24 hours are correct but 7 days, 30 days are not

Back in “good old days” the WPP 30 days showed the same numbers that GA did show

That’s because after 2.3.2 -I think, it’s been a long time- the way WPP updated views count changed from AJAX (Javascript) to template detection (check if single.php was loaded).

Since most spiders / bots out there can’t run Javascript (which is something I wasn’t aware of back then), versions prior 2.3.2 successfully filtered out their page views because the AJAX call was never executed. This is why the statistics would be close to what GA reported. After that change though, WPP started counting their page views as valid ones, inflating the statistics.

Of course, I wouldn’t have made that modification if I knew that. Unfortunately, a long time had passed before I realized what the problem was. WPP 3.0.0 could have been released months ago actually, but I decided to spend a little more improving the plugin and making sure old bugs were fixed for good.

Anyways, this is why the statistics were not that accurate. Unfortunately, what WPP has registered prior to 3.0.0 can’t be corrected so old stats (weekly, monthly, all-time) will still reflect visits from bots.

In your case, however, something’s odd. There should be more visits on WPP than on GA. You’re reporting that GA is actually displaying more views than WPP does. Are/were you using a caching plugin, by any chance?

A long post, I know. If you actually read all that, thank you for your patience hehe.

Something on your site is / has been / was preventing WPP from updating the views count, but it’s hard to tell without further info and/or checking directly on your site.

Possible solutions / ideas that come to my mind now:

• Clear WPP’s data to have a clean start and detect if WPP is correctly updating the views count against what GA reports. I know, you probably don’t like this solution because you’ll lose your data. Another approach would be to delete only the cache data (last 24 hours / week / months) and preserve the All-time data.
• Keep using WPP 3.0.x for a month, and monitor how the stats behave during that period.

If the gap between WPP stats and GA is still too big after that (there will always be a difference, though), then I’d need to take a look into your site to figure out what’s going on.

I don’t have a problem in clearing all data. I just don’t see the point in doing it. How should that help?

You’re right, I was a bit tired when I wrote that. Just go with the second option, stats should normalize after that time.

How about last 7 days? Is there still a huge difference between GA and WPP?

While a difference between GA and WPP is something to be expected, what you’re reporting doesn’t seem right.

I’ll cook up a test version for you to try, and I’ll run some more tests myself. I’ll post back here by Sunday. I’m going to be needing your support to check a few things as well, hope that’s OK with you.