We will walk you through some recommendations for setting up an efficient e-commerce advertising campaign.
At M19, we believe E-commerce Advertising should be easy. We try to enable our customers to focus on business matters and delegate the PPC advertising automatically. Nevertheless, you may be interested to learn more about the principles which guided us when we defined the campaign structure which our AI engine is implementing to run Amazon advertising.
So, we will walk you through some recommendations on how to set up efficient PPC Marketing campaigns. This content speaks mostly of Amazon, but you can easily use these guidelines on other platforms.
On an e-commerce platform like Amazon, your PPC Marketing ads can be displayed in search results, matching a search term typed in by a customer. Or they can be displayed on product pages which are related to your items, including your listings. The main challenge is to identify this list of search terms and product pages which drive conversion to your items.
When setting up your campaigns, you want to achieve two goals:
Why would you care about the less performing ones? Wouldn’t it be better to just negative target these guys? Actually, you want to keep these search terms and product pages active, because you never know if they might drive conversion at some point. For more details about this usual mistake in PPC campaigns (among others), read that post.
And if you want to know what we are going to do with your bad performing campaigns, continue reading this post.
So now, how do we build, update and bid on this list of keywords and product pages which drive conversion to your items? Let say, you are running advertising on a range of 10 products. In order to achieve the goals we talked about, we would recommend you to set four campaigns in the Amazon Advertising Console:
When setting up these campaigns, creating one Ad group per ASIN is a good practice. It provides a good balance between bidding on keywords specific to each product and mutualizing search terms data. In our example, you will have 10 ad groups in each campaigns.
Figure 1 - Here is what your campaigns should look like.
The underlying assumption is that you are able to identify, from your experience and past campaigns performance, the list of search terms and product pages which are good and bad performers. Most of the time, you’ll be surprised by how narrow this list is, especially the list of good performers. It usually boils down to a handful of keywords and product pages per item. You will use this list to bid in EXACT on search terms (or product pages) which you KNOW the performance. (good or bad).
At M19, we use statistical models to predict the conversion, at the ASIN level, of all search terms and product pages which generated at least one click on your items. When we have enough data to accurately predict the conversion, we remove these search term or product page from the auto or phrase campaigns (negative targeting) and move it to the exact search term campaign or the product page campaign. That’s how we KNOW the performance of a search term or a product page. And we do it every day.
So, why do we recommend bidding in exact on bad performers, whether they are search terms or product pages? Because a bad performance, or a low probability of conversion, is as significant a signal as a good performance (or a high probability of conversion). It means low performance conveys as much information as good performance. The only thing which differs is the amount you are willing to bid on bad performers. Easy answer: bid VERY LOW, but bid in exact because you KNOW something about the performance.
One last thing about the auto and phrase campaigns. These campaigns are used to get certainty on the performance of search terms and product pages which you are not able to assess yet. So, when in doubt, don’t be afraid to let a search term sit in these campaigns for a couple of weeks. It is always better to gather more information than making a hasty call on the performance of a search term. The end goal is really to grow the list of negative targeted search terms/product pages in this campaign.
The auto campaign will also capture new or seasonal patterns, and if their performance prove to be significant (good or bad), you’ll be able to bid in exact on them.
Set up four PPC campaigns (auto, phrase, exact and product) for each group of items you want to advertise, with one ad group per item.
For each of your items, build up a list of search terms and product pages whose performance is significant (whether good or bad) and bid in exact on these guys. Of course, bid low (like two cents) on bad performers and high (depending on your profitability goal) on good performers.
Don’t forget to negative target, in the auto and phrase campaigns, all the search terms and product pages which you are bidding on in your exact and product campaigns.
Use the auto or phrase campaigns to assess the performance of search terms and product pages which you DON’T KNOW the performance yet. And use it to capture new search patterns.
Review your assessment of the performance on a regular basis (at least two times a month). Based on your findings, adjust bids and move search terms and product pages from the auto and phrase campaigns to the exact or product campaigns. Don’t forget the negative targeting. You want to keep things mutually exclusive and collectively exhaustive. At M19, this is powered by Artificial Intelligence and done on a daily basis.
Now you know everything about the advanced set up of Amazon PPC campaigns and the principles which guided the structure of the M19 campaigns. Of course, doing this on your own, on a daily basis might be a little cumbersome. If you really want to perform on Amazon without worrying about this complexity, we recommend you run your advertising with the M19 technology. You’ll be surprised how simple it is to use.
We will constantly share insightful articles about Amazon ads with you.
Endlich hat es Amazon seinen Partnern ermöglicht, stündliche Statistiken für Anzeigen auf den globalen Marktplätzen abzurufen — also haben wir getan, was wir tun mussten: Wir haben die Zahlen analysiert und nach einer Möglichkeit gesucht, unser Tool zu verbessern.
Finally, Amazon allowed its partners to retrieve hourly stats for Ads on the global marketplaces, so we did what we had to: we crunched the numbers looking for a way to improve our tool.