity rate x average contract value x 12. The rates below are assumed, and so are the contract values.

Query B gets the lower click-through rate because KD 40 makes a top spot harder to win and a definition query is easy for a results page to answer without a click. Query B also gets the lower visit-to-demo rate because most people reading a definition are not shopping yet.

The arithmetic is short. For Query A, 1,600 searches at 10 percent gives 160 visits, 4 percent of those gives 6.4 demos, 25 percent of those gives 1.6 opportunities, and 1.6 opportunities at $15,000 over 12 months gives $288,000. For Query B, 9,000 searches at 5 percent gives 450 visits, 0.2 percent gives 0.9 demos, 20 percent gives 0.18 opportunities, and that at $10,000 over 12 months gives $21,600.

On these assumptions, Query A produces about 13 times the pipeline of Query B from less than a fifth of the searches, and the order matches the score. That agreement is the check. Push on it by changing the inputs. If Query B's visit-to-demo rate were 1 percent instead of 0.2 percent, its pipeline would rise to $108,000 and still sit below Query A's $288,000. Query B would need a visit-to-demo rate of about 2.7 percent to catch up, which is more than thirteen times the rate assumed here.

Put your own numbers in. If your definition queries really do convert at 2.7 percent, your ranking will change, and that is the method working. The rule is a way to find out which of your assumptions the plan depends on.

How To Turn The Ranked List Into A Quarter Plan

A ranked list is not a plan until it has a cut line. Five steps turn scores into commitments.

1. Draw the cut line by counting the pages your team can ship this quarter, which gives you the number of pages you can commit to.
2. Map each keyword above the line to one page, so every committed keyword has a single owner and no two pages compete.
3. Group related terms into clusters, so one page serves several keywords and the plan lists pages instead of keywords.
4. Send mismatches to the right format, so a comparison query gets a comparison page and a pricing query gets a pricing explainer, with the format written beside each page.
5. Park everything below the line in a dated backlog, so the next review starts from a list instead of a blank sheet.

Clustering is where most of the saving comes from. Three queries about billing software alternatives usually want the same page. Our look at AI tools for topic clustering covers how teams group queries without doing it all by hand.

The number of keywords a B2B SaaS team targets in a quarter is its page capacity, because one page serves one cluster. There is no benchmark that beats counting your own writers and review time. A team that ships eight pages a quarter targets eight clusters.

A low-volume keyword with a high score belongs above the line. Volume is not a reason to keep a high-scoring query waiting. Buyers searching a narrow phrase are few, and each one is worth more.

When the plan is set, the next job is building pages that convert. Our guide to bottom-of-funnel content strategies covers what those pages contain, and a content programme built around buying queries is the service that does the building for SaaS teams.

How To Tell Whether The Ordering Is Working

The metric to track is demo or trial requests by landing keyword, reviewed every quarter alongside pipeline by page. Landing keyword shows you which query brought the visitor. Pipeline by page shows you which pages the money came through. Together they tell you whether a high score turned into revenue.

Compare each page's click-through against the figures in our B2B SaaS content benchmarks before you decide a page is underperforming. A page can rank well and still fail, and a page can convert well on low traffic. Read both numbers together.

Re-score the list every quarter, and sooner when the product changes or a competitor launches an alternatives page. Replace the assumed rates in your sheet with the real ones as they arrive. After two quarters, your own conversion data should be doing most of the work and the starting weights in this article should matter less.

When a high-scored page fails to convert, re-check the intent match before blaming the score. Open the live results page and compare the page type that ranks with the page type you built. A pricing query answered by a long explainer will rank badly or convert badly, and the fix is a different format on the same keyword. Our content audit checklist covers how to re-score pages you already have.

Informational terms stay on the list only with a documented path to a buying page. The path can be a link from the explainer to a comparison page, or a defined next step in your nurture sequence. If nobody can point at the path, the term scores low and waits in the backlog. Informational pages are not wrong. They are only worth building when something leads out of them.

How AI Answers Change The Ordering

AI answers absorb definitional queries first. A question like "what is subscription billing" is easy for a generated answer to resolve without a click, while a pricing or alternatives query still tends to send the searcher to vendor pages. That strengthens the case for ordering by revenue, and the scoring rule needs no change, because definition queries already score low.

The informational terms you keep, because they have a path to a buying page, now have a second job. They need to be the source an AI answer cites, as well as a page a reader finds. Our AEO strategy for B2B SaaS guide covers how to structure those pages, and our answer engine optimization work is the service version for teams that want it done for them.