BookBot recommends using three standard rules to remove keywords which are killing your ad spend.
Non-buyers are killing your ad spend. They click but never intend to buy your book. How do you stop them? Ensure they never see your book! How? Negating the search terms they are using!
What is a search term and why is it so important?
You want to match your book with the potential customer's intent. Someone searching for a vegan cookbook is very unlikely to buy the "Hamburger Bible". Potential buyers show their intent through
- Search - what they type into the Amazon search box
- Products they visit - Amazon assumes if someone visits a vegan cookbook product listing, they likely want to purchase that type of book.
Amazon gives you access to the exact search terms potential readers using (ASINs they visited and search phrases they type) in the search term report.
BookBot reviews this report and tells Amazon you do not want to show for search terms which are costing you money, but bring in no sales. Specifically BookBot uses three criteria for 'negating' (telling Amazon not to show your books) for search terms.
How BookBots finds and turns off search terms?
BookBot uses three default filters. They are designed for you as a publisher (not someone selling beds and a profit margin of $100). Let's look at each one in turn, what it does and why it does it.
"One lucky order" #1 - One sale, high acos and 10+ clicks.
Not recommended for fiction authors with big read thru or large KENP revenue.
The search term made one sale, but it is likely a one off and is now draining your ad spend. Are the search terms relevant? Would you expect them to be there? If not, negate them.
This is most relevant to non-fiction publishers, where Acos is easy to work out (ad spend / sales). For fiction authors with read-thru or high KENP $ levels, we recommend not using this rule as your target acos is likely much higher than 100%.
Know your target acos? Easily create your own version of the filter.
Browser - 10+ clicks and no sales / KENP
The biggest ad spend killer out there for all publishers. People are browsing but not buying, only costing you money! Likely because it is an irrelevant search terms. With 10 clicks, it has been given enough time to prove itself… now time to negate.
No ifs and no buts, time to negate them asap.
Irrelevant search term - 1000+ impressions, low CTR and no sales / KENP
Amazon keeps showing your book to completely irrelevant people (they won't even click on your ad!). Time to tell Amazon to start showing your book to someone else.
Why does this filter matter?
- Limited number of impressions to go around - There are not infinite numbers of impressions. By showing your book for one search term, you sacrifice another search term (which might be a winner).
- Relevance for your book matters to Amazon - If no one is clicking on your ad / book, it sends a signal to Amazon that your book / campaign is not relevant, which in turn impacts your CPC and whether or not your are shown at all.
BookBot removes the search terms which are not getting clicks so Amazon can start showing your books to new search terms. You are training Amazon about what is relevant vs not relevant.
Am I negating too early - the search term might get a sale later
Why the 10 clicks?
BookBot recommends giving a search term time to prove itself, some are losers but some are winners. The average conversion on Amazon Ads is c.10%, whereby you can expect one sale after 10 clicks. If we have not had a sale after 10 clicks, we can be confident (not 100% certain), that the search term will not produce a sale and can be negated.
Why not wait until 100% certain? This would require 30+ clicks which would cost a lot of ad spend across all the search terms. BookBot is striking a balance between being 'confident' and cost.
Why 1,000 impressions and CTR<0.21%?
The average CTR is c.0.21%, we expect search terms to perform above average. 1,000 impressions is enough to be 'confident' that the search term will not receive a sale.