How to Do Keyword Research for a SaaS Blog: A 7-Step System That Starts With Your Product (2026)

Most SaaS keyword research starts with a topic. That's the mistake. It should start with the problem your product already solves, because that's the only list that produces traffic and pipeline at the same time.

Why SaaS keyword research is different from blog SEO

A recipe blog and a SaaS blog can both rank for the same search volume and get completely different outcomes. One sells ad impressions, so any visitor is a win. The other is trying to generate trial signups and qualified pipeline, so a visitor who will never buy is close to worthless, no matter how much traffic they represent.

The high-volume trap: traffic that never converts

It's tempting to build a keyword list around whatever terms show the biggest numbers in Ahrefs' Keywords Explorer or Semrush's Keyword Magic Tool. But a term like "what is project management" can pull tens of thousands of monthly searches and still send you students, job seekers, and people researching a concept for a class assignment. None of them are evaluating software. You'll rank, you'll get a nice traffic chart, and your trial signups won't move.

Business value beats search volume

The better question isn't "how many people search this." It's "how many of the people who search this would plausibly buy what I sell, and how close are they to that decision." A keyword with 90 monthly searches tied directly to your product category can outperform a 9,000 search generic term, because the smaller term is filtered for intent before the visitor even lands on the page.

The three audiences a SaaS blog actually serves (users, buyers, and the champion in between)

SaaS buying rarely involves one person. There's the end user who will live in the product day to day, the economic buyer who signs off on the spend, and often a champion who has to justify the purchase internally. Each one searches differently: users search for "how do I do X," buyers search for "best tool for X" or pricing comparisons, and champions search for proof points like case studies, ROI language, and alternatives to whatever they're currently using. A keyword list that only serves one of these audiences leaves the other two, and the deals that depend on them, on the table.

Step 1: Build your seed list from the product, not the topic

Before opening any keyword tool, write down every feature, workflow, and problem your product addresses in the words your users actually use, not the words your product page uses.

Mine feature names, jobs-to-be-done, and the problem language from your own site

Start with your own navigation, feature pages, and changelog. Every feature exists because it solves a job. "Automated invoice reminders" is a feature name; "how to get clients to pay invoices on time" is the job, and it's the phrase a prospect actually types into Google. Translate each feature into the underlying job before you touch a keyword tool, because that translation is the step most SaaS teams skip, and it's the one that produces buyer-relevant seeds instead of marketing jargon.

Pull seeds from sales calls, support tickets, and onboarding drop-off

Your sales team and support inbox are a free keyword research tool nobody treats like one. Discovery call notes reveal the exact phrasing prospects use to describe their problem before they know your product exists. Support tickets reveal the phrasing existing customers use when something isn't working. Onboarding drop-off points reveal where a job isn't getting done, which is often the same job someone is Googling a solution for right now.

Turn each core use case into a keyword 'stem'

Once you have a list of jobs, turn each one into a short stem you'll expand in Step 2. A use case like "tracking billable hours across contractors" becomes a stem like "time tracking for contractors," which then branches into comparisons, how-tos, and alternative-tool searches. Aim for 10 to 20 stems for a typical SaaS product; more than that usually means you haven't grouped related jobs together yet.

Step 2: Expand seeds into a full keyword universe

Each stem from Step 1 is a seed, not a finished keyword. The expansion step is where you turn 15 stems into a few hundred real search queries worth evaluating.

Free expansion: autocomplete, People Also Ask, Reddit, and Search Console query data

Google's autocomplete and "People Also Ask" boxes are free and current, and they reflect what people are actually searching right now rather than a tool's modeled estimate. AnswerThePublic automates pulling autocomplete-style questions around a seed term. Reddit threads in your product's category surface the raw, unfiltered language buyers use when they're not trying to sound professional, which is often closer to how they'll search than how they'll talk to a salesperson. If your site already has any traffic, Google Search Console will show you queries you're already getting impressions for, including ones you never intentionally targeted.

Paid tools: Ahrefs, Semrush, and when a cheaper tool is enough

Ahrefs and Semrush both let you expand a seed into hundreds of related terms with volume and difficulty estimates attached, and both include a "questions" and "also rank for" filter that's useful for SaaS specifically. If budget is the constraint, Google Keyword Planner is free with an active ad account and gives directionally useful volume bands even if you're not running ads, and tools like Ubersuggest cover the basics at a lower price point. For an early-stage SaaS blog with a short list of core use cases, a cheaper tool is genuinely enough; the premium tools earn their cost once you're managing dozens of clusters and need gap analysis at scale.

Competitor gap analysis: who ranks for your buyers and why

Run your top three to five direct competitors through the same keyword tool's "content gap" or "competing domains" feature. You're not looking for every keyword they rank for, you're looking for buyer-intent terms they rank for that you don't, which usually points to a use case or comparison angle you haven't covered yet. Pay particular attention to any "[competitor] alternative" or "[competitor] vs" pages ranking for your own brand name or category; those are bottom-of-funnel visitors actively comparing vendors.

Step 3: Classify every keyword by search intent

A keyword list without intent tags isn't a prioritized list, it's a spreadsheet. Intent is what tells you whether a term belongs on your blog at all.

Informational, commercial, and transactional in a SaaS context

Informational content earns top-of-funnel traffic and builds topical authority. Commercial and transactional content sits closer to revenue and should be prioritized once you have the authority to compete for it.

Reading the SERP to confirm intent before you commit

Tool-reported intent labels are a starting point, not a verdict. Search the term yourself and look at what's actually ranking. If the top ten results are a mix of forum threads, Wikipedia-style explainers, and listicles, that's informational intent regardless of what the label says. If the results are dominated by vendor pricing pages and comparison tools, that's commercial or transactional intent, and a generic explainer post won't out-rank them.

Killing keywords where the SERP is owned by directories or your own competitors' product pages

Some terms aren't winnable with a blog post no matter how well it's written. If a SERP is locked by G2, Capterra, and three competitor homepages, a new blog post has little chance of breaking in. Mark these terms as "watch" rather than "target," and revisit them once you have enough domain authority to compete, or route that intent to a comparison page on your own site instead of a blog post.

Step 4: Score keywords by business value, not just volume

This is the step that separates a SaaS keyword strategy from a general content calendar. Every keyword gets a business value score before it gets a priority.

A simple 1 to 3 business-value score tied to how close a term sits to your product

Combining value, difficulty, and volume into a single priority number

Once every keyword has a business value score, combine it with keyword difficulty and search volume into one priority number, something as simple as (business value multiplied by volume) divided by difficulty. The exact formula matters less than the discipline of forcing all three variables into a single comparable score, so you're not choosing content ideas by gut feel or by whichever number looks biggest in a tool.

Why a 90-searches/month term can outrank a 9,000-searches/month term in revenue

A business-value-3 term with 90 monthly searches, low difficulty, and an audience actively evaluating tools will often produce more trials than a business-value-1 term with 9,000 monthly searches and an audience that's mostly students and researchers. Volume tells you reach. Business value tells you whether that reach is the right audience. SaaS content calendars that only sort by volume end up with a lot of traffic and very little pipeline to show for it.

Your free keyword scoring template (mid-article CTA)

You don't need a complicated system to run this process consistently. A single spreadsheet with columns for keyword, intent, business value score, difficulty, volume, priority number, and funnel stage is enough to manage a SaaS blog's entire content calendar.

Grab the product-first keyword scoring sheet

Build the template once using the columns above, and reuse it for every new keyword batch. If you'd rather start from something pre-built, most spreadsheet tools have a basic keyword tracker template you can adapt with these fields in under ten minutes.

How to fill it in during one afternoon

Block out an afternoon, drop in every keyword from Steps 1 through 3, score business value and intent by hand (this part resists automation because it requires judgment about your own product), then let the priority formula do the ranking. A first pass on 100 to 200 keywords is realistic in that time window.

Step 5: Group keywords into topic clusters and pillars

Individual keywords don't build topical authority. Clusters do, because they signal to search engines, and to readers, that you cover a subject in depth rather than chasing isolated queries.

One pillar per core job-to-be-done

Each stem from Step 1 should become a pillar page: a comprehensive piece covering the full job-to-be-done, linked out to narrower supporting articles. If "time tracking for contractors" is a stem, the pillar covers the full topic broadly, while supporting posts handle specific angles like compliance, invoicing integration, or tool comparisons.

Mapping supporting articles to each cluster

For every pillar, list the three to eight supporting keywords from your scored list that fall under the same job-to-be-done, and assign each one its own article rather than trying to cram every angle into the pillar itself. This is also where HubSpot's writeup on topic clusters is a useful reference if you want a second framework for the pillar-and-cluster structure.

Avoiding keyword cannibalization across the cluster

Two articles targeting near-identical intent will compete against each other in search results instead of reinforcing each other. Before publishing a new supporting article, check it against existing titles and primary keywords in the same cluster; if two pieces would satisfy the same search intent, merge them or sharpen the angle so each one owns a distinct question.

Step 6: Map clusters to the buyer journey

A keyword cluster without a funnel assignment is only half-built. The other half is deciding where each piece sits in the path from "I have a problem" to "I'm comparing vendors."

Top-of-funnel problem content vs. bottom-of-funnel comparison and alternative pages

Where product-led and comparison keywords belong

Comparison and alternative keywords belong on dedicated comparison pages, not buried inside a general blog post, because readers at that stage want a direct answer fast and a page built for comparison converts better than a narrative post trying to do the same job. Product-led keywords, the ones that describe a specific feature or workflow your product handles, belong in the middle of the funnel, where a reader already understands the problem and is now evaluating how to solve it.

Planning internal links from informational posts to money pages

Every top-of-funnel post should link down to at least one middle or bottom-of-funnel page in the same cluster. This is the mechanism that turns a traffic chart into a pipeline chart: a reader lands on a problem-aware post, follows an internal link to a solution-aware or comparison page, and moves from anonymous visitor to someone evaluating your product.

Step 7: Turn the keyword map into a prioritized content calendar

All six prior steps exist to produce one output: an ordered list of what to write next, backed by a reason beyond "it seemed interesting."

Sequencing by priority score and cluster completeness

Sort by priority score first, but don't let it override cluster logic entirely. A pillar page with a slightly lower individual score may still need to go first if three high-scoring supporting articles depend on it existing. The goal is a complete cluster, not just the highest individual scores published in isolation.

Setting realistic ranking timelines for a new vs. established domain

A brand-new domain with little authority should expect low-difficulty, high-business-value terms to take meaningfully longer to rank than the same terms would on an established site, often several months rather than weeks. Google's own documentation on how search works is a reasonable baseline for understanding why new content takes time to be crawled, indexed, and evaluated against established competitors. Set expectations with your team accordingly rather than judging a six-week-old post against a five-year-old competitor's rankings.

Refreshing and re-researching on a quarterly cadence

Keyword research isn't a one-time project. Revisit your scored list quarterly: pull fresh Search Console data, check which terms moved in difficulty, and add new seeds from any product features shipped since the last round. Treat the keyword sheet as a living document, not a deliverable you file away after the first content calendar is built.

Frequently asked questions

How many keywords should a new SaaS blog target in its first six months? Focus on depth over volume. A tightly scoped list of 20 to 40 high business-value keywords, organized into three to five complete clusters, will outperform a scattered list of 200 keywords with no cluster structure. Completeness within a cluster matters more than raw keyword count early on.

What's the best free keyword research tool for a bootstrapped SaaS? Google Search Console paired with Google's autocomplete and "People Also Ask" results costs nothing and reflects real, current search behavior. AnswerThePublic is a useful free layer on top of that for question-based expansion before you need to pay for a full keyword database tool.

Should I target high-volume informational keywords or low-volume buyer keywords first? Start with low-volume, high business-value buyer keywords. They're typically lower difficulty, closer to revenue, and faster to prove out. Use the early wins to justify investment in the harder, higher-volume informational terms that take longer to rank and convert.

How do I find keywords my competitors rank for but I don't? Run a content gap or "competing domains" report in Ahrefs or Semrush against your top three to five direct competitors, then filter the results down to terms with real business value using the scoring method in Step 4, since not every gap is worth closing.

How often should I redo keyword research for my SaaS blog? Plan a full refresh quarterly, with lighter checks (new Search Console queries, newly shipped features) on a monthly basis. Product changes and new features are the most common trigger for adding fresh seeds outside the regular cadence.

Is keyword difficulty reliable, and how do I interpret it for a low-authority domain? Difficulty scores are directionally useful but modeled on average ranking behavior, not your specific domain. For a new or low-authority site, treat any difficulty score as a floor rather than an exact prediction, and weight business value and SERP intent more heavily than the raw difficulty number when deciding what to target first.

The takeaway

Keyword research for a SaaS blog isn't a research project, it's a translation exercise: turning what your product already does into the language your buyers already use to search for it. Start from the product, expand with both free and paid tools, score by business value instead of volume alone, cluster by job-to-be-done, and map every piece to a stage in the buyer journey. Do that consistently and on a quarterly cadence, and the traffic chart and the pipeline chart start moving together instead of in opposite directions.