5
9 Comments

Google Analytics is bad for SEO

Having external js makes your page load slow.
On top of that, in Europe it is mandatory to have a cookie banner if your website uses tools like Google Analytics.

Today I modified one of my sideprojects (vabiso.com) removing the cookie banner and the google js.
I replaced it with the cloudflare statistics js.
Cloudflare has minimalistic stats, but for my project they are enough.
https://www.cloudflare.com/es-es/web-analytics/

The result was almost a 20-point difference!
Imgur

  1. 1

    Interesting how many posts there are here about Analytics.

    If anyone is interested, Darwin (https://www.darwin.so) is inviting a few qualified people to use the public beta at the end of the month. Will be a cookie-free (optional) service, and will have a free analytics tier which should be best of class. *(email [email protected])

  2. 1

    Google does not penalize sites using g-tag, Google analytics,

    A good read: https://www.seroundtable.com/google-analytics-no-search-penalty-31361.html

    But it can definitely enhance your user experience which is a ranking signal indeed, but keep in mind that figures from Cloudflare can be highly misleading so do it at your own risk.

  3. 1

    Those performance scores can be a little deceptive I'd say, but performance is definitely important. You can probably make up for that by optimising your fonts/font loading and perhaps reducing JS, like swapping jQuery for vanilla JS if possible.

  4. 1

    I'm kind of amazed how long Google took to respond to Mixpanel with GA4 and even still managed to overlook the cookie issue!

  5. 1

    Google Analytics is bad for SEO

    This statement is very provocative. Google Analytics is sometimes bad for website performance AND website performance is sometimes bad for SEO.

    That being said, search engines ranking algorithms are very complex. It is very likely that a difference in a few "performance points" might not affect SEO in any way. For example, the algorithm could just check if the page loads in under 3 seconds, so if your website already loads in 2 seconds having GA on it, removing it won't improve ranking in any way.

    I am not saying this could not be true, but there is no proof of this being true and it's impossible to test. Accurate testing can only done through A/B testing, having a control version (with GA) and a fixed version (without GA) of the exact same domain, which is most likely not possible.

  6. 1

    This is useful to read. Only thing is, I read something somewhere about how Cloudflare's analytics figures are quite off, because they count things like bots so you get bizarrely inflated numbers. I am tempted to try Plausible as mentioned by @gbourne.

    I also love Amplitude, but I don't know how big the library is.

    1. 3

      I am using the cloudflare js (the free one).
      I know there is a paid version that counts the stats directly from the cdn and no js is needed. I guess in this second case is when you can have problems with bots.

      In my case I had the cloudflare and google js active for a couple of weeks (Both at the same time). The statistics were similar.
      It may be that I barely have bots on my website, I don't know.
      In any case, if you can avoid using Google Analytics ... use anything else.

  7. 1

    I've been using Plausible and has been great. Really low overhead and love the simplicity of the interface.

  8. 1

    Thanks for the Cloudflare recommendation. Seems more than enough for my needs.

Trending on Indie Hackers
Is my Landing page too plain? 18 comments Page 404. Why do we give a F$%k about details in design of our products? 10 comments I analyzed >1,000 indie SaaS projects and bought 6 for over $15M in total! AMA 8 comments Roast my idea: Partnerships (and cross-promotions) for SaaS Founders 6 comments Launch IH: 🎉 Vuestic UI - Free and Open Source UI Library for Vue 3 🎉 5 comments Who here use prediction markets, and for what? (linking PolyMarket, the platform with the highest volume) 1 comment