Categories
Web analytics

Alternatives to Google analytics 4

I’m writing the article sort of in response to panic and frequent questions… what to do next if I liked Universal analytics I hate GA4? Are there any alternatives? I get asked this quite a lot.

Disclaimer: I’ve been using Google stack for a long time, tried Matomo, Snoplow, Piwik PRO, Smartlook, MS Clarity, Hotjary etc. But still 98% of the time I run in GA4 + GTM. Still, I’ve been doing Google analytics in web & apps and later GA4 since 2019, and I’ve done Firebase analytics before which they are based on.

  • My minimum requirements for web analytics:
  • Measuring page views
  • Measurement of customized events.
  • Measuring the source of visits throughout the visit
  • Conversion measurement
  • Ecommerce measurement

We’ll go through the possible tools from simple to more complex.

Just measuring page views

You can find plugins to WordPress (WP Statistics etc,), analyzing server logs (AWstats). What it measures is which page was viewed and possibly the source of arrival to that page.

Pros

  • Maximum simplicity
  • It doesn’t need cookie consent because it doesn’t measure the user, but the content output by the server.
  • It can’t be blocked
  • Data is stored on your server, it is not sent to any other parties.
  • Most of the time it’s free, or you pay a one–time fee for a plugin version or software purchase.
  • It doesn’t burden the user’s browser, ie it doesn’t slow anything down because the measurement is on the server.
  • Super easy to install. In wordpress, a few clicks to install the plugin and activate it.

Cons

  • It does virtually nothing. (ecommerce, conversion, working with users)
  • It doesn’t read the article, it displays the page.
  • Data can contain a lot of bots/artificial visits. Sometimes for small sites even 60% of all data. For larger traffic, this dilutes to maybe 10-20% in real traffic.
  • Don’t expect any support, if for example a plugin breaks with an automatic update, you’re just out of luck and lose all your data.
  • Very often users IP addresses are stored and further analyzed, which is against GDPR without consent.

Target audience

  • WordPress blog with no active online marketing.
  • Unpretentious small blog owner who just needs to know how many times which page was viewed and is happy not to have to deploy a cookie bar.
  • The owner of the tool is the site creator/owner.

If that’s good enough for you, why not? :) For a smaller blog with no active online marketing, it may be fully sufficient.

Minimalist 3rd party analytical tools

They often focus on privacy. Namely Plausible, Fathom, Counter, umami.is and others.

Pros

  • Focus on simplicity
  • They play on privacy.
  • Basic installation is simple, code in the page header.
  • Data is with a third party. No servers to worry about.

Cons

  • Nothing complicated to measure it there, practically you only measure page views. Sometimes it can even do custom events. For some tools you can measure the total order value (not products).
  • There is often talk about privacy, but especially in the beginning these companies didn’t have headquarters or owners. And some companies were based in Central America, for example, outside the EU and the law.
  • Minimal connection to other services.
  • Functionally, they often rely on IP fingerprinting and UserAgent, which is NOT allowed to be used in the Czech Republic anyway without active consent, nor touched by an analytics tool. Personally, I don’t like it because cookies can be deleted, while IP address is hard to change. But some people think that IP address is not personal data even if it IS.
  • There are no people who have experience with this, minimal support.
  • Paying for the number of analytics hits sent per month.

Target audience

  • Small blog, simple lead page with totally basic online marketing.
  • Target is a programmer with a desire like privacy, who doesn’t want a cookie bar, but at the same time doesn’t mind doing an IP fingerprint, even though it violates GDPR.
  • An offline marketing person who hasn’t used Google analytics before and doesn’t do active online marketing.

Generally not recommended, wasted time and money. You will hit a limit very soon and find you get nothing out of it but totally basic stats. Which may be enough for some people.

For me I prefer a WordPress plugin or something better, this is something in between.

Live demo Plausible

https://plausible.io/plausible.io

Smartlook, Hotjar, Crazy Egg, MS Clarity etc.

They are user experience (UX) measurement tools, not web analytics tools. Admittedly they often say they do product analytics because they can do funnel and it’s trendy to say so.

Pros

  • Focus on the visual component of what it can sell to bosses and customers. 😉 which IS important.
  • Focusing on the element on the page/feature in question.
  • Better understanding of the individual and their actions.
  • Good at identifying errors on the site.
  • Linking to questionnaires and other UX features.

Cons

  • You can’t really glean much from heatmaps and visit recordings and it’s quite time consuming.
  • Lacks a general more detailed overview of the site, conversions and traffic.
  • Can’t work with ecommerce products.

Target audience

  • UX specialists
  • Product Managers.

These tools were developed for UX, so there’s no point in pretending they’re for something else. But that also means I can recommend them for measuring UX ;), because that’s what they’re good at. So measuring them in parallel with normal web analytics is not a replacement, it’s a bonus.

Personally I like MS Clarity a lot, it’s free, full of features, it have development.

Matomo

An open source tool I’ve been trying out regularly every few years to see if it improves and doesn’t. Still prehistoric in measurement and appearance. Can run on small sites then terribly careful on large sites as it’s not made for that.

Pros of Matomo

  • The main advantage is that you can install it on your own server..
  • By having it on your own server, the measurements are often not blocked by privacy-oriented browsers and plugins.
  • The data is not sent to third parties.
  • Minimal innovation is still the same for years.

Cons of Matomo

  • Requires time and effort to manage and maintain. If you install Matomo analytics on your own server, you have to worry about its security, updates, backups, and troubleshooting any issues.
  • If you use a cloud service, you have to pay a monthly fee based on the number of visitors and actions on your site and you get rid of most of the benefits.
  • It has limited support and documentation. Matomo analytics is an open source project that relies on a community of developers and users. This means you may not get quick and qualified help if you run into a problem or need advice. The documentation is also quite extensive and complex for beginners.
  • Occasionally it breaks with updates (happened to me 2x ) = no one to help you > you lose all your data.
  • There are a few people in CR who can help you with this.
  • For me minimal development, doomed but still holding up.
  • 2.7% of the market
  • Generally their marketing consists of bitching about GA4 and not making their tool better than Google Analytics .
  • My favorite nonsense is that in some places you can’t filter anything. The tables in the base are missing the row with the summary value and other unbelievable things you take everywhere as totally basic.

Target audience of Matomo

  • Website owner with no interest in digital marketing but wants web analytics 😉 .
  • Tool owner is the owner of the site.

Live demo Matomo

https://demo.matomo.cloud/

Piwik PRO

Piwik PRO is a tool that was created by breaking away from the original Piwik (now Matomo). Compared to today’s Matomo, they are stepping into development and making the tool more usable, working with the community, trying to.

Pros of Piwik PRO

  • Probably the closest thing to the Google Universal analytics experience.
  • It’s a combination of Analytics Custom Tag Manager and Consent Manager
  • Ability to collect and process data in compliance with GDPR and other privacy legislation.
  • The ability to create custom segments, reports and dashboards to suit your organization’s needs.
  • When you pay you can have it on-premises or in the cloud etc.
  • Up to 500k visits per month in limited functionality for free
  • Servers in Germany.
  • Connection to Google ads, Google search console
  • In paid version connection to Google big query, Tableau, Looker studio etc.
  • 1.4% of the market

Cons of Piwik PRO

  • 500k visits is not much, so it wants to pay soon and a lot.
  • Need to install and manage Piwik PRO analytics on your own server or use cloud solution from provider, keep an eye on it. (Free version is in cloud and it is possible to have paid one)
  • Not many people to help you with this.
  • There is documentation, but it is more complicated to set up well the first time.
  • The combination of “GA + GTM + Google consent manager” can be confusing, not everything can be set up.

Target audience for Piwik PRO

  • Small to large e-shops and websites.
  • Active online marketing.
  • Owned by the digital marketing department.

The target audience for Piwik PRO analytics are organizations that need full control over their data and want an alternative to Universal analytics. These include e-tailers, media companies, banks, insurance companies, government institutions and non-profit organizations.

For me, a pretty good alternative, although I warn you in advance it won’t be that easy and you will realize how Google Universal analytics were terribly sophisticated. Certainly better than Matomo.

Piwik PRO donť have live demo, without registration.

Update 2024-09-09, hands on experience.

After 9 months of measuring and working with data, I have to admit that this is a good tool in the Universal analytics style. It’s just quite expensive, even for a smaller site.

Installation:

+ GTM Template by Simo Ahava. The limitations come from the tool and not the template
– Dimension numbers 🙂. The thing disgusting already from Universal analytics.
+ You can have up to 200 dimensions and 20 product
– Can’t ecommerce impression view and click
+ In Simo Ahava template there is a GA4 ecom layer translate on Piwik PRO ecom.
– When you want to submit dimensions with pageview. Starting Config… Setting Dimensions.. Submitting page_view
+ Integrate consent with Google consent mode
– Without consent with IP address and device data, visit and what hit = visit = user will break
+ (Which can be fixed with custom reports and adding metrics for inputs.)
+ Namespace separation from Matomo
– Heatmaps with a separate script (could be straight in the template)
+ Custom something like tag management (GTM), but nothing as good as GTM, but it can be used.
– Custom cookie bar/consent mode, I like custom cookie bars, this can’t be set up enough.

Reports

+ Custom realtime dashboards.
– Realistically, this report unless you are a medium that deals with realtime articles, you won’t use it practically
+ It has a lot of basic reports that can be reasonably expanded too.
– Basic reports can’t be modified or hidden. (But you can almost always create a new custom report from them)
+ You don’t wait for data processing (ga4 2-3 days), you have the data +- immediately, including conversion (ga4 up to 12 days)
+ Multiple attribution models (linear etc. ie old ones from UA). There is nothing data drive, which is both a plus and a minus 🙂
– Without paying the data is deleted after 14 months. (So back up to GA4 where the data stays.)
+ Everything is fast, nothing crashes.
– Without paying, the connection to BQ is missing. It’s only for the Enterprise plan (11k+ euro/year).
+ For frew export there is only API.
+ Usable integration into Looker Studio.
– Really clunky Google sheets plugin, as it is only disposable and doesn’t allow updates or editing the export.
+ Connection to Google Search console and Google ads.
– But it’s not clickable, but you want it a project in Google cloud, permissions and keys.
+ There is no sampling, but you can turn it on yourself if you want to render the report faster.
– Traffic from search engine “Seznam.cz” is not marked as organic (I wrote to the developers)
+ Very interesting CDP (Customer Data Platform) in the paid version, but I haven’t seen it in practice yet.

Praxis

– Very few people in the Czech Republic can help you with this. Realistically, I don’t know of anyone who offers it (1.4% of the market).
+ But you can talk directly to the CTO at MeasureSlack.
+ The tool is hosted by you or in Germany.
+ The tool announces a lot of things about privacy
– But at full max privacy setting, it breaks the visit and the user and ruins the reports.
– Above 500k events it starts to price up very quickly , from 600k events for 72 Euro to 11k Euro/year and beyond.

As a result, I believe that Piwik Pro will find its customers. They do a lot of things right. I could see it as an alternative for smaller sites that don’t push marketing or as dual measurement with GA4. I definitely recommend a lot of analysts test it.
The current probably recommended combo for a smaller site might be GTM + GA4 + Piwik Pro + MS Clarity.

Why do I use Piwik Pro myself if I have GA4 everywhere?

+ When I want realtime data from campaigns/events.
+ When I want traffic data within the day (and not 2 days later)
+ When I want to measure conversions and I want to know it now (and not in 12 days)
+ When I have a small site and I don’t want to go to BQ to see the whole data and not just 40% of the data like in the UI.

Snowplow analytics

Snowplow analytics is an open-source platform for collecting and analyzing granular and rich behavioral data tailored for the data application a user wants to build in their own cloud.

Pros of Snowplow analytics

  • The tool addresses the collection of measurement data.
  • You can do anything in there, custom data schemas, maximum customization.
  • You can measure very unique monetization models through it.
  • Measure everything from web, to apps, no limits.
  • Full control over the data, which is great for healthcare, banks etc.
  • Based in London, active community.
  • You pay for your own cloud etc or pay for hosting with SnowPlow.

Cons of Snowplow analytics

  • You don’t get any nice user interface just a database + collector endpoint. So really only for tech geeks.
  • Need to keep it all alive on your own machines/cloud.
  • Pretty hard to get it up and running, you need people who are pretty technical and will design everything from scratch for you straight away. Which is a really big challenge.
  • You have to build everything from data, so you also need to clean the data, count it etc. so you also need solid data analysts. Both technical and for visualizations.
  • 1.4% of the market

Target audience of Snowplow analytics

  • Really large and complex web and other applications.
  • Own the entire data analytics / business intelligence department.

A tool for really hardcore analysts who have a hardcore business riding on data and has totally unique data and data models. Nothing for normal companies. Target companies with maximum emphasis on privacy and or data models what don’t fit the rest of the analytics tools.

Snowplow doesn’t have a demo

I recommend you try Snowplow micro. Or request a demo trial https://try.snowplowanalytics.com/

Adobe Experience Cloud

It’s a fine-tuned adult analytics tool for corporations and it’s really expensive, it takes big contracts for a long time. It’s pretty standardized and it’s got quite a quality.

Pros of Adobe Experience Cloud

  • An adult analytical tool, containing really everything.
  • Years of development, more recently with a drive to innovate.
  • Provides scalability, security and reliability.
  • Good for working in large teams.
  • Good for super-sized corporations and banks.
  • Strong connection to the Adobe ecosystem.

Cons of Adobe Experience Cloud

  • It’s really expensive and requires long-term contracts and commitments.
  • It can be complex to set up and use, especially for smaller and medium-sized businesses that lack resources and expertise.
  • Specialists who can do it are super expensive and very often pretty much keep their know how.
  • Can be difficult to integrate with other tools and platforms that are not part of the Adobe ecosystem.
  • Can be limited in its ability to customize and innovate because it is based on standard solutions.
  • Beware of new features, very often they don’t work very well.
  • It depends terribly on who sets it up for you and how, by there being an awful lot of settings the result can be pretty hell to use.
  • 5% of the market

Target audience of Adobe Experience Cloud

  • Large multinational corporations and their websites.
  • The owner is a foreign corporate 🙂 .

Adobe Experience Cloud is designed primarily for large and global organizations that need to solve complex and diverse online marketing challenges. These organizations typically have large amounts of data, customers and channels that they need to manage and optimize effectively. They also have the budget, staff and know-how to leverage the full potential of Adobe Experience Cloud.

You don’t buy this stuff, you get it in corporate.

Google analytics 4

Are Google analytics 4 so terrible?

No, it’s more of an emotion, bitching about the fact that people had Google Universal analytics taken away from them, which were free and too good. A lot of people have forgotten how long it took to learn Universal analytics and how complicated they are for a newbie.

Pros of GA4

  • Free connection to Google Big Query, which is a feature for advanced analytics
  • Measure web and mobile apps in one place, as well.
  • Work with cookie consent and the ability to measure without cookies.
  • Google ecosystem… Google ads, Google search console, Google tag manager, Youtube, Google cloud, Google consent mode etc.
  • Already resolved the issue with the European Union over privacy and data transfer between Us and Eu, which will surely be challenged again.
  • Over 85% of the market. So there are plenty of people with experience.
  • It’s standard, in a year or two everyone will know how to do it.
  • Rapid development, there’s always news and improvements.
  • Using machine learning to calculate user behaviour, what they didn’t have cookies for, predicting future user behaviour, identifying key customer segments etc.
  • For me, easier to learn than Universal analytics.
  • The whole interface and reports can be more tailored to the project. Custom menus, custom reports.
  • Can do very complex and complex audiences for GA4 and export those to Google ads.
  • Conversion unification with Google ads.
  • Connection to all classic tools like Looker studio, Google sheets, Data API, Google big query etc.
  • Much of the basic setup can be done through the Managment API. Plus there’s a cool project by Ben Kuehn on setup via Google sheets.

Cons of GA4

  • User interface
    • It’s confusing, there’s a reporting section and an explorer and advertising. Each part behaves a bit differently, has different rules.
    • Restriction of filtering options in UI (will be solved in the future)
    • There are not so many pre-made reports. And when you make them the finished new custom reports can’t be copied to other GA4 property.
    • Data computation in the interface that takes mostly 2days.
  • Thresholding, hiding data identifying individual users
    • Confusing GA4 reporting identity defaults, which spoils the impression due to increased anonymization.
  • Need to learn how to use Google big query if you want to do advanced reporting or analytics.
    • Limit of 1M exports to Google big query, then it wants to bypass or pay.
    • Uncomputed data is more difficult to understand and work with.
  • You have to do a large part of the design of an advanced implementation. Which paralyzes people because there aren’t many publicly known best practices yet.
  • Prohibition on working with personal data inside GA4.
  • Angry users, which is a trend.
    • A super tool (UA) that was free was taken away from people.
    • It’s different. People will have to learn it, but that’s true for any new tool. But they don’t want to invest the time to learn it again.
    • People don’t have experience in it.
    • It went off “completely without warning”. So there’s a lot of people who are terribly surprised and ignored tons of emails and warnings to switch from UA to GA4 can’t find any free analysts now because everyone needs them.
    • Analysts are exhausted because it’s multiply challenging to work in it due to ignorance, uncertainty, inexperience. It’s very easy to bitch about it and not address it.
    • Working with the community has been lousy from Google for the last two years, but it’s gotten a lot better, there are tutorials, videos, certifications.
  • Sampling for GA4 UI & data API.  (updated)

Live demo GA4:

https://support.google.com/analytics/answer/6367342?hl=en#zippy=%2Cin-this-article

Want to get away from Google analytics 4?

My recommendation is to measure in parallel in GA4 and in the new tool. Because it’s quite possible that you’ll find that while GA4 is “supposed” to be terrible, much of the alternative tools are multiples worse. So don’t burn your bridges on this one. Most of the time, the misunderstanding of GA4 and the subsequent hatred is just from not having the basic setup right and no one explaining it to you with real examples of your project.

The first companies started running away from Google analytics with the advent of cookie banners, and others tried it for years to follow. It was about 5% with GA having 72-87% of the market, according to various measurements and studies, so there was no big exodus. Those who tried it then very often went back to GA4. First they just learned to use it, they also hit the reality of the alternatives and found them multiples harder to use.

What about me? I tried different variants of analytical tools. As a data geek, I could very well imagine using Snowplow for some projects for reasons of complete freedom and security.

In real life, I stick with GA4. Most of what annoyed me I already understood thanks to 4 years of experience with GA4 (GA web & apps), now I know 95% of it by heart again as with UA. Realistically, I don’t feel like I can see much of a difference there anymore except for a few details. There are more options as I tackle things, which can be challenging at times to figure out correctly.

It is necessary to learn to do things in GA4 according to the current “do your best” and not to worry that it will change in half a year, i.e. always adjust it for the moment and learn to live with changes. A lot of the process of learning with GA4 will sit you on your ass and your ego will crash if you just don’t know for a while and you have to learn to say “I don’t know” and then try to learn and sometimes even give up and put the problem off until someone writes something about it or you meet another analyst and chat. Certainly the ones who don’t screw up GA4 will be multiply valuable to the market and be more mature analysts. Sure it will piss a lot of people off, but that’s fate. But you have to remember that maybe 80% of all clients will have GA4 for the next 10 years, so you can’t really escape it and if you avoid it you’re only hurting yourself, so I recommend you bite the inner block and get on with the learning. Will it be easy? No. Is it worth it? Yes.

Of course there is a bunch of other tools I haven’t had a chance to properly test yet PostHog (I’m testing on this article :) ) , Amplitude, Mix panel (I have on one project but it’s beyond me.). The more tools I tried, I realize how Google Universal Analytics were fancy, oversized and totally unbeatable for its free price.

Realistically I measure to test on this site via wp plugin WP Statistics, GA4, Umami, Matomo, Piwik pro, MixPanel, Amplitude and MS Clarity. The goal is to do an evaluation after a few months to see how it measures or doesn’t measure as well.

Updates

Update 2024-01-12: Personal experience with Amplitude. Underestimates by about 12%. Without paying for it, it’s a terribly limited tool and realistically unusable. I like a lot of the principles there, like naming and categorizing events and their owners etc.

Update 2024-01-12: Personal experience with PostHog. It’s such a great developer tool. It’s not for complex visualizations, but it can already do notebook. AB tests only work if you host it at your place or pay for it. The surveys didn’t work for me realistically and I tried hard (they did a couple of updates, I guess it works now). After disabling/removing the code, it shows me in the data that there is still traffic in the low teens. As far as measurement goes it’s rather HIGHLY oversized, ie it treats it’s bots as real users.

Update 2024-09-09: I’ve made an update around Piwik Pro

Update 2024-09-11: Amplitude has made an update to the UI. I’m not even able to do a source/medium report here, much less a source/medium/campaign 🙂 .

.