[Video] Matomo and Conversion Rate Optimisation

TRANSCRIPT: Okay, now all looks good. Thank you for joining us today, Frederick. The floor is yours.

Thanks. So thanks, everyone, for being here. My name is Frederick, and I’m going to talk about Matomo Conversion Rate Optimisation today.

Just as a reminder, you can join the chat room on the matomocamp.org website. Go into the talk card and follow the link to join the chat if you have questions.

Here’s what we’re covering today: I’ll introduce myself, then go over prerequisites for CRO with Matomo. We’ll look at how to find your first optimisation ideas, how to prioritise them using CRO frameworks, how to use Matomo UX analytics in the process, how to write your testing hypotheses, and how to run A/B tests with Matomo.

I’m a digital analytics and CRO consultant. I’ve worked with many companies and public sector organisations, and I specialise in open source web analytics. I was also the pedagogical director of a digital marketing master’s programme, so I’m experienced in teaching and curriculum development. You can find me on LinkedIn.

To do CRO with Matomo, the first step is defining performance. Performance expresses the level of achievement of objectives. You need to define what the business goals are. For example, an e-commerce site may want to increase the proportion of visitors who purchase. Then you define the achievement level using indicators, thresholds, and segments.

Use two key tools: the KPI framework and the measurement/tagging plan. A KPI framework is a document explaining how you’ll measure performance across different levels of user engagement. You always start with the business goal. At the reach level, you might want to increase site visibility. At the engagement level, you may want to increase qualified users or grow your audience. At the conversion level, your goal is actions with direct business impact. At the advocacy level, you want to build relationships, increase CRM contacts, and grow social engagement.

Each business goal has macro indicators. For example, to increase traffic volume, your KPI might be total site visits. Include metric descriptions, thresholds, Matomo metrics, and data sources. Note that external tools like SEO platforms or social media tools may also be involved. Segments and calculation logic should also be included.

The measurement plan lists actions to be tracked using Matomo, mainly through events. For example, on a product page, you track specific user actions. You define the event structure (category, action, name) and list all events relevant to your goals. This enables both high-level KPIs and granular marketing insights.

Once you’ve built your measurement plan, create the tagging plan. This is the technical specification for dev teams to implement. It defines the data layer structure, tracking code, and custom dimensions for segmentation and detailed analysis.

Now that you have your KPI framework and tagging plan, you can start optimising.

Let’s define CRO. While many define it as increasing the percentage of users who complete a desired action, this can be misleading. For example, dropping prices by 90% might boost conversion rate but harm overall value. So it’s more accurate to think in terms of conversion volume and value, not just rate.

To find optimisation ideas, compare key indicators between segments. For instance, if your average conversion rate is 2%, but mobile is significantly lower than desktop, ask why. Often, mobile users browse first and buy later on desktop. But assuming sessions start and end on the same device, explore differences in checkout funnels.

Use Matomo’s funnel reports and apply segmentation. If the payment step has drop-offs for mobile users, that’s a red flag. You can confirm hypotheses with Matomo UX features.

Before that, prioritise optimisation ideas using CRO frameworks. Heuristic analysis involves reviewing your own site and noting what feels off, but it’s subjective. Relying on data is better. Some frameworks include:

• PIE (Potential, Importance, Ease) by WiderFunnel
• ICE (Impact, Confidence, Ease) by Sean Ellis—though confidence is subjective
• Hotwire’s binary checklist for scoring tests
• PXL framework by ConversionXL using a 0–2 scale

Use these to score your ideas and prioritise.

Once you identify a big drop (like mobile users abandoning at payment), use Matomo’s UX features:

• Session Recordings: View playback of real user sessions to see what went wrong.
• Heatmaps: Visualise where users click (or don’t). Confirm if a button is too low on the page.
• Form Analytics: Identify which form fields cause drop-offs or errors.
• User Feedback: Trigger a survey (e.g. on exit) to ask what went wrong.

You now have a clearer picture. Let’s say mobile users don’t complete the payment step because the « accept terms » button is hard to see. You confirm this with session recordings, events from your tagging plan, heatmaps showing low interaction, and poor click rates. Formulate your hypothesis using the format:

If I [make this change], then users will [do this instead].

Example: If I make the accept terms button more visible for mobile users, they’ll click it and complete their purchase.

Once you define your hypothesis, set up an A/B test. Show one version to a group, the original to another, and compare performance. Use a calculator to determine sample size and duration based on conversion rate, confidence, and effect size. For example, with a 3% base conversion rate and a 50% lift goal, you may need 23,000 visitors per variation.

Then configure the test in Matomo:

• Name the test and write the hypothesis
• Set the number of variants and target pages
• Define success metrics and thresholds
• Set traffic allocation and variation URLs
• Add start/end dates and insert the test code into your site

You’ll see metrics like conversion rate by variation, detected lift, and statistical significance.

For example, changing a top CTA from generic to job-related increased job applications. Matomo showed a clear winner based on significant performance improvement.

That wraps up the session. If you have any questions, feel free to post them in the chat. Thanks again for attending MatomoCamp.