We've just implemented Matomo across all of our web properties here at the
Rockefeller Archive Center, and wrote up a blog post outlining our process
and rationale (there are also some links to additional platforms we looked
at) [1]. We're still too early in the game to be able to draw a direct
comparison to data collected with GA but I suspect we'd be better
positioned to do that in a couple of months.

I can't speak in great detail to the question of accuracy but I think Peter
is right in saying that all of these analytics platforms are measuring
things a little differently. Matomo has some documentation about how it
measures visits, [2] hits [3] and users [4] which may be useful.

Hillel

[1]
https://blog.rockarch.org/moving-from-google-matomo-to-put-the-we-in-web-analytics
[2] https://matomo.org/faq/general/faq_36/
[3] https://matomo.org/faq/general/what-is-a-hit/
[4] https://matomo.org/faq/general/faq_43/

On Fri, Nov 18, 2022 at 9:33 AM Andrew Ward <[log in to unmask]>
wrote:

> I don't mean to hijack the thread, but I'd be interested in hearing what
> other analytics solutions people are using besides Matomo. I need very
> little for my web reports--basically a counter on a few specific pages. We
> don't do any higher analysis or events. I've long felt that Google
> Analytics was too much for us.
>
> Thanks,
> Andrew
>
> On Fri, Nov 18, 2022 at 2:01 PM Peter Velikonja <[log in to unmask]> wrote:
>
> > Hi,
> >
> > I am also interested in getting a better understanding of Matomo.  I am
> > collecting
> > analytics for about 40 public libraries, running Google Analytics,
> Matomo,
> > and another
> > program, Clicky, in parallel for all of them; my original hope was to
> > validate Google Analytics
> > results.  I do gain some insights, and this setup is useful for drilling
> > down on anomalies like
> > robot traffic bulges, but mostly I see variation.  Each program gives me
> > *comparable* results that I
> > wish were more closely aligned; side-by-side comparisons are more
> confusing
> > than revealing to me.
> > My basic conclusion is that none of these analytics programs can be
> called
> > accurate, they each
> > interpret visits or users in their own way, and I'm pretty sure they all
> > use a one-way or
> > 'unreliable' socket protocol, so, unlike with a server log, I can't
> assume
> > every item served is
> > represented.  I'd like to develop better confidence in these analytics
> > programs by getting a deeper
> > understanding of how they really work.  Thanks,
> >
> > peter
> >
> > On Thu, Nov 17, 2022 at 11:14 AM Heidi J. Webb <[log in to unmask]>
> wrote:
> >
> > > Chad and others,
> > >
> > > I'm interested as well! We've been testing Matomo too with some
> struggles
> > > with usage reporting.
> > >
> > > We noticed that Matomo is not tracking all hits. At first, I thought
> > maybe
> > > google analytics was tracking more bots or something but my team
> > > purposefully went to a bunch of sites one day recently and only usage
> > from
> > > one of the 5 devices we were using regardless of browser or location
> was
> > > tracked and logged.
> > >
> > > This is currently on my list to investigate more and it sounds like
> Chad
> > > has dug in much more deeply but since the topic made the listserv, I
> > would
> > > welcome input.
> > >
> > > Heidi
> > >
> > > Heidi Webb | Discovery and Technology Services
> > > Upstate Medical University | [log in to unmask]
> > >
> > > -----Original Message-----
> > > From: Code for Libraries <[log in to unmask]> On Behalf Of
> Haefele,
> > > Chad Mark
> > > Sent: Thursday, November 17, 2022 9:41 AM
> > > To: [log in to unmask]
> > > Subject: [EXTERNAL] [CODE4LIB] Matomo
> > >
> > > With the July 2023 Google Analytics transition looming, we've been
> > > evaluating Matomo as a possible replacement.
> > >
> > > Has anyone been robustly using Matomo who'd be willing to chat about
> it?
> > >
> > > I'm interested in general best practices, but also some specific
> > > troubleshooting:
> > >
> > > Matomo was looking promising until I noticed a concerning reporting
> issue
> > > on low-use pages:
> > > -If I look for usage of those pages on a single specific day, Matomo
> > > reports correctly.
> > > -If I run the same report for a date range that includes that same day,
> > > Matomo reports no usage. We've reprocessed the data from raw logs as
> > Matomo
> > > suggests, without improvement.
> > >
> > > Thank you,
> > > Chad
> > >
> > >
> > >
> > > ---
> > > Chad Haefele
> > > he/him/his
> > > Head of User Experience
> > > UNC University Libraries
> > > [log in to unmask]
> > >
> > >
> > >
> > >
> > >
> >
> > --
> >
> > Peter Velikonja
> > Head of Research, Koios LLC
> > http://www.koios.co
> > *Let's reconnect public libraries with the public.*
> >
>
>
> --
> Andrew Ward (he/him/his)
> Digital Services Librarian
> Troy Public Library
>