Jetpack Site stats about 30% of correct amount

Hi there,

I checked your Jetpack connection and it is working perfectly and no issues I found on our side. Let’s come to stats:

All stats tracking systems collect and report information differently.

For example, Jetpack’s stats do not include visits from logged-in users by default. Many site owners find this useful, since tracking your own visits artificially inflates your traffic. That might be one explanation for the variance between your other stats system and Jetpack’s.

Jetpack also does not report visits in real-time. That is, there’s a bit of a lag between when visits occur, and when they are reported in your stats.

Additionally, there are certain search engine robots and “spiders” that Jetpack doesn’t count, and referrers that are considered spam referrers by Akismet (our spam-catching plugin) will not be reported in your stats. Other tracking systems may be less discerning in what they count toward the stat they report to you, and therefore give you different numbers than we do.

Finally, the different rates at which all these trackers are blocked by the different ad-blocking browser extensions can also affect your stats.

For all of those reasons, different stat tracking methods will rarely, if ever, be in sync. You may find it worthwhile to use tracking data from a variety of stats systems to develop a broad picture of your visitors’ behavior. ??

hi
yes I fully understand about the differences in the details of visits/hits/page views. I’ve been doing this for over 10 years.

however there is something wrong. I am talking about **thousands** of page views per day lower than what it should be.
I take your point about the views of logged in users but that simply wont account for thousands of missed views.

I seem to have an issue like this once every year or so. where there is an inexplicable loss of traffic.

Perhaps there could be a link with the problem and the interaction of Jetpack with Google Site Kit? I’ve had issues with Google Site Kit which sometimes incorrectly reports on direct and indirect traffic (but that is not the case here as, as i say, GA seems to have the correct figures)

(my hosting company stats also show normal bandwidth being consumed etc etc). so it specifically is something wrong with the stats that are showing on jetpack.

• This reply was modified 2 years, 11 months ago by the5krunner.

Hi @the5krunner

I take your point about the views of logged in users but that simply wont account for thousands of missed views.

That’s true, but the point about bots could certainly account for it. There may be certain bots that Jetpack is blocking, where Google is more lenient and counting those visits. If a lot of your traffic is from these bots, then that could account for the difference.

Are there any specific referrers you can see in your Google stats that are missing from Jetpack’s stats?

i’ve run my site for many years, I know my baseline stats and how they relate to the content. It is THOUSANDS below what it should be.

there are virtually no outbound clicks and virtually no Search Engine Terms showing. That is a clear sign to me that something is not recording properly.

Your bot point: Bots don’t make outbound clicks. so any change in bots wouldn’t account for the drop in outbound clicks

I don’t really have a view of what is shown on Referrers. If I look at users from organic search then the figures yesterday are in line with where they were a month ago (which I would expect). Actually, they are slightly higher yesterday.

As I say it doesn’t anywhere near tally with what Google Analytics shows by thousands per day.

I don’t really have a view of what is shown on Referrers. If I look at users from organic search then the figures yesterday are in line with where they were a month ago (which I would expect).

We were hoping for referrers so that we could try to discern if perhaps Jetpack Stats are filtering out some views that we may consider to be bots or spam.

—

Beyond that, I see that you are using LiteSpeed Cache. We’ve had issues with the JavaScript combination and modification in that plugin. Could you disable those features for a bit and see if that makes a difference? I’m not sure it would, but it might be worth trying.

hi thankyou
here is some further info and responses:
—————
another example that might point to amp vs non-AMP traffic:

in the case of the page below, the google analytics traffic for AMP page is close to the total traffic reported in Jetpack. the AMP and non-AMP versions have a broadly similar level of traffic. So, in this case, the jetpack traffic is about half what it should be.

/2022/04/06/coros-permanent-price-cuts-new-models-selected-rebates-from-now/?amp=1
/2022/04/06/coros-permanent-price-cuts-new-models-selected-rebates-from-now/

I recently updated to AMP 2.2.3 but was previously on 2.2.2 in any case. I update plugins as soon as they are released usually. I have rolled back to 2.1.4

Litespeed: I manually record the changes I make to Litespeed Settings and the last changes I made to it were on 12 March when I disabled the two (newish) UCSS settings. So there may be a conflict BUT it wasn’t showing from 12 March to the end of March, give or take. Therefore I doubt Litespeed is the cause.

Litespeed: I’m not exactly sure which settings you are referring to. JS Combine is disabled in any case, maybe you meant that one? (JS Combine External and Inline is dependent on that setting and so is disabled in any case)

JS Minify is enabled. I assume you weren’t referring to that

PHP: I was on an old version of PHP (7.4 , so I’ve updated to 8.1 to see if that makes any difference

I’m happy to try other suggestions as there really is a quite significant problem. My guess would be that this is related to amp plugin/jetpack/google tag maanger/and-or Site Kit by Google

might be no hits recorded on homepage?

Hi @the5krunner

To test this further, I’d highly suggest pausing or disabling any settings that alter or make changes to JavaScript, since Jetpack Stats relies on a script to track the stats on your website.

The stats for the past days won’t change when doing that since they weren’t taken into account in the first step; however, doing this will make it more clear in the next days whether a setting caused some troubles there.

AMP shouldn’t cause any issues with that, as long as you are on the Transitional mode.

i rolled back key software like litespeed, AMP, Google site kit and some others.
i separately note all changes i have made to css/js features in litespeed, I reviewed these historic changes. None were made recently…all were made many months ago when everything was working correctly.

so i have effectively done what you suggest.

please can you be more specific.

it seems strange that a fundamental part of WordPress is so easily broken when GA works correctly. Surely there would be tests you could suggest i run if i were serving ads via jetpack and those ads were impacted by low stats?

These are most of my active plugins
`add widget after content
ads.txt manager
akismet (old version)
AMP – latest on full amp (previosuly on READER mode)
cdn enabler
classic editor
classic widgets
easy google fonts
easy table of contents
fixed widget
FMTC affiliate disclosure
GN publisher
gtmetric
jetpack (latest)
litespeed (latest) – keycdn used, quic.cloud used
paid member subscriptions
reading time wp
redirection
site kit by google
wp rollback
wp review pro
yoast

• This reply was modified 2 years, 11 months ago by the5krunner.

Hi there, @the5krunner,

it seems strange that a fundamental part of WordPress is so easily broken when GA works correctly. Surely there would be tests you could suggest i run if i were serving ads via jetpack and those ads were impacted by low stats?

I am afraid there isn’t much more we could troubleshoot to find out why there is this gap in the stats figure between Jetpack and Google Analytics.

By reading both threads you created about this issue, and the whole conversation above, I do think that the very reasons why you see this gap is what Animesh has explained in his first reply:

`Jetpack’s stats do not include visits from logged-in users by default. Many site owners find this useful, since tracking your own visits artificially inflates your traffic. That might be one explanation for the variance between your other stats system and Jetpack’s.

Jetpack also does not report visits in real-time. That is, there’s a bit of a lag between when visits occur, and when they are reported in your stats.

Additionally, there are certain search engine robots and “spiders” that Jetpack doesn’t count, and referrers that are considered spam referrers by Akismet (our spam-catching plugin) will not be reported in your stats. Other tracking systems may be less discerning in what they count toward the stat they report to you, and therefore give you different numbers than we do.

Finally, the different rates at which all these trackers are blocked by the different ad-blocking browser extensions can also affect your stats.

In a nutshell, some visits are recording and we really can’t expect that they will match 1:1 with another tool, like Google Analytics.

You can use this to your advantage to compare the stats and grab more information about the traffic you get on your site.

Hope that answers your question!

hi stef thank you,

i’m sorry that’s wrong tho

As I said, I changed the WordPress Jetpack settings to include all logged-in users (except me as admin). it makes no difference or very little difference.

I fully understand they can’t match 1:1. However, for many, many months they have broadly matched. and a broad match is fine.

My site is a reasonably popular one with MILLIONS of user pageviews a year (not hits, that’s more and includes robots etc)

Yesterday it is 1:3, which is patently wrong and unhelpful to me.

Example: All the traffic to my homepage is not recorded. It has zero traffic today and yesterday and a few other days too. That’s clearly wrong…GA says so.

this is not a robots or logged in user issue. I would bet a lot of money on that.

Plus there are zero outbound clicks being recorded. Normally I have hundreds or thousands a day. How is that explained by logged in users and robots?

AMP is set as transitional

• This reply was modified 2 years, 11 months ago by the5krunner.
• This reply was modified 2 years, 11 months ago by the5krunner.

14-April – stats today on wordpress are effectively zero.
there REALLY is something wrong. Stats in GA are in the thousands, as normal.

please can i have some help or please can i have this escalated?

thank you

• This reply was modified 2 years, 11 months ago by the5krunner.

I am disabled AMP transitional as that has clearly broken the stats totally

I am trying AMP legacy, tho that appears not to rectify anything

I have reconnected and reconfigured jetpack (again counting logged in users)

I have disabled my ads as that loads lots of 3rd party .JS

still no stats in WP…still ticking over nicely in GA

Hi @the5krunner,

I can see that there is a difference in stats figures between this April and the other months. You reported zero stats for Apr 14th while I can already see 136 views and 46 visitors: as said, the stats are not in-real time and can take a while to update.

I also noticed a slight mismatch while syncing data, which I have just addressed.

Let’s do the sync work (it can take up to 24 hours to complete) and then let’s check again if the discrepancy is still huge.

thank you
yes the stats are now >100 and that happened once i totally disabled litespeed and also moved back to transition AMP.

if you are tinkering at your end should I stop experimenting here?

I am uninterested in slight mismatches of even a few hundred hits/views it is the huge mismatches that I am very concerned about ie thousands.

thank you.