Capitalizing On App Event Optimization

By Ran Milo

Direct response (DR) advertising is all about understanding your ultimate business objective. While a lot of DR focuses on user acquisition, the underlying goal that advertisers look to achieve often goes beyond getting a user to land on a site, sign-up for a mailing list, or download an app.

The smart advertiser understands this simple fact—unless acquired users turn into converted, monetizing users, your return on ad spend will always fall short.

One problem that advertisers face in the world of paid social advertising is the ability to bid for your underlying objective. Bidding for impressions and clicks has little value to an app developer whose goal is to get a user to purchase an item after installing.

Facebook’s CPA bidding, which only charges you when an acquisition (install) has occurred, takes care of such apprehensions of the risk-averse advertiser. However, even in this case, only paying when a user does install, leaves a lot to chance when it comes to driving revenue down the road.

App Event Optimization, Facebook’s latest expansion of its bidding capabilities, allows you to optimize the oCPM bid type for down-funnel events (achieving a particular level of in-app engagement, hitting a paywall, purchasing, etc.).

So now rather than paying per impression and optimizing for users who are likely to install, the Facebook algorithm optimizes for any in-app event that you’ve integrated with the Facebook SDK. So, how can you make the most of it while driving app revenue?

Below are a few tips on how to get the most out of App Event Optimization.

1 Test different events to optimize for (not just purchase)

Now that Facebook’s App Event Optimization allows advertisers to optimize for metrics beyond just the install, most will simply default to optimizing for purchases, as that is the furthest down-funnel metric.

The problem with this approach is that you leave out many mid-funnel metrics, which would likely give the Facebook algorithm more signals to optimize around, untested. It’s also possible that optimizing for the purchase event will raise your CPI to the point that even though you’re driving and monetizing users, campaigns may never yield a positive return on ad spend.

Possible mid-funnel metrics that would be useful for gaming developers to test could be: app opens greater than ‘x,’ reaching a certain level, or hitting a paywall just to name a few. Test a few events with similar budgets to choose the one which drives the strongest performance.

2 Test at scale to give Facebook more optimization signals

As with any algorithm, Facebook’s oCPM optimization becomes more accurate, as it’s fed more data. App Event Optimization should ultimately perform better as you spend more against it, following this logic.

At Bidalgo, we’ve seen across multiple mobile gaming clients that as AEO scales and acquires more users while optimizing for purchases, Install Day ROAS increases with the number of daily installs.

We believe that early monetization is the best indicator of future monetization (in tandem with retention rates), so if AEO can drive users who are monetizing earlier and more frequently it makes sense from the advertiser’s perspective to invest more heavily in this ad product.

3 Test AEO against oCPM for App Install

While it may be tempting to integrate with the Facebook SDK and start pumping your marketing budget into App Event Optimization right away, there are a few caveats worth considering beforehand.

The fact of the matter is that not all apps monetize uniformly. Some apps monetize users immediately and rely on retaining those paying users, and other apps rely on steady daily active user counts and virality to mitigate droves of non-paying uses.

For this reason, it’s important to understand user LTVs within your app as well as the apps payer’s rate (player to payer ratio). Having a solid grasp on these metrics will guide your media buying in terms of finding the ideal price range that you’re willing and able to pay for a user.

Running oCPM for Install allows an advertiser the opportunity to gauge and app’s Payer’s Rate while keeping a goal CPI in mind. Once this Payer’s Rate is known, it becomes much easier to identify how much more you should be willing to pay for an install when optimizing for payers.

4 Change your conversion spec and window

In many cases when an advertiser is entering a new market or launching a new product, there won’t be enough history built up in Facebook’s system to optimize for down-funnel metrics with strong results.

One way to work around this is by changing your conversion spec; begin running your campaigns using the oCPM for Install bid-type to drive significant traffic before switching the conversion spec to Purchase or some other down-funnel metric.

This ensures that you can drive volume at a CPI that you’re comfortable with while feeding the Facebook algorithm inputs before switching to AEO.

Another method that can be used to make up for lack of data is changing the conversion window on your AEO ad sets. Facebook allows advertisers to choose from either a 1-day conversion window or a 7-day conversion window.

What this means is that you can adjust your bid type to either optimize using data from one day prior or the last seven days. As you start running AEO, it can be helpful to work with a longer conversion window (providing more inputs) before shortening it to 1-day once you’re running at scale and can effectively optimize around the most recent purchasers.

Want more details? See the record of our webinar on the topic.

Published on February 14, 2017
Written by
Ran Milo

VP Marketing @ Bidalgo.

Get Started With Bidalgo Today