Granular Targeting Explained

Granular targeting is central to succeeding with digital advertising, and has been for a long time.

I’m pretty experienced with digital advertising at this point but when I got my interview for Google AdWords back in 2003 I had to… Google it to figure out what the hell they did. And I still got it wrong!

I was quite lucky that I first interviewed with Overture (the former name of the digital advertising division of Google’s one-time rival, Yahoo) because I had absolutely no idea how these kind of ad platforms worked, let alone best practices. But I learned fast—Google’s training was excellent.

One thing that was drilled into us constantly was the importance of granular targeting. It might be a little more obvious in 2018, but back then the only really familiar advertising model was the broadcast one. Put your message in giant letters on the billboard. Flood the airwaves with ad spots. Take out full page ads in newspapers and magazines (for the kids: these are paper versions of websites that used to be popular). The basic strategy was a simple, brute force one; if the signal is strong enough, the right people will hear it.

Of course, there is a lot of waste in that model. The ever-present adage from that old world was one from John Wannamaker which was true for almost 100 years. “Half the money I spend on advertising is wasted; the trouble is I don’t know which half.”

In today’s world of Google and Facebook and targeted advertising, we can get a very clear picture of the effectiveness of our ads—click rates, conversions, engagement; we’re drowning in data.

Best practices are pretty well established in 2018 too, and it’s interesting to see how core principles haven’t changed that much in the last fifteen years. One of the most central being the importance of granular targeting.

If you have read Strangers to Superfans, you will understand the significance of developing a picture of your Ideal Reader. For those who haven’t—*glares*—it’s the basis of the new marketing paradigm I introduce in that book which takes Stephen King’s craft-based concept and moves it into the world of marketing, and plots out the process through which your Ideal Reader transforms from being unaware of your work to being its most passionate advocate.

The concept of your Ideal Reader is such a useful one in so many ways—that POV shift to the reader will transform how you craft your product and your packaging as well as your promotions—but to employ it successfully you first need to vanquish a powerful enemy: your Inner Artist.

My beret is firmly on my head when working on a book, but it most definitely gets cast aside when the time comes to sell it, because your Inner Artist will lead you astray. Your natural artistic instinct is to share your work with the world, to broadcast in all directions that you have written A Book and that everyone should read it.

I’ve explained before how dangerous it can be to do something as natural-seeming as telling your friends and family about your new book. I’ve also written about how attempting to ride the coat-tails of a popular author can backfire tremendously — ethical considerations aside. But the issue here is much more fundamental than possibly scrambled Also Boughts. The very sustainability of your promotional efforts is increasingly dependent on developing a sense of your Ideal Reader, and then marketing to them exclusively.

And I mean marketing in the broadest sense because this should inform the stuff baked into the product too: categories, keywords, covers, end matter, email autoresponders, even the tropes. But the most obvious place where you need to get really granular with that targeting is with advertising, and this is the place where we probably screw it up most often, even with the best of intentions.

A clear and simple example might be my self-publishing books. Writers come in all shapes and sizes, so, in theory a book like Let’s Get Digital or Amazon Decoded might have broad appeal, among writers who self-publish or who are curious about it, at least.

However, all the data I have tells me that most of the readership for those books are women between the age of 35 and 54. And if I run an ad on Facebook and only show it to women aged 35-54 it will perform much better than an ad targeting all those interested in self-publishing.

You might think that you are missing a lot of your potential audience like… all men, for example. But that’s the wrong way to think about it. That kind of conclusion is being influenced by the old, outdated broadcast model. You need to start thinking instead in terms of most likely purchasers.

Take Amazon for example. It doesn’t try and show you all the products it has. Instead it rearranges the store, to an extent, for each individual customer, to show them more of the things they are likely to purchase and less of the products they have less interest in. It doesn’t mean that I never have any interest in buying a banana slicer or a Nicholas Cage throw pillow, but my purchasing history clearly shows I’m much more likely to buy ebooks and cheap Chinese phone chargers that break every five minutes.

In the same way, you don’t want to show your ad to all potential customers because that would cost an incredible amount of money. To return to the example of my self-publishing books, Facebook has 2.1 billion people to whom it can serve ads, 199m of which are interesting in Writing.

That sounds like a lot of people, but maybe it’s encompassing people interesting in other kinds of writing, like copywriting and technical writing, and maybe even just consuming it. If I dig a little deeper, 7.2m of those are specifically interested in Creative Writing, which is still a crazily broad audience. Drilling down to Self-Publishing specifically only winnows the field to 1.2m. If I go further and restrict it to those interested in Self-Publishing who live in America and also own a Kindle of some kind, we are still well over a quarter of a million people — and it’s very expensive to hit an audience even of that reduced size.

So I need to drill down further. And if I narrow that last-mentioned audience by women aged 35-54, which a variety of data points have historically and consistently shown me is the demographic slice which responds best to my books, then we get a much more manageable 61,000 people.

Your Inner Artist will be wailing at the 200,000 plus that were lost in that last step alone, thinking how many Precious Readers will be cast aside with this aggressive culling, but being this granular with your targeting — I mean the general approach rather than these specific steps per se — is the only way that you can make digital advertising work for you. (And, in fact, I might well narrow the audience in that example even further.)

You are always far better off drilling right down to the subset of readers who are most likely to purchase. There might be only 10,000 such people when you get granular, but limiting your audience like this is infinitely better than advertising broadly.

Our natural inclination as artists is to be read by as many people as possible, but we must fight against that instinct when it comes to online advertising. We need to target the right people exclusively, otherwise it gets expensive fast and you will run out of money very quickly.

And there will be little coming back to you via book sales, either.

If you want to read more about the concept of the Ideal Reader and how it can improve your marketing, check out Strangers to Superfans.

David Gaughran

David Gaughran

Born in Ireland, he now lives in a little fishing village in Portugal, although this hasn’t increased the time spent outside. He writes novels under another name, has helped thousands of authors build a readership with his books, blogs, workshops, and courses, and has created marketing campaigns for some of the biggest self-publishers on the planet. Friend to all dogs.


17 Replies to “Granular Targeting Explained”

  1. Apologies, David, but I’m going to throw a wet towel into this discussion. The targetted advertising you talk about, not to mention your ‘Ideal Reader’ are all bought at the cost of all-pervasive, internet-wide surveillance.
    The comforting fiction put out by all the tech companies and ad networks is that the data gathered from tracking individuals from website to website is not ‘identifiable’.
    In so far as the data does not specifically ‘name’ each individual, this is true, but what they don’t mention is that it is ludicrously easy to put the pieces of the jigsaw puzzle together to identify a ‘target’. The reason is that all tech companies share the covert data they glean via cookies and web beacons/pixel tags etc. This includes info. people post on Facebook, pictures and geo location data. Put all that together and you know enough about a person to look them up on some public database. Car registration? Medicare? Electoral roll?
    Of course this melding of data is not done by individuals, that would be like searching for a needle in a haystack. Instead, it’s done by algorithms that are extraordinarily good at piecing the jigsaw puzzle together. I’m just amazed that this data has only been used for advertising. Or perhaps it’s had other, more malicious uses, but we simply haven’t heard about it.
    Tin-hat conspiracy theories?
    Sadly no. Those who move in tech circles have known about this for a very long time. They’ve tried to sound a warning but users – you and I – have been too enamoured of the benefits to heed the warnings. There will be a backlash, eventually, it’s just a matter of time. Then this paradigm will crash and burn.

    1. Probably not the place for this kind of discussion but… questions could be asked about the very devices that readers enjoy our books on and the conditions they are manufactured under, and also the surveillance baked into the very internet that we are having this discussion on right now. Tech companies do share a lot of information (with government agencies around the world and I think most people would be quite shocked about the extent of that if they spent any time thinking of it.

      But none of this is new either, and phone conversations and radio traffic have been routintely intercepted by governmental agencies around the world since long before the internet or Facebook or granular targeting was a thing. Here’s the BBC on the Echelon program, for example: http://news.bbc.co.uk/2/hi/503224.stm

      Circling back to the topic at hand, though, the kind of data that is being used here is pretty non-invasive and most certainly aggregated and anonymized. Facebook Interest Targeting is primarily based on pages which you have Liked – arguably one of the least problematic aspects of the entire Facebook operation. To give another example, AMS ads are predominantly served based on the content the user has inputted to the search bar or the content of the page the user is visiting, rather than any personal data. Pretty lightweight stuff overall in privacy terms.

      I can tell you from the advertiser side that I’ve seen some changes since the recent scandals, and Facebook are quietly changing certain metrics and targeting options and closing off access to some data for app developers also. Not saying that will be enough to address concerns (I personally think some of the stuff Facebook was doing was really bad and that concern for users privacy rights is fundamentally absent from the corporate DNA), just noting it.

      Whatever my concerns and your concerns, it should be noted that digital advertising is a $200bn+ market, and that’s not going away because of a few scandals. I don’t know if the changes that these scandals (and new ones to come, no doubt) will usher in will be big or small, but I’d be hugely surprised if the fundamental equation of trading some kind of data for services monetized by ads will change.

      1. Thought provoking article, David. I knew some of it – Australia vs Indonesia springs to mind – but I still find it hard to get my head around how close we’ve come to a real 1984 type situation. And how little everyone cares.
        And you’re right, separating out the commercial from the government agencies is probably pointless, but…I can’t do anything about the governments misusing their power and our data. That’s much too big and hard, but I can do something about the tech companies. They’re cowboys in a world where the true black hats are now almost indistinguishable from the denizens of Silicon Valley.
        I’ve deleted my Facebook and Google accounts, unsubscribed from newsletters, especially the ones using MailChimp, and I’m now using DuckDuckGo and Disconnect as a buffer between my online presence and the leeches out there.
        Given that I continue to use WordPress and Twitter, I acknowledge that my protest is probably futile, but at least I’ve decreased the size of my online footprint, which should make it harder to identify me from those so-called unidentifiable bits of data. I also acknowledge that I’m probably cutting my nose off to spite my face, but there you go, that’s me. 🙂

      2. There used to be a political comedian on British TV several years ago who found out about one of the Echelon bases and discovered that they had forgotten to restrict the airspace above it, so he did the obvious thing and hired a hot air balloon in the shape of Saddam Hussein and flew over it, while calling his mother and reading through the list of trigger words that supposedly triggered the recording of your conversation. Army jeeps appeared within a couple of minutes and hilarity ensued…

        I don’t know if you have read “Cryptonimocon” by Neal Stephenson but it’s a fascinating book and, in the version I had at least, had a long-form article by him on how the architecture of the internet was built across the world and how surveillence is essentially hardwired into it. I think it was for Wired or someone like that and you might be able to dig it up somewhere online.

        The data that governmental agencies hoover up is probably dwarved by what private companies are doing in 2018 though. And even if you avoided Google, Facebook, Twitter and ever other social network out there, you still wouldn’t get away from it, unless you stayed away from every single site that hosts ads (which is nearly all of them these days!). You know those new cookie notices that appeared everywhere since GDPR kicked in? I clicked on one of them to see who they were sharing my data with and the list of companies was… insane. Maybe 1000 or more.

        It’s all pretty nuts.

    2. I just want to comment separately about the concept of the Ideal Reader. Because if you have decided for personal reasons not to use Facebook or have a Page or use that (verrrrrrrry) aggregated and anonymized data on Facebook Audience Insights, that’s your call. But don’t ditch the concept of the Ideal Reader – it’s incredibly useful and you don’t need Facebook data to employ it! You can get a very good sense of your Ideal Reader over time by looking at your Also Boughts (both on your book pages and your author page), or simply by interacting with them. You should still develop a sense of who you are writing for and how they will respond to whatever marketing you decide to do. You don’t need to use digital advertising or aggregated data at all to map out your Ideal Reader’s potential journey from stranger to superfan, or to optimize any of the stages along the way. Digital advertising is definitely a shortcut to getting discovered, and things like that Facebook data will be a shortcut to understanding more about your Ideal Reader, but the whole paradigm doesn’t collapse if you elect not to use those tools (which is your personal choice, of course).

      1. You’re right, of course, my Ideal Readers are out there, I’ll just have to be patient while I wait for word-of-mouth to reach them. Just hope it happens before I hit 90! :D:D

  2. “You are always far better off drilling right down to the subset of readers who are most likely to purchase. There might be only 10,000 such people when you get granular, but limiting your audience like this is infinitely better than advertising broadly.”

    Yep. Have learned that with BookBub CPM ad tests (based on David’s writing on the subject). Once I narrowed my Author Targeting down to one comp author with 7,334 followers, I was hitting over 4% CTR, which is approaching the Goldilocks Zone, it seems. Can’t wait to try this out on Facebook for further refinement.

    1. 4% is very solid for BB. It’s great when you get author targets which work for you. You may not be able to target the same author(s) on Facebook (not all authors are selectable targets), and even if you can target that author, you’ll have to test to see if they are as effective (not all author targets translate well from one platform to another), but at least you have a starting point!

      1. Good point (about author targeting). I’m also running AMS and can add those variables into the FB ad targeting plan. But testing is key, as I’ve learned from you. Thanks.

  3. For some reason, I would have assumed your audience was more men, and skewed slightly younger. But I’m a 37-year-old woman so I guess you know us better than we do! Definitely goes to show how important targeting is!

    1. Those male types do make up a subset of the audience but that demographic band is most prominent. Totally skewed by the fact, no doubt, that they make up a large chunk of those who read the most and write the most, more than whatever inherent appeal I may have to those people!

      Ideally, you will identify several Ideal Readers and craft slightly different messages for each. For example, it’s always said that men prefer plotty blurbs and women lean towards much more character-focused ones, and I would presume ad text on Facebook would be similar, so that’s something you could test, if you had a couple of pockets in your readership on either side of Ye Great Gender Divide.

    1. Aha yes, put my books down right away!

      More seriously though, it’s not about restricting your readership – in fact it will have the opposite effect if you do it right. You seek readers in all sorts of ways and digital advertising/Facebook is just one tool available to you. But to make it work, you have to be granular in your targeting and restrict those ads to the tiny subset which is most responsive to those advertisements. The difference between success and failure with AMS or Facebook or BB ads can be quite small and one little tweak like that can turn a losing ad into a winner. And then those sales will make your book more visible to ALL readers so it’s definitely a net win.

Leave a Reply

Your email address will not be published. Required fields are marked *