The long and short of Goodreads ads

As part of the early marketing of The Obituarist (still just $2.99, available for all devices, oh god please buy a copy), I bought a block of advertising on Goodreads. Well, the campaign has just wrapped up, so I thought it might be useful to look at the details of it all, pull apart my numbers and talk about whether it’s something other writers should consider.

I hope you all like graphs.

Before we start, though – is there any need for me to explain what Goodreads is? Yes? No? Social media site where people list, rate and occasionally review the books they read? Occasional source of INSANE AMOUNTS OF FUCKWIT DRAMA over said reviews, which cause some writers to lose their shit because they didn’t get five stars? Yeah, we all know what it is, and if you don’t, well, it’s worth a look, especially if you’re into genre fiction or like reviews that are mostly series of animated GIFs and the phrase ‘so many feels’.

Anyhoo, GR offer a self-serve ad service – ‘self-serve’ meaning that you create it and they host it, which is fair enough. It’s not a complex ad; just a photo of the cover, a title, a link and a tweet’s worth of text (140 characters). Once it’s all submitted, you then pick a target audience (based on what they already read) and pick a cost-per-click – how much you pay Goodreads whenever someone spots the ad on the right-hand side of their page and clicks on it. That can be as little as 10 cents, and as high as no-seriously-just-hire-a-fucking-billboard – and the higher you go, the more priority Goodreads give to your ad and the more often they’ll show it to target readers. Oh, and you also set a per-day limit on clicks; hit that budget and the ad gets shelved until the next day.

(That’s all pretty cursory; if you want to get more info, here’s the GR advertising page.)

How it worked for me

Back in the second week of June I decided to give Goodreads ads a try. I read through their advice and tried to come up with an appealing tagline for The Obituarist, one that had a ‘call to action’ (i.e. tells the reader to do something):

A social media undertaker gets dragged into a dangerous mystery in this witty crime novella. Click here and add it to your ebook reader!

(Yes, ‘witty’. Come on, it’s a funny book. At times.)

I attached that to the book’s cover and included a link back to its GR page, which I figured would be more useful than its Amazon or Smashwords pages. (Which, yes, makes that ‘call to action’ kinda bullshit.) For the target audience, I went with genre tags – Crime, Ebooks, Fiction, Humor and Comedy, Mystery, Suspense and Thriller. (Humor was a stretch, I admit it.)

Last and most important, I decided to put $60 into the ad campaign – come on, I’m not made of money – with a 50-cent cost-per-click and a 5-click/$2.50 limit per day. I figured that meant the ad would run for 3-4 weeks before running out of money, since obviously I’d be hitting that cap almost every day.

In practice… not so much.

GR provide some nice analytic tools and graphs so that you can watch people ignore your book on a daily basis; here’s how June shook out.

  

What you can see there (click the graph for a bigger image) is that 50 cents don’t buy you a whole lotta pageviews. For most of the month I was getting about 400-500 views of the ad per day, which in turned prompted zero clicks. It was only when the pageviews spiked to 4000-6000 that I got any clicks on the ad. By the end of June I’d amassed 26 361 views and 15 clicks, taking $7.50 from my $60 budget.

Clearly I needed to change things up. So in early July I added a second ad to the campaign – well, the exact same ad, but this one targeting readers of specific crime authors, mostly those that I liked as well. That didn’t have a huge impact, so towards the end of the month I bumped the cost-per-click to 60 cents. Here are the results:

   

Once again there’s a low level of baseline activity punctuated by big order-of-magnitude spikes; my best guess is that those are periods when a significant number of higher-paid ad campaigns finish, leaving room for little fish like me to swim around for a short time before getting crowded out again. And, once again, the clicks tend to only come when we break four figures in pageviews. The second big pageview spike is when I upped the cost to 60 cents, but I can’t tell if there’s a definite correlation to the change or if it’s due to external reasons.

We can also see that targeting by author, rather than genre, does pretty much dick. It might be because most readers don’t nominate favourite authors, or because there’s too much overlap with the genre targeting, but the author-focused ad doesn’t even get 100 views most days.

Anyway, July had 39 398 views but only a disappointing 9 clicks, for a total cost of $4.70.

Moving on to August:

   

Much better! We’ve got more jagged spikes than a pro-wrestler’s teeth here, closer together and higher than before, as are the corresponding clicks. The baseline activity between spiked has also moved up to about 700-900 views per day. (This is also the point where I realised that I hadn’t adjusted the $2.50/day limit on clicks when I upped the per-click cost, so I kinda shot myself in the foot there for the first few days.) It’s also very clear that the author-focused ad isn’t achieving a damn thing; no-one’s seeing it, no-one’s clicking it. Still, it does no harm by existing.

Stats for the month: 83 800 views, 34 clicks and a spend of $18.80.

By September I felt that the campaign was dragging, so in an attempt to amp it up I changed the text of the ad to this:

Chandler meets Facebook in this crime e-novella as a social media undertaker is dragged into a dangerous mystery. Available in all formats.

 No call to action (or exclamation marks), but it’s a more accurate and (I think) more interesting précis of the book. What kind of effect did it have?

   

Umm… I think maybe there’s a slight improvement in how many clicks I got on the good days, but that’s just total guesswork. Also, despite not changing the price-per-click, the number of pageview spikes fell right back – confirming, I think, that that’s entirely due to external factors and the number of campaigns competing for eyeballs on a given day.

Also, yay – one author-ad click! Hooray for the cult of personality.

Monthly stats are 78 489 views, 25 clicks and $16.40 spent. Why the extra 40 cents? Because at the end of September I saw that the budget remaining was a multiple of 70 for the first time and decided to bump the cost-per-click again. This time I also remembered to up the daily limit as well.

And thus October, where the campaign trundled along before ending about two weeks in.

   

The graphs pretty much speak for themselves at this point. Stats for the month: 44 335 views, 18 clicks, the last $12.60 gone from the budget.

Was it worth it?

For 60 bucks I got 272 383 page views over four months. That sounds pretty damn baller on the face of it. But that’s only the first data point. More importantly, those views translated into 101 clicks on the ad. Well, okay, a hundred clicks doesn’t suck.

But of course, not every click is a sale, or even more than a flicker of interest. It’s a little difficult to work out exactly what those 101 readers did after clicking – I think the data is there, but I can’t find it in GR’s records right now – but I can see that during the course of the campaign, 37 strangers added The Obituarist to their list of books to (maybe) read. If all of them buy a copy, and assuming a rough and largely inaccurate average of $2 royalty per book (it varies depending on who buys it and from where), then I’m looking at $74, or a total profit of $14. And that’s best case.

Not, um… not the most amazing result.

On the other hand, it’s not a god-awful result either; it’s not like I just pissed the sixty bucks up a wall. Sales-per-click is a crude metric and one that can only disappoint. On a social media site, it’s also about visibility and exposure; it’s about finding readers and then getting them to boost and pass on the signal. This is the start of that, not the end, and as a start I think it’s pretty viable. I may do another round of ads later on, or I might look at Facebook ads instead. Or both.

My recommendations

So if all those graphs didn’t send your brain into vapour lock, and you’re thinking of going the GR route for advertising your own books, here are five quick recommendations based on my experience.

1. Target genres, not authors. It’s really clear from this data that the author-ad was completely useless. Well, maybe not completely; it did garner two clicks, but then again I might have got those clicks from the other ad at some point. Certainly, though, you need to make the primary ad in your campaign a genre-focused one, with an author-focused one only as backup.
2. Set your cost-per-click above 50 cents. I definitely got more exposure and clicks when I upped the price to 60 cents, and I suspect I would have had more improvement at the 70 cent mark if the campaign hadn’t ended. I think a dollar per click is probably on the high side, though. 60-80 cents would be my mark.
3. Write a decent ad. This shouldn’t need to be a recommendation, it should be obvious – but I’ve been looking at other people’s GR ads these last few months, and most are completely terrible. Like, ‘incoherent gibberish that doesn’t even tell you the name of the bloody book in question’ terrible. Forget the marketing talk and the ‘call to action’; if you can string together 140 characters that make sense, you’ll have a much better chance of standing out from the trainwrecks.
4. Back it up with other activity. Goodreads isn’t just an advertising platform; like any social media site, it works because of its communities and their energy. If you become part of those communities, readers are more likely to recognise your name and style and be interested in your work. Just be sure to do so in a genuine way, rather than ramraiding forums to spruick your book and then fucking off again. Be open about how much you love books, talk to other readers and make connections; that gives you a base level of visibility that can be raised by the ads. (This is the bit I have to work on.)
5. Have realistic expectations. You will probably not become an overnight sales sensation from GR ads. You will probably not blow out your budget in two weeks. You will probably get fuck-all clicks and you won’t be able to meaningfully control the ebb and flow of pageviews. With luck it’ll pay for itself; without luck it won’t lose that much money. It’s just another arrow in your quiver, another frog in your blender; set the ad, let it go, do something else and don’t worry about it.

—

Right, well, there’s 1900-odd words on putting out a 140-character ad. Never let it be said that I can’t talk endlessly about pretty much anything.

Next week – dialogue! I’m not very good at it and now you can be too!