Keyword Research for a New SaaS With Zero Domain Authority

Why normal keyword research fails new sites

Standard advice says to find keywords with high volume and low difficulty. For a domain with no authority, that advice breaks in two ways.

First, keyword difficulty scores in tools like Ahrefs, Semrush and Moz are estimates based mostly on the backlinks of the current top results. They do not know your site, and they miss weak pages that rank only because nothing better exists.

Second, high-volume terms are usually informational or broad. "What is a CRM" gets lots of searches and almost no buyers. "CRM for solo real estate agents" gets far fewer searches and a much higher share of people ready to try a tool.

So for a new SaaS, the goal is not traffic. It is ranking for the queries closest to a purchase, where the competition is beatable.

Start from buyers, not tools

Before opening a keyword tool, collect the words your buyers use. Spend an hour on:

• Sales and support conversations: the exact phrases people use to describe the problem.
• Reddit and forums: search "[problem] reddit" and note how people phrase questions. Our guide on finding customers on Reddit has search operators for this.
• Competitor review pages: complaints on G2 and Capterra reveal "alternative" and feature-specific searches.
• Google autocomplete and People Also Ask: type your core term plus each letter of the alphabet and save suggestions.
• Your own Search Console (once you have it): queries with impressions but no page targeting them.

You should end up with 100-200 raw phrases. Only then use a tool to check volume and see related terms.

The buyer-intent modifiers that work for SaaS

Combine your core terms with modifiers that signal someone is choosing a tool:

• Alternative: "[competitor] alternative", "[competitor] alternatives for [audience]".
• Versus: "[competitor A] vs [competitor B]", "[competitor] vs [your category]".
• Audience: "[category] for [role or industry]", for example "scheduling software for tattoo studios".
• Use case / job: "how to [job] automatically", "[job] tool".
• Integration: "[tool] [other tool] integration", "connect [tool] to [tool]".
• Template / example: "[document] template", "[workflow] example".
• Pricing: "[competitor] pricing", "cheap [category]".

These pages convert because the searcher already knows they want a solution. In our experience, a single well-targeted alternatives page can bring more signups than dozens of general blog posts.

Check SERP weakness by hand

For each candidate keyword, search it in a private window and look at the top 10. Count the signs of weakness:

• Reddit, Quora or forum threads ranking in the top 5.
• Pages that do not directly answer the query, or are outdated.
• Thin pages (a few hundred words, no examples, no data).
• Sites with a low Domain Rating or Domain Authority ranking in the top 5. See DA vs DR explained for what those scores mean.
• Few results with the exact phrase in the title.

Scoring rule of thumb: three or more weakness signs in the top 5 means a new site has a realistic shot with a clearly better page and a few relevant links. Zero or one sign, with big brands and high-authority publishers everywhere, means park it for later.

A simple prioritization template

Put every keyword in a sheet with these columns and score each 1-3:

• Business value: 3 = your product is the answer, 2 = your product helps, 1 = related topic only.
• Intent: 3 = choosing a tool, 2 = trying to solve the problem, 1 = learning.
• SERP weakness: 3 = many weak results, 2 = mixed, 1 = strong results only.
• Volume (tool estimate): note it but do not score it heavily.

Priority score = value × intent × weakness, range 1-27. Publish everything scoring 12 or higher first.

Example rows:

• "Calendly alternative for therapists": value 3, intent 3, weakness 3 = 27.
• "how to reduce no-shows at a therapy practice": value 2, intent 2, weakness 3 = 12.
• "what is appointment scheduling": value 1, intent 1, weakness 1 = 1.

Do not trust zero-volume labels

Keyword tools estimate volume from limited data, and they routinely show "0-10" for queries that actually get searched. They also group close variants together.

Signs a low-volume keyword is worth writing for anyway:

• It appears in autocomplete or People Also Ask.
• People ask it on Reddit or in your support inbox.
• Competitors have a page targeting it.
• It is a natural variation of a proven query.

A page ranking first for twenty related low-volume queries can bring more qualified visitors than a page ranking ninth for one high-volume term.

Turn keywords into a first 30-page plan

Group keywords into clusters around one problem each, then publish in this order:

1. 5 alternatives and comparison pages (highest intent).
2. 5 audience or use-case pages for your best segments.
3. 10 how-to guides that solve the problem and naturally show your product.
4. 5 templates or free tools that can earn links.
5. 5 supporting pages answering specific questions in each cluster.

Link every supporting page to its cluster's main page and back. Publish two to three pages a week. New domains also need some authority, so combine the plan with a modest number of relevant backlinks. If you do not have time to write them, our blog post service produces buyer-intent pages from your keyword list.

Measure and re-prioritize monthly

After 6-8 weeks, open the Search Console Performance report and look at queries per page:

• Pages with impressions at positions 8-20 are your best opportunities. Improve the content, add internal links and a few backlinks.
• Queries you never targeted but get impressions for are new keyword ideas.
• Pages with zero impressions after two months may have an indexing or intent problem.

Keyword research is not a one-time project. The first plan gets you indexed; Search Console data tells you where you can actually win.