Docket vs Google Search Console
Search Console is free, it is Google's own data, and no audit tool replaces it — this one included. It is the single source of the queries your site actually appeared for, the clicks and impressions they earned, Core Web Vitals measured on real visitors, and any manual action taken against you. A product sold to you as a Search Console replacement is lying. Verify your property first. This page is about the question Search Console does not answer.
Four things only Google can tell you
None of these is derivable from a crawl, by anyone, at any price:
- Queries, impressions, clicks and average position. Demand lives in Google's logs. A crawler on your laptop can read every word of a page and still have no idea whether a human ever searched for it.
- Index status, from the source. Whether Google holds this URL, which canonical it chose over the one you declared, when it last crawled, and why a page was excluded. A crawler can establish that a page is crawlable. Only Google can confirm it is indexed.
- Core Web Vitals from real visitors. Google's help page is explicit: The data for the Core Web Vitals report comes from the CrUX report, measured on actual Chrome users and reported at the 75th percentile.
- Manual actions and security issues. A human reviewer's decision about your site, or a hacked-content notice. No crawl surfaces either, because neither is on your site.
If your property is not verified, stop reading and go and do that. It costs nothing, it takes five minutes, and every paid tool in this category is worse without it.
The shape of what Search Console reports
Search Console is a record of what has already happened to pages Google has already visited. That sentence describes its limits more usefully than a feature list, and Google documents most of them itself:
- It does not crawl on demand. URL Inspection handles one URL at a time, and Google's published quota for the API is 2,000 queries a day and 600 a minute per property. A site of 5,000 pages cannot be inspected in a day even by script.
- A page with no impressions has no row. The Performance report is built from appearances in search results. A page you published this morning, or one Google has never indexed, is simply absent — and absence is usually the case you were trying to debug.
- The tables are truncated on purpose. Google's own documentation: Our tables can show a maximum of 1,000 rows, so some rows might be omitted. The same page adds that it might not track some queries that are made a very small number of times, and that most reports only cover a representative sample of a site's URLs. The chart totals and the table will not reconcile, and that is the intended behaviour.
- It runs behind. Google's figure, again from its own help: Normally, however, collected data should be available in 2-3 days.
- It reports; it does not sequence. Page indexing groups excluded URLs by reason. It does not weigh one reason against another, or against how long a fix takes, and it will not tell you which to spend Tuesday morning on.
Every one of those is correct behaviour for a reporting surface owned by the search engine. None of them is a defect. Together they are the outline of a different question.
What Docket does to the same site
Docket crawls the site as it stands right now and runs 96 checks — technical SEO,
copy, page speed, structured data, local visibility, AI search access and marketing
conversion — then ranks every finding by impact against effort and sorts them into four phases.
Each finding states what it costs you in plain language and carries the exact markup to paste.
It writes a client-ready PDF and can score you against a named competitor. The crawl runs on
your Mac: no account, no telemetry, results in ~/.docket.
The difference is time-shape more than feature list. Search Console answers questions about last week, on pages Google has met. Docket answers a question about this minute, on every page you have — including the one you published an hour ago.
A division of labour, question by question
| The question | Search Console | Docket |
|---|---|---|
| Which queries did I appear for, and what did they earn? | The only source there is | Cannot know |
| Is this URL indexed, and which canonical did Google pick? | Google's own verdict | Reports what you declared, not what Google chose |
| What are my Core Web Vitals on real visitors? | CrUX field data, 75th percentile | Fetches the same CrUX values, unless offline |
| Do I have a manual action? | The only place it appears | Cannot know |
| What is wrong with the page I published an hour ago? | No data for 2-3 days | Answered now |
| What is wrong with a page that has never had an impression? | Not in the report | Crawled like any other page |
| Which of these 60 findings do I do first? | Grouped by reason | Ranked by impact against effort, in four phases |
| What exactly do I paste to fix it? | Explains the concept | The markup itself |
| How do I hand this to a client? | Export the tables | Client-ready PDF |
| How do I compare against a named competitor? | Not its job | Side-by-side scoring |
| Where does my site's data end up? | Google's servers | Your Mac, in ~/.docket |
Deliberately no green ticks and red crosses in that table. Nine of those rows are not a scoreboard; they are two instruments pointed at different things, and colouring Google's cells red would be the dishonesty this page opened by refusing.
On cost: Search Console is free and has never been anything else. Docket is $199 paid once.
Where Docket defers: Core Web Vitals
Docket does not measure Core Web Vitals. Its speed checks are causes —
render-blocking resources, uncompressed images, page weight — and a cause is not a metric. The
metrics come from real visitors, which is why Search Console builds that report from CrUX rather
than from a lab run. Docket has one optional check that fetches the same CrUX values through the
PageSpeed Insights API; it is one of four checks that reach off your machine, and
--offline turns it off. Google's copy is the authority and the property-wide
grouping is in Search Console, so confirm a speed fix there. For the lab side of the same
problem, see Docket vs Lighthouse.
There is a hole underneath both tools, and small sites fall into it. A URL needs enough real Chrome traffic before CrUX will report on it at all, and Google does not publish how much: An exact number is not disclosed. Search Console's own help adds that A URL group without threshold data for both LCP and CLS will not be on the report. On a low-traffic site neither tool hands you a field number. Fix the causes, and read the origin-level figure instead of a per-URL one.
How the two fit into a week
- Verify the property in Search Console. Nothing substitutes for this step.
- Read Page indexing and Performance for what Google already thinks: which pages earn impressions, which are excluded, and for what stated reason.
- Crawl with Docket for what is true today, including every page Search Console has nothing to say about.
- Work the ranked plan, re-crawling after each change. That costs nothing locally.
- Come back to Search Console a few days later for the verdict, because that is the only place a verdict exists.
The recommendation is boring, which is usually the sign it is honest: keep the free one forever, and add the crawler when the question is what to change rather than what happened.
Checked 10 August 2026. The factual claims about this product were read from its own pages on that date: the URL Inspection API quota is "2000 QPD" and "600 QPM" per site (source); "Our tables can show a maximum of 1,000 rows, so some rows might be omitted" (source); "Normally, however, collected data should be available in 2-3 days" (source); Google "might not track some queries that are made a very small number of times", and most reports "only cover a representative sample" of a site's URLs (source); "The data for the Core Web Vitals report comes from the CrUX report" — field data from real visitors, reported at the 75th percentile — and "A URL group without threshold data for both LCP and CLS will not be on the report" (source); how much traffic a page needs before CrUX will report on it is not published: "An exact number is not disclosed" (source). Google changes its products and its documentation; if something here has gone out of date, tell us and it will be corrected. Nothing on this page argues that you should use Docket instead of Search Console, because you should not.
Docket is a one-time download for macOS. No account, no crawl credits, and the crawl runs
on your machine. Four optional checks fetch data it cannot produce alone; --offline
turns all four off. Download Docket →
Common questions
Can Docket replace Google Search Console?
No, and nothing can. Search Console is the only source of your real queries, impressions, clicks, Core Web Vitals measured on real visitors, and manual action notices. It is free and it is Google's own data. Docket audits the site itself and assumes you already have Search Console set up.
Why would I need an audit tool if Search Console is free?
Because Search Console reports on pages Google has already crawled, and says nothing about a page with no impressions yet. It also groups its findings by reason rather than ranking them. Docket crawls the site as it stands today and returns an ordered fix plan with the markup to paste.
Does Docket measure Core Web Vitals?
No. Docket checks the causes of poor performance and can fetch the same CrUX field values Search Console shows, through the PageSpeed Insights API, as one of four optional checks that --offline turns off. The authoritative view is Google's and it is in Search Console. Confirm a speed fix there, not in Docket.
Is the URL Inspection tool enough to audit a site?
Not at any scale. It inspects one URL at a time, and Google documents the API quota as 2,000 queries a day and 600 a minute per property. A site larger than that cannot be inspected in a day, and inspection reports indexing status rather than telling you what to change.