Analytics, DeveloperHub
Analytics · Search, feedback, health

Every search your
docs failed.

Readers tell you what is missing every time they type into your search bar and get nothing back. DeveloperHub logs all of it: top queries, click-through, dead ends, page helpfulness and link health, in one dashboard you can act on this afternoon.

14-day trialNo card4.8★ on G2 Search analytics docs →

Search analytics, reader feedback, link health and an activity log. See what each plan includes →

analytics / search 7d30d6m
Searches
1,842
 
No results
4.1%
 
Helpful
87%
 
02 · Gaps

A dead-end search is a support ticket with a delay on it.

The no-hit report lists every query that returned nothing, ranked by how many people hit the same wall. It is the shortest to-do list in your documentation backlog, and the agent will draft the missing page from it. OPSWAT ran this loop and cut support ticket volume by 15%.

Ranked by frequency

Twenty-three people looked for the same thing and left empty-handed. Write that page first.

No-click searches too

Results came back and nobody opened one. The wording is wrong, or the page is.

Hand it to the agent

One click turns a gap into a draft, written against your repo and the rest of your docs.

Explore AI features →
Searches with no results last 30 days
3 gaps · 51 searches ended nowhere
03 · Feedback

Helpfulness scores sat next to the traffic.

A page read four thousand times with a 61% helpfulness score is the most expensive page you own. Ratings and written reader messages land in the same dashboard as the search data, so the highest-impact edit is obvious rather than argued about.

Sentiment score per page, charted day by day
Most and least liked pages ranked side by side
Unread reader messages flagged, with the page they came from
Slack pings the moment a message lands, plus a bi-weekly digest of broken links
Explore reader feedback →
Helpfulness by page · this week
Webhook setup312 views93%
Getting started1,204 views87%
Authentication4,081 views61%
Rate limits886 views44%
Authentication: read most, liked least. Two unread messages both ask for a cURL example.
4 pages below 50% helpful · needs attention
04 · Health

The rot you cannot see from the inside.

Broken links, orphaned pages, prose nobody can parse and an edit history nobody can reconstruct. Four reports, all running whether or not you remember to look.

Link analyser

Every link checked, internal and external.

A prioritised report of broken links, URLs that now redirect, and pages nothing links to. Each one comes with a suggested fix you apply in a click.

Link health · 1,284 links checked
/api-basics/authentication200
/legacy-api/v1/tokens404
!external-tool.com/docs301
/guides/old-quickstartorphan
Activity log

Who changed what, and when they changed it.

Every edit, publish, version bump and permission change, attributed and timestamped. Filter by action type or by author. Useful for compliance reviews, onboarding, and the morning after an incident.

Project activity
S
Sarah published Getting started
v2.5.0 · 2 hours ago
J
James edited Authentication
OAuth 2.0 scopes section · yesterday
AI
Agent drafted Rate limits
from the no-hit report · yesterday
Readability

Grade level, reading time, Flesch score.

Scored per page, so the guide that quietly turned into a specification gets flagged before a reader gives up on it.

Getting started4 min readgrade 8
Authentication11 min readgrade 15
Authentication reads three grades harder than the rest of your docs. Ask the agent to simplify it.
Google Analytics

Page-level data, without touching the theme.

Paste a GA4 tracking ID and page views, sessions, bounce rate and time on page come back from Google, alongside the native search analytics. No code changes, no tag manager project.

Tracking IDG-XXXXXXXXXX
StatusConnected
Data streamdocs.acme.dev
05 · And the rest

No third-party dashboard to configure.

CSV export

Queries, feedback and page metrics, out to your own BI tools.

Google Analytics

One GA4 tracking ID and full session data comes back.

Date filtering

Any custom range. Set it to the days after a rewrite and see where the numbers landed.

Per-version analytics

Which version readers are on, and which one they have abandoned.

Readability metrics

Grade level, reading time and Flesch score per page.

Slack notifications

An instant ping on every new reader message, and a bi-weekly email digest of the links that broke.

Data is only useful if somebody books the meeting.

The teams who get the most out of this treat it as a fortnightly ritual: open the no-hit report, open the feedback list, fix the top three, ship. Then look again in two weeks.

100%
of queries logged and analysed
15%+
drop in support tickets
4.8★
average rating on G2
14-day
free trial, no card required

“The feedback loop works great, we have biweekly meetings to enhance the documentation according to search analytics and user feedback. It is no longer just about writing documentation; it is about writing, analysing, gathering feedback and iterating to provide the best experience possible.”

George Prichici
George Prichici
VP of Product, OPSWAT
Read the OPSWAT story →

Find out what readers
could not find.

Import your docs, wait a week, and read the no-hit report. Fourteen days free, no card, no sales call unless you want one.