Keyword tools sometimes report “0” for terms that still get real searches. Other times the term genuinely has little to no demand. A tool’s zero does not tell you which. Target these terms selectively: when Search Console, customer language, or the live SERP backs up real intent, and when the page is cheap enough to write that being wrong doesn’t cost much.
TL;DR
- A tool-reported “0” is that tool’s output, not proof of zero demand. It does not prove hidden demand either. Treat the term as unknown until you have evidence from somewhere else.
- The Keyword Golden Ratio (KGR) only works on nonzero terms under roughly 250 searches a month. A tool-reported zero makes the ratio undefined, so that case needs a different check.
- The volume number does not decide the budget. Business fit does. Fund a page when it names a real problem you solve, the SERP shows an opening, and there’s a plausible path to a conversion.
- Treat it as a portfolio, not a single bet. One low-volume page rarely moves much alone; a cluster of them, routed into an existing topical hub, compounds.
- AI visibility is a secondary reason to do this. The core case is demand and business fit. Where this content shows up in AI answers is a working hypothesis, not something Google documents.
Why keyword tools miss real demand
Volume methods differ by tool. Google Keyword Planner reports averaged historical metrics and groups some close variants under its own documented rules. Third-party tools apply different data sources and models of their own.
At the long tail, all of these methods get less reliable. A reported “0” is what one tool produced under its own data and grouping rules. That is all it is. It doesn’t prove the demand is zero, and it doesn’t prove the demand is hiding somewhere, which is a thin basis for killing a brief.
- Reporting floors differ by tool: small real demand can show up as zero, a wide range, or no estimate at all.
- Close variants get grouped inconsistently: Google can match a paraphrase to the same underlying intent when it ranks pages. Keyword tools do not always treat those variants the same way.
- New phrasing can lack any query history to estimate from: Google has said that 15% of daily searches are ones it’s never seen before in that exact form. Even so, a tool can sometimes still infer demand from related queries.
- Search Console records your own site’s impressions, not total market demand: it only reflects queries where your pages already had visibility. Its Performance report also omits anonymized queries.
When these terms deserve budget
Business fit and downside decide the budget. Reported volume does not. Fund a page when the query names a concrete problem, comparison, or local need your business serves. The live SERP has to show a real opening rather than entrenched competition, and there has to be a plausible path to a conversion.
A very specific query can point to buying intent, or to a trend that is just starting. Someone typing an oddly precise phrase is often further along in deciding something. But precision alone guarantees neither.
The same query can just as easily be a one-off curiosity search with no repeat pattern behind it.
Good fits tend to cluster in emerging industries, local markets, and product-support niches. Think of a page for something like “how to configure schema markup for a used car dealership”. It might only draw 30 monthly visitors, and those visitors are close to exactly what you sell.
A single page like that rarely does much on its own.
The value shows up when several of them add up to genuine traffic, each one covering a real variation of how people phrase the same underlying need. They also give you topical coverage a competitor working only from tool-reported volume won’t build. Keep what it costs to write and maintain them proportionate to how uncertain the payoff is.
What low difficulty adds, and where KGR fits
Reported volume and difficulty measure two different things. Volume is a demand estimate. Difficulty is a vendor’s competition estimate, and vendors weight it differently: some lean heavily on the current top results’ link profiles, others don’t. A low score is a prompt to check the SERP yourself, not proof there’s no real incumbent.
For nonzero terms under roughly 250 estimated monthly searches, the Keyword Golden Ratio (KGR) is still a fast filter. It’s a practitioner heuristic from Doug Cunnington. Divide the approximate number of Google results for an allintitle: search of your phrase by the estimated monthly volume. The heuristic treats a result under 0.25 as its strongest opportunity range, worth a closer look, not an automatic green light to publish.
If the tool reports zero, the ratio is undefined. That case needs the Search Console and live-SERP checks instead of a KGR calculation.
How I find these keywords
- Start in Search Console where the site already has visibility: filter Performance for queries with real impressions and low or no clicks. Then see whether an existing page already targets that intent before you create a new one.
- Use autocomplete and People Also Ask as prompts, not demand counts: they reflect what is currently being surfaced, shaped by relevance and popularity. No verified volume figure sits behind them. Validate what you find against Search Console or the live SERP.
- Forums and communities where your actual audience asks questions tend to surface the real phrasing keyword tools miss.
- Competitor gaps, checked in Ahrefs or Semrush, show what similar sites aren’t covering. A gap can also mean the query has low business value rather than an opportunity. Confirm it before you commit a brief.
Building it into a content strategy
- Cluster by intent rather than by keyword list. Group phrasings by what the searcher is trying to do. Then let one well-built page answer the cluster instead of spinning up a thin page per phrase.
- Connect each cluster to a relevant existing topical hub. Internal links help both users and search engines find the page and understand its context. They do not guarantee rankings on their own. But an isolated page with nothing linking to it has a much harder time getting discovered and crawled at all.
- Write in your audience’s own language, not a keyword-stuffed version of it. The point is matching how people really express the need. Natural phrasing that covers a few close variants on one page usually beats chasing one exact string.
- Measure it in Search Console and your own analytics, not the keyword tool you found it in. A tool that couldn’t estimate the volume won’t reliably report the outcome either.
What to expect, and when to check back
Set a review date, not a result date. There is no universal timeline for when a low-volume page starts producing, so don’t promise one. Check indexation separately first, then look at impressions, clicks, and whether the page is pulling in adjacent queries you didn’t explicitly target.
Decide per page from there: keep it as is, expand it, or merge it into a related page. Dropping it counts as a decision too.
Established sites with existing authority tend to see signal sooner than a brand-new domain with none.
A secondary case: AI visibility
AI visibility is a secondary test here, not the business case. The core reason to do this work is demand and business fit. Google does not document narrow, highly specific pages as more likely to appear or get cited in AI Overviews or chat-style answers. Treat any advantage there as a hypothesis worth testing and watching, not a confirmed mechanism.
I still treat broad, granular coverage of a topic as a reasonable bet under that hypothesis. Judge it by what you observe in your own visibility and results, not by what you assumed going in.
Where this doesn’t work
This isn’t a substitute for covering genuine head-term demand your business depends on. It isn’t a volume play by itself either.
A pile of disconnected zero-volume pages is more likely to create maintenance overhead and thin, overlapping content than real rankings. Use it as a completeness layer around topics you already own. Merge overlapping intent instead of publishing one page per phrase. For regulated or high-stakes topics, that means adding subject-matter review and stricter publication controls before you write, not skipping the demand altogether.
Sources and evidence limits
The claims here about how keyword tools work come from Google itself: the Keyword Planner historical-metrics documentation and the Search Console Performance report documentation. The “15% of searches are new” figure comes from Google’s Search quality blog.
KGR is different. Doug Cunnington published it as a practitioner heuristic, and it has no Google documentation behind it. The AI-visibility section is explicitly a working model, not a cited Google mechanism.
FAQ
Should I target zero search volume keywords?
Selectively. Target them when the query names a concrete problem, comparison or need your business actually addresses, confirmed by Search Console, customer language or the live SERP, not just because the phrase looks obscure. Build them as a cluster routed through an existing topical hub, not as isolated pages or a standalone traffic strategy.
How is the Keyword Golden Ratio calculated, and does it work for zero-volume terms?
Divide the approximate number of allintitle results for your phrase by its estimated monthly volume. The heuristic treats a ratio under 0.25 as its strongest range for terms under roughly 250 monthly searches. If the tool reports zero, the ratio is undefined, that case needs Search Console and live-SERP checks instead of KGR.
Why do keyword tools show zero volume for terms people actually search?
Tools use different data sources, grouping rules and models, and all of them get less reliable at the long tail, so small real demand can report as zero rather than an exact low number. Close paraphrases don’t always group consistently across tools, and genuinely new phrasing (about 15% of daily searches, per Google) has no history to estimate from yet.
How long before a zero or low search volume page shows results?
There’s no universal timeline, so set a review date rather than a result date. Check indexation separately, then assess impressions, clicks and adjacent queries before deciding to keep, expand, merge or drop the page. Sites with existing authority tend to show a real signal sooner than a brand-new domain with none behind it.
If you want this run as a topical strategy instead of handling pages one at a time, that’s what the BUXS Framework is for. Book a call if it would help to talk through where it fits your site.
Pozdrawiam serdecznie!
Direction and strategy: Szymon Slowik. Research and drafting support: LLM. Reviewed, corrected and finalized: Szymon Slowik.
Created by Szymon, edited with an LLM… see any difference? 🙂