amzstrategyconsulting.com

Amazon places both sponsored and organic search results in the same page and both types of results compete for the same search queries from would-be customers, if your product is already highly ranked organically for a given search term, how can you know if your Amazon PPC campaigns are actually bringing in more sales instead of just taking sales that you would have gotten anyway?

Enter incrementality testing, I know of two ways you can test this:

Method 1 :

Separate your advertising campaigns into three types:

Branded, generic and competitor.

Branded keywords will have a high cannibalization rate, generic less so and competitor KWs will be the most incremental. You can further separate the generic KWs into high organic ranking and low organic ranking if you want more granularity.

The point of separating them is so that we can test individually by campaign type, starting from the most incremental to the least incremental campaigns.

The next step is to set-up an experiment for a set period of time (4 weeks might be good):

1 Choose a group of ASINs that will serve as the test group, and another set of similar ASINs as a control group.

2 Do not make any changes to the test or control groups during the experiment period (like running promos off-amazon).

3 Increase the advertising budget on the Competitor campaigns for the test group, don’t make any changes to the control.

4 Test for 4 weeks

5 Compare the change in total sales for the test group vs the control group. Did the test group increase its sales at a higher rate than the control group? If so then advertising was incremental, if not then you’re probably at the optimal level

6 You can run further experiments on the other campaign types (generic and branded) to determine their incrementality.

7 Additionally you should look at incremental contribution margin. It’s possible that we are adding new customers but actually making less profit.

Method 2: 

1 Choose a specific search term to test

2 Go to the Search Query Performance Dashboard, filter by the ASIN you would like to test, and use the data from the column “ASIN Share of purchases” as a benchmark.

3 Increase spending on the specific term, for the specific ASIN

4 Monitor weekly (Search Query Performance shows weekly, monthly and quarterly performance)

5 If ASIN Share of purchases increased then your spending is likely incremental

6 Remember to also check your contribution margin

Method 2.2:

Additionally if you sell multiple products that are substitutes of each other, you want to know if your advertising campaigns are incremental as a whole, that is if you increase spending on a search term for product A from your catalogue, you’re not stealing sales from product B also in your catalogue. 

The way you do this is by using the Brand View in the Search Query Performance dashboard and using the “brand share of purchases” instead of the ASIN share of purchases.

If it increased then you’re not cannibalizing (completely) from your own catalogue, if it remained the same while the ASIN share increased, then you are likely cannibalizing from your own catalogue.