How long do you give AMS ads to produce sales?

[JRTomlin]

I have a keyword list that I am using pretty successfully for my medieval historical trilogy that I plugged in to a new ad for another medieval historical series. They are similar in being both related to Scotland and biographical fiction so I felt the same keyword list should work.

I started it running the day after Christmas, am getting a fair number of exposures and about 1 click for every 1000 exposures, but ONE sale. 🤦

With the lag in sales reporting, I am trying to figure out when to pull it. Since it is getting clicks, it is obviously costing me advertising budget. I am baffled at getting clicks and no sales though. I don't usually have that happen. 

Any advice? Should I go ahead and pull it after a week or give it a few more days/weeks to see?

[Lee]

I think it depends how much you are prepared to risk. Personally I don't see it as a risk, I see it as paying for information that I can use and refine.
I used to run a lot of Adwords years ago and do quite well with it. Theoretically AMS should work reasonably similar, but in the limited amount of time I've been monitoring AMS, it's very different from all other types of advertising I have done and much harder to find any kind of consistency. I can run the same AMS ad twice and get very different results.

Unlike Adwords, it seems to take much longer to find patterns with AMS, so refining also takes longer. Because of this I will leave an ad running for at least three or four weeks before I change anything. If you mess with them any earlier then you can't gather much information from it so it just becomes a lucky dip. With that said, if I see a consistent pattern quite early on, then I will do something with it, either pause or change the bid. Sometimes it's obvious that a particular keyword might be a bad choice early on.

I am in non-fiction. I have no idea about fiction, it may behave very differently.

[Arches]

I have a keyword list that I am using pretty successfully for my medieval historical trilogy that I plugged in to a new ad for another medieval historical series. They are similar in being both related to Scotland and biographical fiction so I felt the same keyword list should work.

I started it running the day after Christmas, am getting a fair number of exposures and about 1 click for every 1000 exposures, but ONE sale. 🤦

With the lag in sales reporting, I am trying to figure out when to pull it. Since it is getting clicks, it is obviously costing me advertising budget. I am baffled at getting clicks and no sales though. I don't usually have that happen. 

Any advice? Should I go ahead and pull it after a week or give it a few more days/weeks to see?

What I would suggest is to be very skeptical of the AA dashboard sales reporting column. Three reasons: first, sales do lag notoriously; second, the sales don't show page reads if you're in KU; and third, the numbers AA does report are often crap when they eventually do come in.
So the question is how to know if an ad is working. It's tough to know in detail. The only way I know is to look at your KDP dashboard or use a service like Book Report. Those sales numbers seem to be timely and accurate. If you're only running a single ad at a time, it's fair to attribute new sales over historical levels to the ad.
The worst problem is how to manage specific keywords, I think the best you can do is to use the data provided to make intelligent guesses about which keywords are productive at specific bid amounts.
Yeah, the sales reporting sucks, but I do see real sales come through KDP that seem to be related to particular ads. Best of luck.

[JRTomlin]

Thanks for the suggestions.

With the vagaries of sales especially around the holidays, it is difficult to tell. Historically, my sales have gone up pretty sharply around the 1st of January. I am not in KU. When I was, that really made analyzing results impossible.

I never mind not making money on an AMS ad as long as I do not lose money on it. I have a very high sell through on all my series, so all I ask of it is that it feeds readers into the series.

I am skeptical that that ad is doing that in spite of feeling okay about the keywords. They are providing the clicks. That is not my best selling series, but it has decent sales so I am puzzled. I am going to give the ad another week and see how it looks then.

ETA: The AMS dashboard. 

Yes, it is notorious. I tend to take its word on clicks because otherwise I don't even have a starting point for analyzing results. I know we have no reason to trust the report on clicks, but I think otherwise I might simply go nuts (if I'm not already).