Book Promo: $15K Ads Drive 3.5x ROAS in 2026

You can’t just write a good science book and hope it finds an audience. You have to put it in front of the right readers with a smart digital ad strategy. A well-run campaign can turn a title that seems obscure into a bestseller, and it’s proof that even the most specialized books have a hungry audience if you know how to reach them.

Campaign Teardown: “Cosmic Whispers: The Search for Dark Matter”

Here’s the breakdown of a campaign we just wrapped for “Cosmic Whispers: The Search for Dark Matter,” a non-fiction book written for educated laypeople and science fans. The main job was to drive pre-orders and get some early sales velocity going before the official launch. We had eight weeks to work with, from mid-September to mid-November 2025, and a total budget of $15,000. We set our targets directly: get at least a 2.5x Return on Ad Spend (ROAS) and sell a minimum of 1,000 books.

Strategy and Targeting: Pinpointing the Curious Mind

Our whole strategy was built on slicing our audience as thinly as possible using Meta Ads Manager and Google Ads. We knew that the “dark matter reader” was a diverse bunch, ranging from amateur astronomers and physics undergrads to general science buffs who just like a good documentary. On Meta (Facebook and Instagram), we built a few different audience segments. The first was an interest-based group targeting users who followed topics like “astrophysics,” “quantum mechanics,” “NASA,” and specific science popularizers like Neil deGrasse Tyson. The publisher also gave us an email list of science book reviewers and early adopters, which we used to create a custom audience. From that list, we then built 1% and 2% lookalike audiences to find more people with similar profiles. For Instagram specifically, we focused on really visual creative, like astronomical photos paired with provocative quotes from the book. Google Ads was all about capturing search and display traffic. For our search campaigns, we bid on keywords you’d expect, like “dark matter book,” “cosmology explained,” and “physics popular science,” but we also went after long-tail keywords like “what is dark matter made of” to catch people who were just starting to get curious. On the display network, we didn’t just let Google decide where to run ads. We used managed placements on sites we knew our audience read, like ScienceDaily and Phys.org, making sure our banners showed up right next to relevant articles.

Creative Approach: Beyond the Cover Image

We knew we couldn’t just throw the book cover on every ad and call it a day, since different platforms and people respond to different things. For Meta, we ran tests on three main ad formats:

• Image Ads: These used stunning, high-res photos of galaxies, nebulae, and some cool concept art about dark matter, with short, punchy quotes from the book laid on top.
• Carousel Ads: This format let us walk users through different chapters or big ideas from the book, giving each a quick summary and a call to action.
• Video Ads: We made short (15-30 second) animated videos that broke down a complex concept like dark matter into something easy to grasp, then ended with the book cover and a “Learn More” button.

On Google Search, the ad copy had to be tight and focused on the benefit to the reader, promising to make complicated science understandable. Headlines like “Unravel the Universe’s Deepest Secret” and “Dark Matter Explained: A Journey Through Cosmic Mysteries” did really well. For the display network, we created a mix of static images and HTML5 animated banners that would look good on any device or site. We kept our calls to action (CTAs) consistent across everything: “Pre-order Now,” “Discover More,” or “Get Your Copy.”

What Worked: Data-Driven Insights

This campaign worked because we were all over it, A/B testing constantly and tweaking budgets every day.

Our remarketing campaign on Meta was the runaway winner. We retargeted anyone who hit the book’s sales page but didn’t buy with a simple 10% discount offer. That small group converted at an incredible 18.2%, a huge jump from the 3.5% we saw from cold audiences, and our CPA dropped to just $5.50. That’s your proof that following up with warm leads is where the easy money is. On Google Search, we found that keywords like “dark matter explained for beginners” and “popular science books physics” brought in the highest CTRs (2.8% and 2.5%) and the cheapest conversions. This told us people were actively looking for accessible info, which was a perfect match for how the book was positioned. We immediately funneled more budget to those keywords. The video ads on Instagram were also a huge hit, especially the animated ones that made abstract ideas visual. We had one 20-second video explaining “gravitational lensing” that got a 45% view-through rate and a CTR of 2.1%, blowing our static image ads out of the water. For a dense topic like this, moving pictures clearly do a better job of grabbing and holding attention.

What Didn’t Work: Learning from Iteration

Some things definitely flopped at first. Early on, our broad interest targeting on Meta for things like “science news” or “physics” got us a ton of impressions but almost no engagement or sales. The CPA on those ad sets was hovering over $25, which was totally unsustainable, so we killed them within the first week and moved that money to our more specific interest groups. We saw a similar problem with some of our Google Display placements. Running ads on general news sites, even in their “science” sections, just wasn’t specific enough and led to a lot of wasted clicks. We tightened our managed placements to focus only on hardcore science blogs and educational sites. That one change dropped our display CPA by about 30% in a couple of days. Another mistake was overlooking mobile optimization on the landing page itself. Our ads looked great on phones, but we found out through Google Analytics that some of the interactive graphics on the book’s landing page were loading slowly on older devices, causing people to bounce. We worked with the publisher to get the page optimized, which cut load times by 1.5 seconds. Fixing that small technical issue boosted our mobile conversion rates by 10%.

Optimization Steps Taken: Agility is Key

We were in the accounts constantly. For the first two weeks, we reviewed performance daily, then shifted to three times a week once things stabilized.

We ran a ton of A/B tests. For example, we quickly learned on Meta Ads that benefit-driven headlines like “Unlock the Universe’s Deepest Secret” worked way better than just descriptive ones. We even tested CTA buttons and found that “Get Your Copy” did slightly better than “Shop Now” for this audience, probably because it feels more aligned with an educational product. Budget reallocation was a daily task. We were constantly shifting money away from ad sets with a poor ROAS and pumping it into the winners. By the third week, about 70% of our Meta budget was going to our remarketing and lookalike audiences, with the other 30% feeding our best-performing interest groups. On Google Ads, we were just as aggressive, raising bids on high-converting keywords while pausing the ones that had high CPAs or bad quality scores. We also built out our negative keyword list aggressively on Google Search. Any time we saw money being spent on irrelevant searches like “dark matter game” or “dark matter movie,” we added those terms to our negative list immediately. That simple cleanup probably saved us an estimated $500 over the life of the campaign. Another key optimization was adjusting our ad schedule. After looking at the data, we saw conversions were much higher in the evenings and on weekends, which makes sense, that’s when our audience had downtime to browse and read. So we set our ad delivery to be more aggressive during those peak hours, bumping our bids by 15% to make sure we were visible when people were most likely to buy. Finally, we kept an eye on what competitors were doing. It didn’t dictate our strategy, but it gave us useful context. For example, when we noticed a competitor was focusing heavily on ads that looked like documentary clips, we decided to double down on our book’s unique narrative-driven approach to differentiate our ads.

Reflections and Future Directions

So what did we learn? A few things stand out for anyone selling a niche science book. First, you absolutely have to know your audience. Generic targeting is just lighting money on fire. Second, using social media for discovery and search engines to capture active interest is a powerful combination that works together. Third, you have to be willing to constantly test creative and move budget around. It’s not optional if you want a positive ROAS. Our final 2.96x ROAS and nearly 1,500 sales blew past our original goals. I think we could have pushed it even higher, though. If we had used dynamic creative optimization from day one, the platforms could have found the winning ad combinations even faster. Going forward, we’ll also make short-form video a bigger part of the plan right from the start, since it clearly performed so well here. There’s a huge audience for good science writing, and with the right ad strategy, publishers can connect their authors directly to those readers and sell a lot of books.