If you want to be competitive in PubFight, getting a sense of the sales potential of a new title is key. One way to do this is to look at the sales of comparable titles. SalesData can be a useful tool for this. It can both help you find some comparable titles and then use them to predict sales.
Interested? No? Too late! You've already started reading...
Step 1: Finding Comparative Titles
To do any kind of predictive analysis for a new title, you'll need some comparable titles. One source is publisher-provided comp titles in CataList. If you have access to them, great! Save those ISBNs and use them in step 2.
But perhaps there are no comp titles listed or you just wish for more. SalesData makes it easy to put together a list of comparable titles from the author's previous works.
You can use the Bestseller Report to do this. Click on Bestseller in the menu and then on the criteria page, do the following:
Set the market to All Markets.
Set the Reporting Period to Lifetime.
In the Author/Contributor field, enter the name of the author (or authors if you want comps for multiple titles).
Optional: In the Format field, select a format that matches the new title. This should help reduce the amount of duplicate editions you need to sort through later.
Optional: Under Pub Date, select a start date that's 2-4 years in the past and leave the end date blank. Recent titles are more relevant for our purposes, so this can help cut down on the noise for those prolific authors.
Then download the report and manually filter it down by removing titles:
that are not first editions;
that are boxed sets, etc.;
that are written by a different author with the same name;
that have subjects that are dramatically different from the new title; and,
that don't match up nicely with the new title.
Now that you have a nice set of comparable titles, you can dig into them in more detail using the Multiple ISBN Report (see step 2).
Step 2: Predicting Sales
By using the Multiple ISBN Report, you can see how your comp titles performed during their initial 13 weeks of sales and use those numbers to give yourself an idea of how well your new title will perform.
On the Multiple ISBN Report criteria page, do the following:
Select the "Use list of ISBNs" tab.
Paste the list of comp ISBNs you gathered in step 1 into the ISBNs field.
Select "Over Time" for how you want to compare the ISBNs.
Select "All Markets" for your market.
Select "Initial 13 weeks from publication dates" for the Reporting Period. (Another valid option is "Initial 13 weeks from first sale dates," but I prefer to use "from pub date" because it aligns the results on pub date and I feel it makes comparisons between the titles easier when trying to predict how the sales will shape up week to week.)
Select "Units Sold" for viewing the results.
Run and download the report.
Step 3: Analysis
The sheer number of variables at play makes concrete predictions impossible, but with careful analysis you can get a sense of a new title's sales potential. For PubFight purposes, the "Weeks 1-13" and "Weeks 1-52" columns will give you a good estimate of each comp title's potential sales during the PubFight timeframe. The weekly columns are a good indicator of what to expect week to week and can be useful when planning your print runs.
Here are some general things to consider:
Pub date matters.
Look for comp titles with a pub date in the same season as the new title. (Comp titles with pub dates outside of the same season should still be considered, but their sales will not be as indicative of potential sales as those titles that have pub dates within the same season. Comp titles with pub dates very late in the year will probably not be useful as sales predictors either. Sales tend to drop after Christmas.)
The more recent the pub date, the stronger weight it should be given. Generally.
Subject matters.
Titles with subjects that don't match the new title or that are outside of the author's wheelhouse should be given lesser weight.
Book format matters, too!
A collection of short stories from an author known for novels or, similarly, a graphic novel from a fiction writer might not be a good comp title to use.
If you have multiple comp titles, try averaging each week and graphing the data.
Finally
Remember what legendary data analyst Ted "Theodore" Logan is fond of saying: "Past success does not guarantee future performance!" The data can only tell you so much—don't forget about awards, movies, and general buzz when making your selections.
