This section will help you understand the logic behind the auto-generated rules in the Quick Rule Setup, and it can act as a guide for those of you wanting to begin your own rules from scratch. In order to determine if a keyword or target is effective for your book, we will need to look at our ad engagement indicators:
Impressions
Clicks / CTR
They will help you when setting up rules like "if I reach more than X impressions, but I barely got any clicks and no sales or reads, I want to pause that target because it is getting exposure that could go to other, better targets."
How many impressions to a click are enough to determine if a target will get us conversions (purchases / borrows) ?
Usually 1000 impressions per click is considered enough to determine if there is interest in an ad. That said, since Amazon can take some time in updating sales / reads information and there is no cost per impression, there is rarely harm in waiting a bit more, for say 1200.
If after 1200 impressions there is 1 click or none, then the target will likely not be effective, so we can consider pausing it. This way, we will stop giving attention to targets that are unlikely to get us conversions, and focus on others that may have a higher possibility. (Rule in BooksFlyer: No interest at all)
It is easier to wait a bit longer for impressions, since we don't spend money on them, but what about clicks? Especially for some competitive genres with high bids, each click without a sale can feel frustrating — which takes us to the next question.
How many clicks should we wait to get a conversion (borrow or sale)?
Let's take into account the average conversion rate (number of full reads or orders you get from the clicks in your ads) one can expect when advertising on Amazon. A quick Google search will show a majority of marketing articles pointing at average conversion rates of around 10%. 1 in every 10 clicks is the one that brings the sale/read. This rate can be higher or lower, so the average is a good starting point.
In an ideal world, where Amazon Advertising delivers perfect information, all up to date at the exact moment it happens, and where people purchase the book (or start reading it) immediately after clicking on its ad — rather than several days later — we could generally assume that 10 clicks is enough info to know if a target will convert or not. What a wonderful dream.
As you may have guessed, this is not the case. Ad-related sales and reads can get updated a few days, or even weeks, after they happen (note that Amazon is processing a lot of data in just a matter of seconds, so it is only normal that it can take some time to reflect everything properly) and a person that clicks on an ad of your book today, may choose to buy it a few days later (ever heard about sales attribution?).
What does this mean?Â
Basically, if you choose to pause a target only after 10 clicks, there could still be missing data on your sales or reads. Ever noticed a bunch of sales/reads in a paused keyword that you’re pretty sure weren’t there when you paused the keyword? That’s what I’m talking about, and it hurts.
You could wait for at least 5 more clicks to decrease (rather than pause) the bid by a fixed amount or percentage. We are talking about only beginning to decrease a bid after 15 clicks, and little to no return. (Rules in BooksFlyer: Highly exp targets, low clicks; Expensive targets, low clicks)
For more drastic measures, like pausing the keyword or applying the automated rule, wait for 20 at least. (Rules in BooksFlyer: Clicks, no interest; Enough data to automate)
Tip 1: You can wait for a few more clicks if you have a larger budget, and/or if you get a good value from one sale/read ($9.99 price point for ebooks, for example).Â
Tip 2: You need to get clicks in order to get sales / borrows, which means spending money before knowing if you will make it back. If you have a limited budget and are not yet getting much from your ads, start small with two or three campaigns, ensure you keep your spend limited and slowly build on that.