I ran an ad campaign since the beginning of the month on Amazon. Here's what I learned.
- Amazon's ad campaigns take DAYS after approval to actually start serving impressions. Give it time. It took me 4 days.
- My sales rank was in the 20k-30k range, some all time highs for this book. I did not see that volume in sales, which indicates to me that the majority of engagement came through borrows. (People can borrow through their kindles)
- Amazon's reporting is Horrible. To the point that it's nearly impossible to judge the success of a campaign. You get impressions, clicks, spend, and sales but the data is not broken down by day and there may be up to three day lag before a sale ends up in the report.
- They don't do any calculations. There's basic stuff like conversion rate or click through. To calculate conversion % you have to divide the total sales by the royalty for the book and divide that by the number of clicks.
- Borrow activity (KENP) does not show up in the sales data at all. But based on my sales rank, I know for a fact that it's there.
I'm promoting a $4.99 title by interest (Science Fiction & Fantasy: Adventure, Fantasy, Sword & Sorcery) at ten cents per click. Run time 8/1-8/15.
I'm a debut author with no broad following or name recognition.
Why ten cents? Well because I didn't know that it would take four-five days to start seeing activity. More on pricing below.
I had immediate good performance and strong sales rank (for me, anyway) throughout the early part of the campaign. Here I measured click through (How many people clicked on an ad), conversion (how many people bought an ad) and net profit %.
Note: These are the aggregate rankings on each day, not the daily activity. Sample: 40k impressions 470 clicks
Click Through: This number was the most solid and predictable of the three. It started at 1.7% and ended at 1.2%, steadily declining day over day. This tells me that my ads hit a point of saturation quickly and the quality of leads diminished over time as people became inured to the ads.
Conversion Rate: I averaged about 3%. This also went down with the click through. Again I suspect the window for quality leads closes pretty quickly, especially if impressions are served multiple times. (i.e. if I didn't buy it the first time I'm not going to buy it). On my best day I converted at 5%, other days I had zero sales. Again the data was delayed so it's impossible to say which number is accurate save for the aggregate.
Net Profit: Based on the hard number of books sold I spent 6 dollars on my campaign. However... there were definitely borrows. With the KENP payouts I will never, ever know how many pages read over the coming months-years how many of those were from this campaign. I do feel confident the campaign has been slightly profitable when factoring those in.
Cost Per Click
Toward the end I lowered my bid to five cents and still got impressions. At that point however I wasn't getting sales. If I started at five I would have seen 50% profits. Since I was curious I put together a CPC matrix based on book price.
If you expect to sell 5 books for every hundred clicks at 2.99, then a ten cent CPC is your break even point. (I would add 5% conversion seems pretty optimistic for an ad) .If you want to make 50% ROI then you need to cut the numbers in the table below in half.
I've highlighted where my title sits in the chart below.
Amazon PPC ads can be a tool for modest success with the right inputs. Bidding as low as possible seems like a safe strategy.
They set a minimum budget of $100. It's difficult to see spending all of this (at least in the sci fi/ fantasy interest bucket). So this was not a scaleable solution for me. It would take months to spend $1000 and even at 50% profit... it's not a money machine.
I would use it again to goose my sales rank or lead people to a particular title. It paid for itself, mostly and got me exposure.