Google Keyword Planner vs the paid keyword tools

A search volume is an estimate of monthly searches for a query in a market, averaged over the last twelve months. It is not a count of last month, and it is not a promise of visits.

• Keyword Planner is Google's own, built for advertisers. It shows ranges to accounts without ad spend and figures to accounts with it, and it groups close variants together.
• The paid tools estimate from Google Ads data, clickstream panels and their own models. That is why two of them disagree about the same phrase, and why both are approximations.
• A cluster's volume is the sum of its members, which is why a page written for one head term is usually worth several times the number next to it.
• A query that shows an AI Overview or a featured snippet sends fewer clicks than its volume suggests.

• Precision of the volume: exact figures, ranges, or a model of the two.
• Grouping: whether close variants are folded together or shown separately.
• Difficulty: an estimate of how hard the first page is, which the Planner does not offer for organic search at all.
• Clusters: whether the tool groups queries that one page could answer.
• Rivals: whether you can see who ranks for the query today, and with which page.

• Being Google's own data, one step from the source rather than two.
• Telling you a phrase has essentially no demand, which a modelled tool sometimes rounds up into something that looks real.
• Commercial signals: the advertiser competition and the bid range say whether a query is worth money to someone.
• Seasonality, in the monthly breakdown, which settles whether a term is fading or simply out of season.
• Costing nothing, which matters when you are deciding whether this discipline is worth a subscription at all.

Its weaknesses are equally clear: ranges rather than figures without spend, grouped variants that hide which phrasing people use, no organic difficulty, and an interface built for buying ads rather than planning pages.

• Difficulty estimates, so you can tell a query you might win from one held by five large sites.
• The results page beside the query: who ranks, with which page, and how strong they look in that tool's model.
• Clusters and parent topics, which turn a list of five hundred phrases into twenty pages to write.
• Questions and long phrasings at scale, which is where a small site's first traffic usually comes from.
• History, so you can see a term rising rather than guessing from one snapshot.

None of that changes the underlying truth: every number is an estimate. What you are buying is the work of turning a query list into a page list.

1. 1

   Start with the queries you already have

   Search Console's Performance report lists the searches that have shown your pages, with impressions, clicks and average position. These are real, they are yours, and they are free.

2. 2

   Add volumes from Bing Webmaster Tools

   Its keyword research is free and comes with volumes. Bing's share is smaller, so read them as a shape of demand rather than Google's exact figure.

3. 3

   Size the head terms in Keyword Planner

   Ranges are enough to separate a term with demand from one without.

4. 4

   Read the results page for intent

   Type the query. Guides mean informational, product pages mean transactional, lists and comparisons mean the searcher is choosing. That tells you what to write more reliably than any metric.

5. 5

   Collect the questions

   Autocomplete and the People also ask box give you the sub-questions your page should answer, at no cost.

• When you are choosing between topics every week rather than every quarter.
• When you need to know who holds a query before committing a week to writing for it.
• When the list is long enough that clustering by hand has become the bottleneck.
• When a market you do not know well is on the plan, and you have no Search Console data for it.

Before those apply, the free stack is not a compromise. It is the same decisions with a slightly slower path to them.