Why You Should Ignore Your Sitewide Conversion Rate (And What You Should Look At Instead)

Conversion rate.

We all want it to go up right?

If any of you genuinely just said no, you should leave, QUICK – before the pop up with ideas to improve your ecommerce conversion rate! (Jokes, I haven’t set that up yet).

Some companies use a generic conversion rate figure as a guide for overall success and look deeper to improve.

However, I’ve encountered many online retailers who use this number as gospel and just look to improve that number in all their AB testing.

This is quite frankly, bullcrap!

Comparing year on year differences in conversion rate can be extremely misleading so many other variations can affect the change, such as:

  • Change in device spread – mobile’s increase of device share has slowed but is still growing and it’s likely you have a higher percentage of mobile traffic now than you did last year. This is nothing you can control.
  • More traffic on your homepage – if you’ve been running large brand awareness campaigns, you’ll have more traffic on your homepage. This will affect the overall number for your site (normally in a deceptively positive way).
  • More traffic to your blog – if you are running a cohesive SEO led content campaign through the creation of keyword targeted articles to drive more links to the content and internal links through to your main money pages, you’ll likely also experience an increase in blog traffic. This will likely decrease your conversion rate overall (although might increase your email sign ups, but separate issue).
  • More traffic to category/product pages – similarly to the homepage and your blog, your category and product pages have different conversion rates, and depending on your tactics in different marketing channels the percentage of traffic might vary between the two – which would skew the numbers in comparison.
  • Going Viral – An obvious one, but if you had a piece of content, a product, or story about your brand/site that goes viral and throws thousands or hundreds of thousands of visitors to your site, you’ll definitely experience lower conversion rates.
  • Change in traffic sources – generally speaking, Google Shopping campaigns convert better than organic Google traffic. So if you have just added Google Shopping campaigns and didn’t have them last year – this will drastically change your conversion numbers (and will also add more traffic to your product pages). Also works the other way, if you’ve maxed out the exposure of your shopping campaigns, but have had a big 12-month push on SEO, you’ll have more organic traffic across your site.

So what I’m saying, is that it’s very unlikely that you’ll have the exact same distribution of traffic source, device type, landing page spread as you had 12 months ago.

Let’s look through some examples…

(Note: These are all hypothetical numbers based on real experiences people have come to me and asked.)

Example 1

Let’s say you had a big push on your SEO in the last 12 months, particular focusing on getting more traffic to your categories.

You add more internal links.

You improve your category structure.

You produce great supplementary content on your blog (with internal links).

You build great links, get your products featured on big news sites.

You’re feeling wonderful. But your conversion rate is down.


Month March 2016 March 2017 Change
Sitewide Conversion Rate 2.43% 2.18% -10%

So you look at conversion rate on a more granular level…

Month March 2016 March 2017 Change
Homepage Conversion Rate 3% 3% 0%
Category Page Conversion Rate 1.5% 1.5% 0%
Product Page Conversion Rate 2% 2% 0%

No change. WTF!

So what could be causing the conversion decrease?

No change on each level, but combine it with the traffic mix.

Let’s have a look at the organic traffic by categorising the landing pages of the organic traffic…

Month March 2016 March 2017 Change
Homepage Sessions 10,000 15,000 +5,000 (+50%)
Category Page Sessions 20,000 50,000 +30,000 (+150%)
Product Page Sessions 15,000 25,000 +10,000 (+66%)

Your goal of driving traffic to your category pages (there’s so much potential there) has worked! And even has benefited other page’s traffic too (YAY!).

But at the same time, you’ve reduced your overall conversion rate of your site (which isn’t a bad thing in this case) but driving a lot more traffic to the site.

Example 2

You’ve spent 3 months aggressively trying to improve your conversion rate, you conduct quite a few split tests – some have won, some have lost, many were inconclusive (always the way!).

You compare your performance compared to before you started your work….

Month March 2016 July 2017 Change
Sitewide Conversion Rate 2.43% 2.18% -10%


You’ve done all that work. And your boss is mad that all you’ve done is make conversion rate worse.

But, but, bu…

The implemented winning tests should surely mean overall it increases right?

You look at the checkout flow in Google Analytics and it says the conversion rate through the checkout is better.

You compare the checkout flow on mobile as well, that says it’s improved too.

What might have happened?

(This is one of many options…)

Someone had turned off all retargeting adverts so site traffic dropped about 5%, but wasn’t noticed as organic traffic had gone up 5%, and the retargeted traffic was converting at a higher rate than organic.

So whilst you improved your checkout flow.

Your traffic source mix had changed.

What You SHOULD Look at (And look to individually improve)

At the best least you should look at conversion rate for…

Mobile Traffic, that’s Desktop, Tablet & Mobile

Site Area, that’s homepage, product page, category page (there can be more but these are the basics)

So you should end up with something like this:

Jan ‘17 CR Review Desktop Tablet Mobile
Homepage 4.26% 3.74% 1.86%
Category Page 1.33% 1.19% 0.89%
Product Page 2.13% 1.73% 1.02%

Note: Numbers purely for filling out the table, not real numbers or benchmarks.

You might think this will add a lot of extra workload, but it should be easy enough to set this up within a Google Analytics dashboard depending on the URL structure of your site.

Here’s one I made earlier to get you started.

conversion rate dashboard

How To Set It Up

Homepage – (Default: Landing Page = /) This should be correct straight away providing your homepage doesn’t sit in a subfolder.

Category Page – (Default: Landing Page contains /shop/) You will need to edit these sections to allow for your category pages to be picked up, this might be changing the ‘Landing Page’ field depending on your site structure, or you might have custom dimensions or other ways to track.

Product Page – (Default: Landing Page contains /products/) You will need to edit these sections to allow for your product pages to be picked up, this might be changing the ‘Landing Page’ field depending on your site structure, or you might have custom dimensions or other ways to track.

PRO TIP: Custom Dimensions – If you don’t know how to set these up, Branded 3 have a great guide, go check it out. Note: Not specifically for this type of dimension but the steps you follow would be very similar.

Be advised – this is probably still not enough detail for many companies, some would look add another level of maybe country – although the tables can get very messy so whilst you want more visibility.

You should go deeper in certain areas on a specific case by case basis, more on that now…

What To Segment Further?

Congratulations, you made the first step to knowing the fuller picture of your conversion rate.

I applaud you.


But wait, there’s more…

You can create further segments directly within Google Analytics to see an even more relevant number.


For whatever reason, you can only sell your product within the United Kingdom.

But you look at your Google Analytics and see that you get visitors from over 60 different countries (a part of you thinks ‘Yay’ – but then you realise that this doesn’t help you) and almost 50% of your traffic is coming from these countries).

This traffic is crap, for want of a better word, and you’ll probably see that they don’t convert either (SHOCK!).

Picking up this traffic would be a natural part of your business – so it rarely does any harm, apart from clouding real conversion rate of (actual potential) customers on your site.

How To Set It Up

For our example, you would simply create a segment for ‘UK Visitors’ – which will only show visitors visiting your site from the UK.

If you add this segment onto your dashboard we previously set up you will notice that your conversion rates go up (YAY!), whilst you might see this as a reason to celebrate, in reality, you are just seeing a closer representation of your REAL conversion rates.

There are plenty of other ways you can learn from segmenting this data – from your traffic channel (Organic, Paid Search, Social) – different types of traffic will respond differently so it’s always good to keep an eye on them.

So if you haven’t already done it, go set up your Google Analytics dashboard now.

And start being more aware of your REAL conversion rate.

Leave Your Comment

Your Comment*

Your Name*
Your Webpage