Ask any person in SEO what they use – you’re more than likely to get a laundry list of tools that work to tackle SEO from all angles. With an explosion of data produced by these tools, decision-making becomes much more complex with the data sitting in different places. You can easily get overwhelmed staring at graphs and dashboards without being able to make that data actionable.
“Data is the new oil”, you’ve probably heard this quote ad nauseam in the past couple of years. Heck, we even got the data to prove it:
But we don’t put crude oil in our cars... do we?
In this multipart blog series, we’ll take SEO data (Google Search Console, crawl data, etc.) on a journey from their source into useful and actionable reports to make informed decisions on what to do next. No fancy or expensive tools are needed, we’ll stick to simple tools we can use for free like Google Sheets and Google Data Studio. You won’t need an Engineering Degree to do this either. We’ll move our way up from simple data processing, throw Machine Learning algorithms in the mix, and bring different datasets together to cross-reference the data to extract true value.
As a final step, we’ll put our designer hat on, and format the reports by adding some color and design to make them shine. We want to see the twinkle in a client’s eyes, am I right?
The Spreadsheet... the good old rows and columns filled with numbers, borders, and colors... you probably love and hate it at the same time. As much as we live in a world where databases store gigabytes (even terabytes) of data that can be queried in milliseconds, with tons of tools out there visualizing those results, I strongly believe the best way to get a feel for your data and iterate on building your reports is by using a plain and simple spreadsheet as a sandbox... fancier tools come in at later stages.
Ready to dive in?
Log into Google Search Console and select the search property of your choice, then go to the ‘Search results’ section. Pro tip: this might not apply to all use cases, but I prefer to work on Domain properties to suck in as much data as possible, then filter in the data processing and reporting stages afterwards if I’m interested in a particular subdomain.
Now select a Date range from the ‘Compare’ tab, my personal preference is 28 days, but any date range works, just make sure you use the same date ranges consistently across the board going forward.
The more the merrier, so check all 4 key metrics: clicks, impressions, CTR and position. Then hit ‘Export’ and select ‘Google Sheets’.
The data is now ready in Google Sheets, separated in tabs, waiting for us to be processed. To understand which search queries have lost visibility and which queries attracted new eyeballs to our website, let’s create a fresh new tab, select the first cell ‘A1’, and enter this formula:
The QUERY formula allows us to select specific pieces from a larger dataset and execute simple aggregate functions on the resulting data at the same time. If you are familiar with the SQL query language, you’ll immediately recognise this syntax, here’s the breakdown:
This leaves us with a list of all search queries and their difference in metrics between the last time period and the previous period. That isn’t very actionable yet, so let’s figure out the top winners and losers for each metric.
To get a better view of the actual winners and losers, we’ll select a Top 10 for each, per metric. Select an empty cell next to the existing data, let's pick G1, and enter:
Another query, this time with a few new directives:
A very similar query can be applied to cell I1, to find the losers. Only this time we need the values lower than 0 and sort the results in ascending order.
For the 1st part in this series, we’ll leave it here for now. This part has taught you how to export the entire Search Console dataset into a Google Sheet, and then process that data by executing QUERY formulas on them to extract meaningful Winners and Losers. There are a few more things you can do we’re leaving as an exercise to try on your own:
In the next part of this series, we’ll introduce Machine Learning algorithms to fuzzy match search queries with a brand name and extract all branded searches, including typos and misspellings. Make sure to subscribe to our newsletter to be notified when this article drops.