SaaS Keyword Research: Finding Founder Search Intent

Most SaaS keyword research produces a giant spreadsheet nobody uses. Marketers dump 3,000 keywords from Ahrefs, sort by volume, and pick the ten with the highest numbers. Then they wonder why the resulting content ranks nowhere and converts nothing. The problem is not the tool. The problem is confusing search volume with buyer intent. A keyword with 5,000 monthly searches means nothing if the searchers are students writing essays, not founders evaluating tools. This guide walks through the keyword research process we run inside every QwiklyLaunch 45-day sprint. It surfaces the queries founders and buyers actually type when they are ready to spend money, prioritizes clusters by revenue potential, and produces a content roadmap you can execute in a quarter. It is the exact method that has moved SaaS domains from zero to five-figure monthly organic pipelines.

Start With Jobs to Be Done, Not Seed Keywords

Traditional keyword research starts with a seed keyword and expands outward. That approach misses queries founders would search but nobody has documented. Start instead with the jobs your product does for a user. Write out the top 10 jobs in plain language. For a project management SaaS, jobs include tracking team workload, running sprint retrospectives, managing async standups, prioritizing tasks, and forecasting delivery dates. Each job becomes a research entry point. Ask what someone with this problem would search when they first realize they need a solution. What would they search when comparing tools? What would they search when trying to implement the solution?

Query intent types every SaaS needs

• Problem-aware: user knows they have a problem but not the solution category
• Solution-aware: user knows there is a category but is evaluating options
• Product-aware: user knows specific products and is comparing
• Implementation: user has chosen and needs help using
• Job title or role: user identifies with a persona

Every cluster you build should cover all five intent types. Missing intent types means missing revenue.

Mining Real Conversations

Keyword tools only show queries that already have search volume. They do not show emerging queries or queries that appear in real conversations but not yet in search logs. Mine actual conversations for both. Sources include Reddit threads in relevant subreddits, LinkedIn posts from your target personas, sales call transcripts, support ticket subject lines, and community Slack messages. Extract phrases users actually type. These become your long-tail differentiators. When your competitors are all optimizing for the same 100 keywords from Ahrefs, ranking for the 500 queries only found in Reddit gives you a moat.

Practical mining workflow

1. Export five months of sales call transcripts
2. Search for phrases starting with how do I, what is, best way to, and I need
3. Extract 200 candidate phrases
4. Run each through Ahrefs to check if it has any search volume
5. Even zero-volume phrases can become keyword group seeds

Volume Is a Lie, Difficulty Is a Lie, Intent Is Truth

Search volume estimates from Ahrefs and Semrush are directionally correct but often wrong by a factor of two or three on long-tail queries. Difficulty scores are worse. They average domain authority of ranking pages, which ignores whether those pages actually match intent. A keyword with difficulty 45 might be easy to rank for if the top 10 pages all misinterpret intent. Prioritize by intent match, not by volume or difficulty. Look at the top 10 results for a candidate keyword. If they are all blog posts and you plan to publish a product page, you probably will not win. If they are all outdated blog posts and you can publish a fresher, more comprehensive one, you probably will.

The Cluster Framework

Group keywords into topical clusters. A cluster is a hub keyword plus 6 to 12 supporting keywords that share intent. The hub becomes a pillar page. Each supporting keyword becomes a spoke post. Every spoke links to the pillar. Every pillar links to every spoke. This structure signals topical authority to Google and captures long-tail traffic while building topical trust. Use a tool like Keyword Insights or Cluster AI to automate clustering, but always review manually. Automated clustering makes mistakes on ambiguous queries.

For a SaaS with three primary use cases, aim for three to five clusters at launch, then expand quarterly. This aligns naturally with a growth and marketing content roadmap and gives your writers a clear brief for every piece.

Prioritization by Revenue Potential

Not all clusters are equal. Score each cluster on four axes to prioritize. First, monthly search volume across the cluster. Second, buyer intent, from 1 for informational to 5 for ready to purchase. Third, competitive difficulty, from 1 for open field to 5 for dominated by DR 80 sites. Fourth, product fit, from 1 for tangential to 5 for direct match to your value proposition. Multiply the four scores. Prioritize clusters with the highest total. This method reliably surfaces clusters worth investing in and pushes vanity clusters to the bottom.

An example prioritization

• Cluster A: 8,000 monthly volume, intent 3, difficulty 4, fit 5 = 480
• Cluster B: 2,000 monthly volume, intent 5, difficulty 2, fit 5 = 100
• Cluster C: 15,000 monthly volume, intent 2, difficulty 5, fit 3 = 450

Cluster A wins on totals. Cluster B is smaller but has the highest intent and lowest difficulty. Both are worth building. Cluster C is a trap because low intent means low conversion.

Bottom of Funnel First

New SaaS domains should always start with bottom-of-funnel keywords, not top-of-funnel awareness content. Bottom-of-funnel queries include tool A vs tool B, best software for X job, alternatives to product Y, and pricing for category Z. These queries have fewer searches but far higher conversion rates and are usually less competitive. A single bottom-of-funnel post that ranks in the top three often drives more pipeline than 20 top-of-funnel posts combined. Awareness content matters, but it should come after you have captured demand from users who are already in market.

Long Tail and Zero Volume

A keyword with zero reported search volume is not always worthless. Ahrefs rounds down anything below 10 monthly searches to zero. A cluster of 50 zero-volume queries can collectively drive 500 clicks per month. Long-tail queries also convert at much higher rates than head terms because they signal specific intent. Build a habit of publishing pages for oddly specific queries you find in sales conversations. These pages become the foundation of your topical authority even if individual traffic is low.

Tools Worth Paying For

Two tools matter most. Ahrefs or Semrush for keyword databases and competitor analysis. Pick one and use it deeply. Google Search Console for actual query data from your site, which is the only fully accurate source you will ever have. Add a clustering tool like Keyword Insights or CustomGPT if you are working with more than 500 keywords. Skip niche keyword tools that promise magic. The core workflow is: pull competitor keywords from Ahrefs, filter by intent, cluster, prioritize, then supplement with Search Console data from your existing pages.

Mapping Keywords to Pages

Every prioritized keyword needs one destination page. Never target the same keyword with two pages. That creates cannibalization and confuses Google. Build a keyword-to-URL map spreadsheet with columns for primary keyword, secondary keywords, URL, page type, publish status, and current position. Update weekly. This spreadsheet becomes your content operations backbone. Without it, teams drift into duplicate coverage and gaps in critical clusters. Building this map is one of the first deliverables inside our SaaS development engagements when SEO is in scope.

Competitor Analysis That Reveals Real Gaps

Every keyword research project should include a competitor gap analysis. Pull the top three competitors' organic keywords using Ahrefs Content Gap or Semrush Keyword Gap. Filter for keywords they rank for and you do not. Sort by intent match to your product. This surfaces the specific queries where you have the biggest ranking opportunity because a competitor already proved the query converts. Do not try to win every gap. Pick the 20 to 30 highest-intent gaps and add them to your priority list. Gap analysis run quarterly consistently produces the highest-ROI additions to your keyword universe.

Gap analysis workflow

• Pull organic keywords for top three competitors
• Cross-reference against your own keywords to find gaps
• Filter by buyer intent and product fit
• Prioritize gaps with 100 to 5000 monthly search volume
• Assign each priority gap to a content brief

Query Intent Interpretation With SERPs

The SERP tells you what Google thinks a query means. Before writing content for a keyword, look at the top ten results. Are they blog posts, product pages, videos, or category pages? What word count do they average? What structure do they use? Do they include images, tables, or embedded tools? Match the format Google rewards. Publishing a blog post for a query where every ranking result is a product page means you will not rank regardless of quality. This SERP analysis takes five minutes per keyword and prevents weeks of wasted content effort.

Handling Cannibalization Before It Starts

Cannibalization happens when two of your pages target the same keyword. Google picks one to rank and ignores the other. This dilutes your ranking authority for the query. Prevent cannibalization by enforcing one keyword per URL in your keyword-to-URL map. When two pages naturally overlap, consolidate them into one comprehensive page and redirect the weaker URL to the stronger. Audit for cannibalization quarterly using Search Console query filters or a dedicated tool like SEMrush position tracking. Early prevention is far cheaper than later cleanup, so bake the discipline into your content operations from day one.

Search Console as an Ongoing Research Tool

Google Search Console shows queries your site already ranks for. This data is more valuable than any third-party tool because it reflects your actual buyer behavior. Every month, filter Search Console for queries where you rank in positions 4 to 20 with over 100 monthly impressions. These are queries where a small optimization can push you to the top three. Update the page, add supporting content, and rebuild internal links. This process produces faster ROI than writing new content because you already have the ranking foundation. Founders who ignore Search Console data leave enormous compounding value on the table.

Search Console workflows to run monthly

1. Filter queries by position range 4 to 20 with impressions over 100
2. Filter for pages with high impressions and low CTR to find title tag rewrites
3. Sort by clicks to find your top-earning pages for deeper investment
4. Look at queries appearing in impressions but not ranking to find content gaps
5. Track new indexed URLs to catch orphaned pages Google finds through backlinks

Emerging Query Detection

New queries emerge constantly. A term that had zero search volume last quarter might have 500 monthly searches this quarter. Detect emerging queries by monitoring Reddit for new terminology, watching Twitter engineering circles for new patterns, subscribing to newsletters in your niche, and tracking new dictionary entries. Emerging queries have low competition because established sites have not yet targeted them. Capturing them early builds ranking authority that compounds when the query becomes mainstream. This is how new SaaS often outranks legacy players on new categories, by moving first when established brands are still writing about old terminology.

Voice and AI Query Considerations

Voice search and AI answer engines like ChatGPT search change how queries look. Voice queries are longer and more conversational. AI answer engines synthesize from multiple sources rather than returning ranked links. Optimize for both by writing content that answers questions directly in the first paragraph, using natural language rather than keyword stuffing, and adding FAQ schema to help AI extract answers. Do not rewrite existing content for voice or AI. Just add these enhancements to new content. Both channels are still small compared to traditional search but growing quickly enough to matter within 24 months. Founders who invest now in AI-friendly formatting build a durable ranking advantage for the next wave of search behavior. Structured content with clear headings, direct answers, and factual citations wins in both traditional Google results and emerging AI answer surfaces. The pattern is the same as good writing has always been, so the investment pays back regardless of which channel dominates.

How Research Fits a 45-Day Sprint

Inside a QwiklyLaunch 45-day sprint, keyword research happens in the first week. We build the universe, cluster it, prioritize, and produce the keyword-to-URL map by day seven. The next four weeks are execution: shipping the priority pillar pages and the first batch of spoke content. By day 45 the site has 20 to 30 pages targeting mapped keywords. This works because the research is scoped and time-boxed. Founders who let keyword research drag on for months delay everything downstream. Do the research fast, ship the content, then iterate based on real Search Console data.

Keyword research is not a phase you finish once. It is a habit you run every quarter. Real Search Console data will surface queries you never predicted. Mine that data monthly and update your map. Over time you build a proprietary understanding of your buyer's search behavior that no competitor can copy. If you want a team that treats keyword research as an operational discipline rather than a one-time project, talk to us about your launch. You can also see how we structure content roadmaps on the projects page.