Presented by AdColony

There’s more than one way to skin a cat, as they say, and there are even more ways to run a global performance campaign. In the early days of mobile user acquisition, clients had one goal in mind: Get a ton of downloads and DAUs, and fast. They typically had a hands-off approach, letting the ad network be flexible in its approach and optimization practices, as long as the immediate outcome was met. Nowadays, however, clients want to know what happens after the install, as they are more concerned about profitability and LTV.

While we, like most mobile ad platforms, welcome more active involvement from clients, especially at the beginning of a campaign, there are times when you just need to trust. Granted, that’s a loaded word these days, as brand advertisers face a paradox of trust with agencies, but standards for transparency and accountability are gradually improving. With that comes trust, and with trust comes performance. Because when you trust your mobile ad partner, and don’t interrupt campaigns mid-flight or make reactionary budget adjustments, it frees them up to do their job — which is to meet your outcomes.

Here is the DNA of a healthy mobile performance campaign, and how you can help:

A clear view of goals, and a strong opinion on ONE goal

What, ultimately, is going to drive your ROAS? Is it retention? If so, how are you measuring? D1, D7, D14 or into Month 2? If this is the case, spending more money on apps where we naturally see more retention will be part of the strategy. Or is there another custom metric or event you are looking at? Focus on which single metric you want to optimize toward the most. (We know it’s hard, but you must prioritize.)

Avoid this pitfall: Killing scale with overly aggressive ROAS goals. Say you want 20% or even 30% — and it’s achieved, but among hundreds of apps, only a handful produce that result. Your instinct says to focus on those few, right? But then you sacrifice scale. Instead, go ahead and focus on those, but also expand and explore new inventory, new creative, etc. to keep scaling your campaign. 

PIE (Post Install Event) data and other ways to identify your ideal user

With that, your partner can create lookalike models to build the right target audience for you; i.e. users who have an affinity to the type of app you want to promote (e.g., hypercasual, card game, racing game) or to find the right users that have propensity to spend in the app. But it’s often not so simple; social casino apps don’t want to run ads for other social casino apps because that’s their competition. Most publishers want to avoid user churn and hence we’ve seen more and more publishers blacklisting competitive apps. When this happens, your network will find another way to get those users.

Avoid this pitfall: Share your PIE data, and insights on audiences that have worked well for you, but don’t get too stuck on past performance. Give your network the freedom to explore new user identities.

A mix of creative types, optimized for context

Creative has become the single most decisive factor in mobile performance campaigns. It is important to A/B test creative message iterations. But it is not sufficient to just throw multiple assets at the network, as many as 50 different creatives, and say, ”just run these all at once.” That approach simply doesn’t work anymore — there’s not enough supply and not enough learning.

What we now know is there’s a time and a place for it all, but you must be strategic, even in your initial approach. For example, in hypercasual games, we see playables and pause-and-play videos perform far better than any other type. But every network has its own supply composition, and what works on Facebook might not work on our video network or vice versa.

To be smart right out of the gate, work with a partner that can A/B test from the start, and automatically substitute in new creative to replace the sub-performing one, so you’re constantly improving as the campaign runs.

Avoid this pitfall: Too many creative optimizations, or too little. Follow your agency or mobile partner’s best practices; they have experience in what is the right amount. 

A note on IPM/install rate

IPM refers to Installs per 1,000 impressions. It’s important for performance marketers to closely watch this metric. This metric is absolutely critical to the marketability of the game or app that you are promoting. Low IPMs or install rates will make it difficult for your app/game to be competitive in good quality supply. Your game/app is competing with thousands of other games at any point in time. Machine Learning algorithms will pick the best ad that pays the highest CPMs. These CPMs are directly related to IR or IPM of your campaign. If your campaign creative can’t provide competitive CPMs, it makes it extremely difficult to drive user acquisition for your campaigns. Pay close attention to this metric and work with your network partner to improve the IPMs.

The right bid prices

A large part of the in-app gaming supply is behind mediation or traditional waterfalls. These waterfalls allow you to reach users with different session view positions. With dynamic bid pricing, you can reach users early on, and so much depends on where you are — first position in the waterfall or twentieth position. Be exploratory in your line items to find the right users, and once you find them, take advantage of optimizations.

Avoid this pitfall: Overly aggressive optimizations and loosely defined metrics. Base yours on clearly defined benchmarks: install rates, IPM, retention rates, or even ROAS. 

Once you feel confident that your campaign has these four core components, there’s no reason for you to be up at night thinking about your performance ad spend. Go into this holiday season with a feeling of trust that your campaigns will be running smoothly and at their best, while you spend that time instead with your friends and family. And, doesn’t it feel good to make money while you sleep?

Vikas Gulati is Head of Global Performance at AdColony.

Sponsored articles are content produced by a company that is either paying for the post or has a business relationship with VentureBeat, and they’re always clearly marked. Content produced by our editorial team is never influenced by advertisers or sponsors in any way. For more information, contact