Results in a Nutshell
I started slowly at $5 a day and canceled ads that clearly weren't working after a $2 expenditure or a hundred exposures without a click. By spending $58.84 over two weeks, I garnered 245 new names, bringing my total to a more respectable 400 names. My cost averaged 24 cents a subscriber, which was my target range. When I first posted a new ad, a good one would snag 10 percent of the viewers at around 15 cents each. Over time, this would drop. I paused ads that started costing more than 50 cents a name.Key Takeaways
- Keep your budget small, $35 over a week. When I tried the "more is better" approach, multiplying to $25 a day, the amount of money expended went up by a factor of three, but the rate of clicks stayed the same.
- People who are too young may grab the free stuff but don't necessarily have the disposable income to buy your e-books. Fourteen-year-olds might read my video-game novel, but they don't pay for it. Don't waste your limited ad budget on them on the first pass. People who are too old may not relate to my books. Therefore, I chose the age range between 28 and 64.
- Make sure to make your first "narrow the audience" filter people with Kindle readers. This limits any audience to people who may take action and become fans. Unfortunately, only about 5 percent of US Facebook customers identify as having e-readers, and they sometimes won't intersect strongly with your intended targets.
- Make sure your targeted audience is between 300K and 1 million. This isn't as easy as it sounds. I added UK readers in because although the US has 5 times the Facebook population, we have only 3.8 times the readers. That means we're less literate. Also, some categories I tested scored many more hits in the UK!
- There was a direct correlation between how well people responded to these ads and how well the tested nine books performed in Amazon sales (multiply by forty). So if you want to test-market a new release, spend five bucks and do it here first.
- The images you pick must be text free, uncluttered, and eye-catching. This is hard to do. Start with a clip from one of your covers, but if that doesn't work, consider going to Dreamstime. Their site has a one-week, five free download trial that I highly recommend using for your experiments.
- People don't speak in books on Facebook; rather, they speak in movies and TV series. By forcing yourself to pick two shows that most exemplify your book, you also communicate better to your Amazon audience.
- Fantasy is a harder sell, reinforcing my earlier experiences, where I can sell 2.5 times the science fiction with the same effort.
Images that Worked
What works on Facebook is doubly important because with similar constraints, what works here should also work in Bookbub ads. These were my most successful ads.
1. Alias meets Armageddon (zoom on Jezebel's Ladder cover)
2. Hot Zone in North Korea. What could another SARS epidemic trigger? (image used on K2 cover)
Strangely, 76 percent of the respondents were women. Therefore, I limited the ad to women only so that I would have a higher hit ratio. I'm running a second campaign switching the starting analogy to "The Stand."
3. Giants, evil spirits, and blood-feuding wizards guard the Doors to Eternity (top of Dreams of the Fallen)
I used "Lord of the Rings" and "Game of Thrones" as the audience base.
4. Like Babylon
I used Han Solo and Firefly fans to start with. This one did much better in England (35-50 percent depending on the time of day). That tells me I should promote my Gigaparsec series more in the UK.
5. Batgirl meets Orphan Black. Spies, quantum computers, & gummy bears. Quantum Zero Sentinel (Dreamstime image)
Since 90 percent of the clickers were men, I limited this to them in order to boost the relevancy.
Be careful targeting images, because some of them only scored among lonely 45-year-old guys late at night--not what I was aiming for.
In the Details
For those who like to see the proof, here's the spreadsheet.
| Title | Sales | Subscriptions | Projected | left to sell |
| jezebel | 4578 | 89 | 3568 | done |
| doors | 3207 | 36 | 1443 | oversold |
| void contract | 610 | 27 | 1083 | 473 |
| k2 virus | 585 | 43 | 1724 | 1139 |
| foundation | 339 | 19 | 762 | 423 |
| quantum zero | 227 | 24 | 962 | 735 |
| scarab | 144 | 4 | 160 | |
| contagion | 77 | 1 | 40 | |
| Messenger | 56 | 2 | 80 |
I have 25 novels in my catalog and limited funds for ads. This data helps me to direct future advertising dollars and effort. Clearly, the lower three don't have a market, no matter how good they are. The top two have hit Amazon bombs and sold more than I expected over the years. The sweet spot is in the middle. K2 Virus is seems like I will get the most bang for the buck, followed by QZS with the Batgirl reference, and Void Contract in the UK. Unfortunately, Foundation doesn't have a good connect rate, and people don't often read the rest of the series. So I'll hold off on that and try it in the UK at a later date, once I learn the ropes with Bookbub ads.
