Bids by Badger Algorithm Not Hitting Target ACOS

Why isn’t the Bids by Badger Algorithm hitting its target?

Mathematically, the perfect bid is a calculated metric and there is no emotion to it. Your bid (CPC) should be set to your ACOS target percentage of your average RPC (revenue generated per click). For example, if your average RPC is $1 and you’re targeting a 30% ACOS, then your average CPC should be $0.30.

Let’s look at a few pieces of the Bids By Badger Algorithm to understand why we might be more than 10 points away from target ACOS.

The most common scenario: Your short-term data is much different than your long-term data. 

In our analysis of all clicks running through the Ad Badger platform, we found that the average CPC does not change by a great deal month to month or week to week. 

Let’s assume you have a 30% Target ACOS, but you are currently at 60%. There are really only two ways to get your ACOS down from 60% to 30%. 

  1. Double your conversion rate. (Doubling the number of sales from the same ad cost.)
  2. Cut your CPC in half. (Reducing the ad cost by half, while maintaining the same revenue generated per click.)

PPC can only influence the CPC in this equation, which is why Amazon Sellers should always try to increase the conversion rate on their product pages.

Let’s say we look at 90 days of data and see that you have a 10% conversion rate. However, in the most recent 30 days, your conversion rate is just 5%. In general, this is a massive, atypical, abnormal drop in conversion rate. In our research, products do not change this dramatically. Most likely there is competition selling a similar product with more reviews and maybe even a cheaper price point.  This could also be caused when your product goes out of stock or if you get a bad review.

Bids by Badger will update daily with your new 40-day conversion rate and continue to scale down your bids. Within 1-3 weeks, this will usually correct itself as your 40-day conversion moves closer to the newer conversion rate. 

Scenario 2: Low conversion data confidence. Mathematically, it’s correct, but there will be swings.

In this scenario, we likely have a low level of data confidence. Let’s say you have just 10 conversions of data and 20 clicks. That’s a massive 50% conversion rate. 

Mathematically, your bid will be set to reflect this huge conversion rate. But we need to wait for more data confidence to get this number closer to what your eventual, leveled-off conversion rate will be. Note: You can set a custom bid cap so that the algorithm will never exceed whatever limit you set. 

The biggest issue with low-conversion data is the swings. One day you may have a 50% conversion rate, and another it might be 25%, and another it might be 75%. More data confidence will keep this stable. As we continue to learn more about bidding on Amazon, we’ll update our algorithm to be more confident with less data.

Scenario 3: No Conversion Data. We’re buying data right now. 

We wanted to build Bids By Badger into a system you can turn on from Day 1. Want to drop 100 keywords and turn on the bid algorithm? Do it. We’ve got your back. 

We have special systems to balance data discovery (can we get more clicks?) and not spending too much (we want more clicks but don’t want to break the bank).

Once it gets a conversion, we’ll start calculating bids based on the revenue generated per click. Until then, we’ll inch it up and inch it down based on how many clicks we have. 

We hope this article helped you understand why the Bids by Badger Algorithm might not always hit your target ACOS.

If you have any more questions for us, chat with us using the bubble in the bottom right-hand corner of your screen or email us at thebadger@adbadger.com.

Thank you for choosing Ad Badger!


How did we do?


Powered by HelpDocs (opens in a new tab)