What to measure, finding your KPI

The art of reporting

As an SEO Manager, properly reporting on the SEO performance of your site is key. By not tracking the correct KPI (Key Performance Indicator) you are taking the risk of making inappropriate decisions, which essentially could negatively impact the outcome of your SEO efforts.

In previous article, we’ve focused on defining an SEO strategy that aligns with your business goals. We’ve also introduced the benefits of working towards a North Star Goal

Essentially, you should be focusing on producing 2 types of reports:

  • A report tailored for the stakeholders or decision-makers within your company
  • A report designed for yourself and your team

Each report should contain elements pertinent to its audience, (we’ll elaborate on this later).

‘Measuring for reporting’ vs ‘measuring your success’

A common pitfall when measuring the SEO performance of a campaign is to measure for the sake of reporting only. In this scenario, too many (irrelevant) KPI are tracked. While this might make a report look impressive, nine times out of ten those reports are not even looked at when shared with your manager or colleagues.

Instead, you should make your report as simple as possible, focusing exclusively on your (business) goals. That’s what we call measuring your success.


Key Performance Indicator

We often overlook what the true meaning of a KPI is, especially when reporting on SEO performance. The reason behind this fact is that an SEO campaign touches on a multitude of measurable components and that we have the tendency of measuring / tracking them all.

For instance as part of your SEO efforts you might need to focus on acquiring backlinks, so you may decide to report on how many backlinks have been acquired each month. While this could make sense if you wanted to report on the performance of a backlink campaign, it does not necessarily qualify as a KPI for SEO. Why? Because it’s not a KEY performance indicator, it’s only a performance indicator.

The obvious question you may have is: how and when does a Performance Indicator (PI) become Key (K)?

It’s actually simple. For a PI to become K, you should answer the following question: does this PI directly influence my North Star Goal? If your answer is 100% positive, then you’ve got a winner.

Keep in mind that you don’t need many KPIs. While there are no guidelines on how many you should be tracked, think quality over quantity.

Reverse engineering your North Star goal

To make sure your reports contain the correct KPIs, you should work backwards: starting with your North Star Goal (which should be a reflection of your business goals) and then follow this approach:

(note: for clarity, let’s assume that our North Star Goal is: ‘increasing our organic entrances by 130% over the next 12 months in the USA, resulting in 340 sales per month’)

  1. Breakdown your North Star Goal in separate elements: in our example, those elements would be ‘organic entrances’, ‘USA’, ‘sales per month’
  2. For each element, fill the following table
      Element            What do I need to measure?      Is there a dependency?
Organic entrancesEntrances coming from organic channel only Yes: Visibility of products and collection pages
USA Traffic going to the US section of site only No
Sales per month Purchase event Yes: Conversion rate of site

3. Revisit the items that have a dependency and list the sub-elements that can influence their performance. In our example, this could be:

(a) Visibility of products / collection pages: organic performance of each individual page, the load time of those pages, the mobile-friendliness of those pages.

(b) Conversion rate: the layout of product pages, pricing, the relevancy of content. 

4. Separate those new sub-elements in 2 categories: ‘can be measured or assessed’, ‘cannot be measured’:

(a) Can be measured or assessed: the organic performance of each individual page, the load time of those pages, the mobile-friendliness. Content relevancy.

(b) Cannot be measured: page layout, pricing.

5. Finally, repeat step 2 for the new sub-elements that can be measured or assessed (i.e. ignore the ones that fall into the category ‘cannot be measured’):

              Sub-Element    What do I need to measure?     Is there a dependency?
Organic performance of each individual pageOrganic entrances of our collection / product pages Yes: Click-through rate
Load time of those pages Average load time of those pages Yes: Our web hosting, page code, 3rd party integration
Mobile-friendlinessAlert in Google Search Console related to pages not being mobile friendly Yes: Page layout, loading time on mobile device
Content relevancyBounce rate and average time on page   No

Once you’ve completed this exercise (with no more dependency), you will have a set of KPI that are 100% aligned with your business goals and, therefore, should be measured and tracked monthly.

In our example, we would measure / report on the following (for the US section of our site only):

  • Overall organic entrances of our site, per month
  • Individual organic entrances of our products and collection pages, per month
  • Average ranking position and CTR of our key pages
  • The total amount of ‘Purchase Event’ per month with a breakdown per product
  • The conversion rate of our site for our organic traffic only
  • The average load time of our site on desktop and mobile devices
  • The average load time of our key pages (products and collection) on desktop and mobile devices
  • The bounce rate and average time on page of our key pages on desktop and mobile devices
  • Alerts coming from Google Search Console related to the mobile-friendliness of our key pages

Note: you may come across a situation where the data you need cannot be pulled from tools such as Google Analytics or Google Search Console. In this case, you will need to get that data from the business department of your company or check if you can automatically integrate them into your report. Google Data Studio (which is the tool we recommend to produce SEO performance report) can connect with a variety of data sources. We will cover this topic later in this article.

Leave a Comment

Your email address will not be published. Required fields are marked *