What Are Hits in Google Analytics?

Someone asks how many hits your site got last month, and the honest answer in 2026 is that the question no longer has a number attached to it. Google Analytics stopped counting hits years ago. The metric that replaced it counts something different, and reporting one as though it were the other is how people end up telling their boss traffic tripled overnight.

The word has not gone away, though. It survives in old dashboards, in server log reports, in agency templates nobody has updated, and in the vocabulary of anyone who learned analytics before 2020. So it is worth knowing precisely what a hit was, what it is not, and what the current equivalent is.

This guide covers the direct answer, the hit types Universal Analytics recognised, how hits relate to sessions and pageviews, what GA4 replaced them with, and where the term still legitimately appears.

What Is a Hit in Google Analytics? The Direct Answer

A hit was a single packet of data sent from your website to Google Analytics. One interaction, one hit. Load a page, that is a hit. Click a tracked button, that is another hit. Complete a purchase, that is several.

Hits were the atomic unit of the Universal Analytics data model. Everything above them was assembled from them: a session was a group of hits from one visitor in one sitting, a pageview was a hit of the pageview type, and a user was the person those sessions belonged to.

Crucially, a hit was not a visitor and it was not a visit. A single person reading four pages and clicking two tracked links generated six hits, one session, and counted as one user. That ratio is why "hits" was always the most inflated-looking number in any report, and why it became the favourite metric of anyone who wanted their traffic to sound impressive.

The word itself comes from web server logs, where it has a different and even broader meaning. That shared vocabulary between two unrelated measurement systems is the root of most arguments about what a hit is, and it predates Google Analytics entirely.

Universal Analytics stopped processing new data for free properties on 1 July 2023, and for paid Analytics 360 properties a year later. So in 2026 there is no live Google Analytics property anywhere that is still counting hits.

Not sure which numbers in your analytics actually mean anything? Get a free SEO audit and find out which metrics are worth reporting on.

The Hit Types Universal Analytics Recognised

Not all hits were the same. UA classified every incoming packet by type, and the type determined which reports it fed.

Pageview Hits

The default. The tracking snippet fired one automatically whenever a page loaded, with no configuration needed. These populated the Behaviour reports and were what most people meant when they said "traffic".

Event Hits

Fired manually for interactions that were not page loads: video plays, file downloads, outbound link clicks, form submissions, scroll depth. Each carried a category, an action, and optionally a label and a numeric value. This rigid four-field structure is the single biggest difference from how GA4 works.

Ecommerce and Transaction Hits

Purchases sent transaction and item hits carrying revenue, tax, shipping, product names and quantities. Enhanced Ecommerce extended this to the whole funnel: product impressions, add-to-cart, checkout steps.

Social, Screenview, Timing and Exception Hits

The long tail. Social hits tracked shares and likes. Screenview hits were the mobile-app equivalent of a pageview. User timing hits measured how long something took, such as an asset loading. Exception hits logged errors and crashes.

There was no way to tell these apart from the headline number alone. Two properties both reporting 100,000 hits could be a content site with pageviews only and an ecommerce site with a tenth of the traffic and exhaustive funnel tracking. The number described the instrumentation at least as much as it described the audience.

The practical consequence of all this: you could not compare hit counts across properties unless you knew exactly how each one had been instrumented. A site with aggressive event tracking generated several times the hits of an identical site with none, on identical traffic.

Hits vs Sessions vs Pageviews vs Users

This is where most of the confusion lives, so here is the hierarchy, from smallest to largest.

• Hit — one packet of data. The smallest unit. Not a person, not a visit.
• Pageview — one hit of the pageview type. A subset of hits.
• Session — a group of hits from one visitor in one sitting, closed after 30 minutes of inactivity by default.
• User — the person those sessions belonged to, identified by a cookie.

A worked example. One visitor arrives, reads three articles, watches a video and downloads a PDF, all in fifteen minutes. That is:

• 1 user
• 1 session
• 3 pageviews
• 5 hits minimum (3 pageview hits plus 2 event hits), more if scroll tracking was enabled

Five hits, one visitor. Anyone quoting the hit count as an audience size is overstating it fivefold, and the multiplier changes with the tracking setup rather than with the traffic.

Notice that the hit count is the only figure in that list you cannot derive from the others, and the only one that changes if you add tracking without adding traffic. That is the whole problem with it in one sentence.

Universal Analytics also capped hits: 500 per session, after which further hits were dropped. Free properties had monthly collection limits too. These caps were a real source of silent data loss on heavily instrumented sites, and are a large part of why the event-based model that replaced them was designed differently.

Reporting on numbers you cannot explain is a liability. Get a free SEO audit and get a clear read on what your analytics is telling you.

How to Find the Modern Equivalent in GA4

There is no Hits report in GA4 and there will not be one. The model changed rather than the label.

Step 1. Accept that everything is an event now. GA4 collapsed every hit type into a single concept: the event. A page load is a page_view event. A purchase is a purchase event. A video is video_start and video_progress. There are no separate hit types to reconcile.

Step 2. Look at event count, not hit count. Open Reports, then Engagement, then Events. The Event count column is the nearest structural relative of the old hit count — the raw number of data packets collected.

Step 3. Do not treat event count as traffic. GA4 fires events automatically that UA never counted as hits, including scroll, session_start, first_visit, user_engagement and outbound click. Event count is therefore typically higher than the old hit count for identical traffic. This is the single most common mistake in UA-to-GA4 comparisons.

Step 4. Use the right metric for the actual question. For audience size use Users or Sessions. For content performance use Views. For engagement use engaged sessions or the specific event you care about. Event count answers "how much data are we collecting", which is an instrumentation question, not a marketing one.

A useful sanity check when comparing the two eras: pull the event count for a date range, then subtract the automatically collected engagement events you did not ask for. What remains is roughly comparable to a UA hit count, and the gap between the two figures is usually large enough to surprise people seeing it for the first time.

Step 5. Check your event volume against the current limits. GA4 has its own collection and cardinality limits, and they have been revised more than once. Check Google's current GA4 limits documentation rather than relying on a figure from a blog post, including one this old.

Migrating reports and not sure what survived the move to GA4? Get a free SEO audit and start from a clean baseline.

When the Word "Hits" Still Legitimately Appears

The term is not purely historical, which is part of why it keeps causing trouble.

Server log analysis. Web server logs genuinely count hits, and always have — one HTTP request, one hit. Every image, stylesheet and script on a page is its own hit. A single pageview can be fifty server hits. This is the oldest meaning of the word and the most inflated.

The Measurement Protocol. Google's protocol for sending data server-side still uses hit-adjacent terminology in parts of its older documentation, because it predates the GA4 rename.

Legacy dashboards and templates. Looker Studio reports, agency decks and internal spreadsheets built before 2023 often still carry a field labelled Hits, now wired to something else entirely or quietly broken.

Google Ads and third-party integrations. Some connectors and billing descriptions written against the old API still describe data volume in hit terms, because the contracts and documentation outlived the metric.

Other analytics platforms. Some tools outside the Google ecosystem still use "hits" as a first-class metric with their own definition, which may or may not match UA's.

If someone hands you a hit count in 2026, the useful follow-up question is which of these four they mean. The answers can differ by an order of magnitude on the same site.

Limitations and Traps

Hit counts cannot be compared across properties. They are a function of instrumentation depth as much as traffic. This was true in UA and remains true of GA4 event counts.

Historical UA data is gone. Google removed access to Universal Analytics data after the shutdown grace periods expired. If nobody exported it, there is no hit history to compare against, and no, it cannot be recovered from the interface.

UA-to-GA4 trend lines are not continuous. Because the counting model changed, any chart spanning the migration has a discontinuity in it. Treat the two eras as separate datasets rather than one series.

Event count is not a stable metric either. Google has added and renamed automatically collected events more than once since GA4 launched, so an event-count trend line can step up for reasons that have nothing to do with your site or your tracking. Annotate your reports when the platform changes under you.

The 500-hit session cap silently truncated data. If you are working with exported UA data from a heavily tracked site, some sessions are incomplete and there is no flag telling you which.

Hits vs the Alternatives: What to Report Instead

If your goal is to communicate performance rather than to sound busy, the choice is straightforward.

• For audience size — Users for unique people, Sessions for visits. Never event count.
• For content performance — Views, segmented by page path. This is the closest honest descendant of "hits" as most people meant it.
• For engagement quality — engaged sessions, average engagement time, or a specific conversion event. These have no UA equivalent and are the real upgrade in GA4.
• For technical diagnostics — event count is genuinely the right metric, because the question is about data collection volume.

The practical rule: if you cannot explain in one sentence what a number counts, do not put it in a report. "Hits" fails that test for almost everyone who uses it, which is the strongest argument for retiring the word.

Want to know which of your pages actually earn their traffic? Get a free SEO audit and see the numbers that matter.

Final Thoughts

Hits were the Universal Analytics unit of collection: one interaction, one packet of data. They were never a count of people, and the ratio between hits and humans depended entirely on how thoroughly a site was instrumented. GA4 replaced the whole hit-type system with a single event model, so the direct modern equivalent is event count — a number that is useful for diagnosing your tracking setup and actively misleading as a measure of traffic.

The reason this matters beyond trivia is that the same confusion repeats itself with every metric Google renames or retires. Unique page views went the same way, and understanding what replaced unique page views in GA4 is the same exercise applied to a different column.

So when someone asks about your hits, the right move is not to find a number. It is to ask what they actually want to know, then give them Users, Sessions or Views instead.