10 things to analyse that will improve website performance

A check list of techniques for conversion rate optimisation

What are the most effective data points to track that tell you how well your website is functioning? The purpose of most websites is either to drive ecommerce sales, drive leads or drive engagement with content. Some of these are applicable to all of these, while some are quite specific.

The key metric we should look at is Conversion Rate, the rate at which your users took a key action, or achieved the thing you want them to achieve.

This is by no means an exhaustive list, but it will give you a good foundation of understanding of your user flow.

Key user actions and micro conversion rates

Every website especially ecommerce websites need to understand their key microconversions by session and by user:

  • Item view / Product Detail Page / Info page view rate %
  • Add to basket / More info click rate %
  • Checkout start rate %
  • Add delivery information rate %
  • Add payment information success rate %
  • Product promotion / carousel click through rates %
  • Creative interaction rates %

You can build up your funnel and identify key areas for improvement, counting by both sessions and users, with the primary metric being users that completed each step. The difference between the two will give you key insights about the characteristics of your users and give you a good data point to base outbound marketing to.

Form / Checkout abandonment Rates

I have included this as a separate point because it is so important. Optimising your checkout flow or lead generation form is the single biggest action you can take, because all users who try to convert, must go through this process.

Quit often it is the single biggest point of failure. If you lose 30 – 40% of your users during this step, this is probably about average. Small differences in completion rate can have massive impacts for large traffic websites. Track this, and constantly monitor it.

Common page paths to conversion

Common paths to conversion

You know your landing page conversion rates, but to enrich this data you can count the number of times a page was included within a converting session. You can attribute a page value score based on the number of times a given page assisted a conversion or hindered it.

You will quickly identify the pages that contribute nothing to your performance. Think of the 80/20 rule here, most pages will not! It will focus your attention and give you best practices.

Do this for page groupings as well for a more top-level view before drilling down. It is best to only compare pages that have the same page type, as pages closer to the conversion in a natural funnel will have higher page value scores.

Form completion page referrer

For Lead Generation websites, it is crucial to know the page that referred the user to the form page where they converted. Let’s say you have information pages about different services, but all of them direct to one contact form page.

You can identify the conversion rates of the referrer pages and identify top and bottom performers and analyse further reasons why.

Add to basket rates in detail

If you’re driving shopping or Performance Max campaigns to product pages, it is crucial to know the add to basket rate of these pages. In my experience, add to basket rate has the largest bearing on whether a user converts based on multi session journeys.

  • Product page add to basket rate %, by product category and product
  • Product page add to basket rate % by price bucket – split by page type
  • Product list page add to basket rate %
  • Product promotion carousel add to basket rate %
  • % split of add to basket events by page type
  • Cart page upsell success rate %

Collecting this data will help you identify how your users interact with your website, where they add to basket, and the things that influence that decision, you can strategically merchandise the website to leverage these user preferences.

Conversion rates of a given event

How much impact does an action have on the website conversion rate? For example the user conducting a site search or downloading a file or interacting with a widget. Assessing these will give you an indication (not definitively) of how it impacts conversion rate.

Again, consider where the event is in relation to your natural funnel, and use with caution. Segment the data by page type to get the most accurate results.

On-site search term conversion rate

Take the insight one level lower and drill down into the on-site search terms and their respective conversion rates. Knowing this information will tell you

A) what are the most frequently wished for products, or information that your users want.

B) How effective has that search query been in resolving the user’s issue and achieved a successful conversion?

Knowing this data will indicate which topics your website is best at dealing with, and give you an insight on what the primary convers of your potential customers are, and whether or not you are addressing them.

Conversion and user journey insights linked to SEO search queries

So much focus around SEO optimisation is on volume, because ultimately if it’s free traffic, then more is better right? Well, SEO isn’t free in reality, and it depends if you’re attracting relevant traffic, based on your keywords you could be ranking very high for keywords that have massive volume, but ultimately drive traffic that will never convert.

The key opportunity is to look at the conversion rate by keyword, look at the keywords that contribute to the higher conversion rates and focus your SEO optimisation on them. If you are paying for SEO optimisation, you need to go after those keywords most relevant to your business. You need to utilise Google BigQuery for this, as you need to join the Search Console and GA4 datasets to gain this insight.

Common re-entry pages and conversion rates

If a user has multiple sessions, look at the pages that are commonly entered on their second or third visit. This will help indicate the content that users look for when shopping for considered purchases over multiple session interactions.

Lifetime value of signed up users

The most important marketing metric is LTV, Lifetime Value. If you can recognise a user again from when they signed in, you can attribute your revenue to the first campaign you saw that user. Further still, you can gain a true marketing return on investment, giving you confidence to spend what you need to attracting customers first time.

Standard GA4 will have lots of direct traffic in the traffic acquisition tables. But using BigQuery, and by collecting your first party user ID, we can stitch together previous sessions and interactions and get a view of the individuals (not cookie IDs) that browse then return to your website.


Do you need to know how effective your website is at converting your users?

Get in touch or schedule a free consultation call.


Read more about Digital Analytics

  • 10 things to analyse that will improve website performance

    What are the most effective data points to track that tell you how well your website is functioning? The purpose of most websites is either to drive ecommerce sales, drive leads or drive engagement with content. Some of these are applicable to all of these, while some are quite specific. The key metric we should…

    Read more

  • Privacy Sandbox: What is happening to 3rd party cookies?

    You may have seen some noise online about the phasing out of third-party cookies and the introduction of Google’s Privacy Sandbox. This post is to explain what this means in varying aspects of digital marketing. There’s lots of emotive & negative language to describe this change. Some may call it a “cookie-pocalypse”, be very wary…

    Read more

  • Alternatives to GA4 (Part 1): Where is Digital Analytics Heading?

    What other Digital Analytics tools are available? There are many Digital Analytics Tools that are alternatives to Google Analytics 4 (GA4) on the market today. These existed before GA4, but have come to the forefront because of the consensus that standard GA4 can be limited for many basic use cases. In this article I’m going…

    Read more