In my previous articles on attribution and acquisition (found here and here), there’s been a good amount of shop talk that might seem more appropriate for an advertising conference than a game developer’s site. However, it’s important to understand that the way everyone is bidding on ads -- including game developers -- is broken. There’s a good chance that you’re overbidding on some of your ads, and you don’t even know it; or, on the flipside, are getting a great deal and have no way of showing it in your reports on ad spend ROI.
The good news is that technology is catching up. The big data revolution has made it possible to attribute more accurately. Predictive analytics allow game developers to track LTV in addition to CPI. Social data can be taken into account to target users based on their influence and eliminate the data from bot traffic.
In short, we’re getting better at this. In this final article, I want to take this conversation out of the theoretical and move into some of the real-world applications of better, data-driven attribution models.
The obvious first application is user acquisition. With better advertising models, you can pinpoint exactly which advertising source is bringing in users and causing them to sign up for your game. No more vague attribution. If an advertising source isn’t working, decrease your spend or cut it out. If it is working, put more capital toward that source.
So if it’s so straightforward, why aren’t more people using data-driven models? Well, they’re fairly new, and only recently advanced enough to justify the spend. Publishers also don’t want you to use them, to put it lightly. It’s easier for them to make a profit when you don’t really know how much the ad is worth, so you’re blanketing the web or overspending. But what they don’t realize is that they’re undervaluing ads, too. This could be a win-win situation when adoption rates increase.
The solution spans more than just acquiring initial users, too. Advertising isn’t the solution for everything, of course, and this is by no means a comprehensive how-to guide on player acquisition, retention and monetization. This is simply to show the different ways advertising can be used beyond the initial player acquisition.
Retargeting: You know that potential player that stops by your site but doesn’t actually sign up? The one that got away? Use ads to be more strategic when retargeting those potential players. You can cross-reference analytics data to find your most valuable targets with the advertising sources that take them to your site, then make and appropriate real-time bid on the ads that will reach them.
Cross-Promotion: Similarly, you can cross-promote your game through ads. Does “related products” sound familiar? Again, by uncovering what sources your players are coming in from (and similar games they play, sites they visit, or products they buy), you can cross-promote and bring them onto your site.
Churn: In another scenario, you have a player that's about to churn (according to whatever analytics program you’re using). The bad news is that you might not be able to stop them. The good news is you don’t have to lose them as a customer. You can advertise another game in your portfolio to them, targeting based on the most valuable sources you found.
Upselling and Conversion: For the players who aren't at risk, you can still use outside ads to upsell. Figure out who isn’t at risk for churn, or who are the most influential players. Give them a sweet deal through a targeted ad, and bump them up to the next pay level.
Keeping Creative Honest. The last 5 years has seen an explosion in ad tech, but the unintended consequence has been the slow morphing of advertising from a creative business into an arbitrage-focused optimization process. As a result, the content of the ads has become seemingly less important. However, good data-driven models will show which creatives have the best impacts. Totally aside from the ad publisher, we can now know which ads are working. We strongly suggest that “working” includes the resulting players’ impact on other players as well. But however you derive LTV, you can and should now trace it back to your messaging. This gives you a tool to evaluate your creative team, as well as a metric for them to justify their own efforts.
So what can game developers do to improve their attribution standards now? Stay informed. Advertising isn’t always top-of-mind for developers, but its importance will only grow in the coming years. Do your homework before implementing an analytics program, and reconsider the way you’re measuring your advertising. Consider using data-driven models, and at the very least, stay vigilant on the way you’re currently measuring your ROI. As the saying goes, sunlight is the best disinfectant, and more data only shines more light on the murky world of ad values.