In this article, you will learn how to segment your products using sales data to better manage Google campaigns. We will focus on segmenting them into best-selling and worst-selling products by assigning dynamic labels to each group. This is a step-by-step framework developed by one of our agency partners, Austin Becker, and the templates and solutions referenced throughout the article are based on this approach.
By leveraging this segmentation, you can create targeted campaigns for your top-performing products, such as the top 10 or top 25, while also addressing underperforming products with tailored strategies. For example, you might allocate higher budgets and aggressive bidding strategies to your best-sellers, while testing new creatives or offers for your worst-sellers.
The real power of this method lies in automating this segmentation process. By dynamically updating these labels based on real-time sales data, your campaigns will always reflect the latest performance trends. This ensures that your Google campaigns remain optimized, allowing you to focus your ad spend on what’s working and quickly adapt to changes in product performance. Additional resources and examples can be found on the Austin Becker Marketing website.
You need an automated system to segment your products based on sales data, because it makes no sense to work with outdated data. Automation means that your data is updated every day or even several times a day, so your data is always fresh.
When it comes to best-selling products, trends can change quickly. For example, inflatable water toys may be among your top 10 best-selling products in the summer, but you won't sell more than a few in the winter. On the other hand, your best-selling product in the winter may be wooden sleds for babies. So you can see that the list of your best products based on sales revenue can change regularly, even on a daily basis.
What is necessary to do if you want your data to be up to date is to automate this process completely so that based on the last 30 days, your items would be ranked according to the revenue in relation to the revenue of other items in the store.
Austin Becker has created a product data template in which he has set up labels for products and created formulas that help automate the process of assigning specific labels to specific items.
You can use this template as a base for your own product list. Using it, you divide them into Top 10, Top 25, Top 50, and Top 100. Your list to always be updated according to the sales revenue for each item based on the last 30 days.
Here’s how you can do it:
You need to import the list of your active item IDs to the sheet you work on. You can do it using Google's AppsScript extension and the ready-made script template at this link:
What you need to do is to get your product feed live URL from your DataFeedWatch account. You will find it in your shop under the specific channel.
Then paste your feed URL into the script into xlmURL. Hit run and voila!
It will take all those active items, and it'll put them in the sheet titled “Title Lookup Key”.
You must ensure that the product data you use is always up to date, so don't forget to schedule appropriate updates.
So between 2 a.m. and 3 a.m. every day you'll have your sheet set up and ready to go with the latest data.
You need to import sales revenue data every day. We'll give you an example if you import it from GA4. The easiest way is to use a data import tool such as Adveronix, SyncWith or Supermetrics.
Adveronix fetches the requested data and inserts it into the spreadsheet. In our example, it involves item ID and item revenue.
You need to schedule an import that takes place once a day at around 4 a.m.
This way you will get fresh GA4 data in a spreadsheet every day.
You should assign the seller's rank based on specific item revenue, and it can all be done with ready-made formulas, which you can copy.
To explain better, what the formula does is take the revenue from your raw data from GA4 which you’ve imported into a sheet and then the revenue gets ranked with another specific formula and then based on that rank you add a label which is a manually entered piece of text, for example top 10 top 25 top 50 or other names you wish. The label is then being added based on the criteria you set up.
Then we have a final output of your IDs in the column on the left and all of the bestseller ranks in the column on the right.
What we have at this point is a process that runs steps 1, 2, and 3 between midnight and 3:00 a.m. every day. Step 4 is a step that depends on steps 1, 2 & 3 taking place.
If you don't run the first three steps, then your sales data won't be updated and you'll be importing old data back into your feed for your custom label. You must therefore schedule a final update inside your DataFeedWatch account once all the previous steps have been completed - this will take place at 4 a.m. This way, you will receive the latest data.
Then, in DataFeedWatch you need to create a custom label - let’s go with custom_label_0:
It's simply a matter of asking the DataFeedWatch system to find everything in column A (item ID in column A in our example) and everything next to it in column B, and assign it the value you have created.
What you are getting is that you have created the attribute custom label zero, and the value assigned to this custom label zero for this specific item ID will be top 10 or for another top 25 and so on.
Segmenting your products by sales data and automation changes the way you manage ad spend. Instead of guessing which products deserve the biggest budget, you can rely on a data-driven system that updates daily.
This dynamic approach ensures that your campaigns are always targeted at your current top performers, whether it's a specific “Top 10” list or a broader “Top 25” segment. You no longer have to worry about seasonal fluctuations or outdated bestseller lists dragging down your performance.
Download the template created by Austin Becker, linked in the article, to start building this system yourself.