How to Read Search Volume Data (and When It’s Lying to You)
- Filed November 10, 2026
- Keyword research, guide 3 of 6
A search volume figure is an estimate, built by a specific tool from a specific data source, for a specific place, time window and definition of “the same keyword”. Google’s own Keyword Planner defines average monthly searches as “the average number of times people have searched for a keyword and its close variants based on the month range as well as the location and Search Network settings you selected” (Google’s Keyword Planner forecasts help). Every part of that definition can make a number misleading: “average” hides seasonality, “close variants” merges searches you may want to separate, and “settings you selected” means the figure changes with location and network. Third-party tools add their own modeling on top. Read volume as an indicator of relative demand, with a range of uncertainty, never as an exact count.
Where volume numbers come from
| Source | How it estimates | Main caveats |
|---|---|---|
| Google Keyword Planner | Google's historical search data, for advertising planning | Groups close variants; averages over the selected months; can show ranges instead of exact figures; low-volume or sensitive keywords may be unavailable |
| Third-party keyword tools | Their own models and data sources, which vary by vendor | Methods are proprietary; figures differ between tools |
| Google Search Console | Actual impressions for your site | Only queries where you already appear; anonymized queries omitted from tables |
| Google Trends | Relative interest over time, indexed 0 to 100 | Not absolute volume |
Google’s Keyword Planner help also notes that historical statistics like average monthly searches “are only shown for exact matches”, and that keywords with very low search volumes or considered sensitive “aren’t discoverable or forecastable” (Keyword Planner help).
Six ways volume misleads
1. Grouped variants. Keyword Planner reports volume for a keyword “and its close variants”. “Running shoe” and “running shoes” may share one figure. Adding them together double counts.
2. Averages hide seasons. A keyword averaging 10,000 searches a month might get 40,000 in December and 2,000 in June. Check monthly trends before planning around an average.
3. Location settings. A national figure says nothing about a single city. Set the location to the market you serve.
4. Tool disagreement. Two third-party tools can report volumes that differ by multiples for the same keyword, because they model differently. Neither is “right”; compare relative sizes within one tool.
5. Volume is not clicks. Searches for a query are shared among ads, AI Overviews, featured snippets, local packs, videos and organic results. A keyword with high volume and a results page full of features can send fewer organic clicks than a smaller keyword with plain results.
6. Zero is not zero. Tools can report low or no volume for long, specific queries that people do search. Search Console impressions for your own pages reveal some of that demand.
Comparing estimated demand with its uncertainty
Suppose a company is evaluating three keywords. The figures are illustrative.
| Keyword | Keyword Planner (monthly) | Tool A (monthly) | Tool B (monthly) | Search Console impressions per month (current page, position ~8) | Reading |
|---|---|---|---|---|---|
| project management software | Range 10K to 100K | 74,000 | 49,500 | 18,200 | Large demand, competitive; tools agree on order of magnitude |
| project management for architects | Range 100 to 1K | 320 | 880 | 1,140 | Tools understate; Search Console shows more real demand |
| gantt chart template free | Range 10K to 100K | 22,000 | 60,500 | none (no page yet) | Tools disagree by nearly 3 times; check the results page and Trends before committing |
What the comparison shows:
- The first keyword’s figures are consistent in order of magnitude even though they differ.
- For the niche query, the site’s own impressions exceed both tools’ estimates, a signal that tools understate niche demand in this case.
- For the third, the disagreement between tools is large: Tool B’s figure is 60,500 / 22,000 = 2.75 times Tool A’s. Plan with the lower figure until other evidence (Trends, Search Console data from a test page) narrows it.
A reading checklist
- [ ] Note the source, location, language and date range of every figure.
- [ ] Check monthly trends, not only the average.
- [ ] Check whether variants are grouped before adding keywords together.
- [ ] Compare relative sizes within one tool rather than mixing tools.
- [ ] Look at the actual results page to estimate how much demand reaches organic results.
- [ ] Use your own Search Console impressions where you already rank.
- [ ] Treat low or zero volume on specific queries as “unknown”, not “none”.
What volume is good for
Volume helps rank opportunities against each other alongside intent and buying stage, size a market roughly, spot seasonality and detect rising or falling interest over time. It is a poor basis for traffic promises. A forecast built on volume should carry ranges and state the tool and settings used.