La Riposte

Tuesday, April 10, 2012

KDP Select “FREE for FRIDAY” Promotion Recap

Continuing the analysis of the KDP Select program and particularly the use of “free” days to promote short stories, I recently concluded the third of my planned “FREE for FRIDAY” promotions.

The genesis, some of you might recall, was that when I released Happily Ever After as a KDP Select title, I used all 5 free days at once, and while the book did well, with 190 downloads on the first day (and 656 total) the number of books downloaded for free dropped off considerably after the first 2 days of the promotion.

Thus, with Seven Lives to Repay Our Country, I decided to test the effect of 1 and 2-day promotions, and also to find out just what effect social networking played on the efficacy of such giveaways.

Finally, I wanted to establish a baseline for the “KDP Select Bump”, if such a phenomenon actually existed.

So, what did I learn from this promotion?

Well, a few interesting things.

Results: Seven Lives to Repay Our Country rose to #8 on the Fiction/Genre Fiction/War list, and #35 on the Fiction/Short Story list, with 102 copies given away.

Observations:
  • Social networking saturation. The first “FREE for FRIDAY” event got 45 signups on FB and 11 on Goodreads. This one got 5 and 6, respectively. This shows that a potential disadvantage to splitting up your promotions may be that you lose the attention of FB / GR fans on subsequent promotions. For example, suppose you have 100 fans on each site. The first time you do a promotion, you invite them all, and 60 sign up. They download your book, but on a subsequent campaign, they probably won’t sign up since they already have it.
  • People don’t Share. I posted a couple of notes on FB that specifically asked people to click “SHARE” to help the note go viral. Total shares? Four. My observation is that unless something is really cute, sad, funny, or bizarre (and usually image-based) people simply don’t share. They may “LIKE”, but that doesn’t have the same impact.
  • Effectiveness of Twitter. Since I didn’t really use Twitter intensively in the first promotion, it’s probable that my Tweeps didn’t suffer from the same sort of saturation, and Twitter is also a platform that lends itself better to sharing a link and a short blurb with a lot of people. This time, I sent 22 total tweets (@ an average of 2 people per tweet) and got 21 Re-Tweets as a result (about a 50% return.)
  • Guest Blog / Author Interview. I had a an author interview at Shut Up & Read and a guest post at Believing in Horses, both of which referenced the promotion. Judging from the scarcity of the comments at both sites, I believe these probably did not measurably impact the current campaign, but since they’ll be out there for a while, they may lead to paid sales in the future.
  • Staying Power. When the promotion ended at midnight on April 6th, this story was ranked 12th in Fiction/Genre Fiction/War list, and #35 on the Fiction/Short Story list... But the interesting thing is that it stayed on the free list for the next five hours, actually rising to #10 by 3 am, and #8 by 5 am, before finally disappearing from the list at 6 am.
  • The “Kindle Select Bump” - it's too early in the month to tell, but past data indicates there should be an increase of about 4 times sales over months without a "free" promotion.
Analysis:

There was little apparent impact from the Facebook / Goodreads events and the guest post / author interview. And based on the second “FREE for FRIDAY” (where I intentionally did no promotion) we know that no promotion = low free downloads. Thus, the only factor that seems to have driven downloads was the Twitter campaign.

Those 22 Tweets to 44 followers generated Re-Tweets that reached approximately 97,292 people. If we subtract out the number of people who probably downloaded just based on the story being on Amazon’s “free” list (using the number from the un-promoted free day as our baseline) we see that about 80 downloads were probably due to Tweets that sparked a reaction.

Thus, we hypothesize that in the future, we can expect 4 downloads per Tweet sent out, and of course, we’ll test this with our final promotion.

Finally, we may have a partial explanation for the “Kindle Select Bump” – if the book stays on the “FREE” list (even though the price shows 99 cents) for 5 hours Pacific Standard time, that means anyone picking books off that list as late as 8:59 am on the East Coast, or 1:59 pm in Europe will see the book at or near the top of the list, but if they click to get it, they’ll see that it costs something – a price they may be willing to pay based on its ranking.

Future Plans:

Based on this information, I will plan to use the last 2 “free” days for this title to run a promotion that ties together everything learned from this one and the 3 promotions which have preceded it, and tests the hypothesis about the “return on investment” for Tweets, the optimal length of a promotion, use of a “Core Group” of supporters, and the extent of the “KDP Select Bump”, which so far appears to increase sales of all books by 4 times the number sold in a month without a promotion.

Hopefully this synopsis will be valuable food for thought for any authors currently planning their own future promotions, so if you know anyone who might benefit from this information, please forward them the link!

Best to all,
Edward

2 comments:

  1. Edward, Thanks so much for posting this -- I'm planning on enrolling in KDP Select in a couple of weeks, and have been studying the trends as much as possible to figure out what the best strategy is for my novel. This post provides some great, firsthand analysis and insight into the process!

    ReplyDelete
  2. Jen, I'm glad you found it useful... Here are links to two other posts that discussed the earlier KDP Select promotions I did in more detail, in case you're interested.

    http://read-write-listen.blogspot.com/2012/02/wednesday-blog-hop-kdp-select-results.html

    http://read-write-listen.blogspot.com/2012/03/row-80-update-and-kdp-select-release.html

    Have a great day!
    Edward

    ReplyDelete