The mobile performance marketing metrics that matter to the big spenders and what we can learn from them
User Acquisition is intense. Success requires UA specialists who are multi-talented; resilient, curious, analytical, and more besides. And because user acquisition represents the ‘engine for growth’ for app-first businesses, if it misfires, the business can stall or fail. So, performance marketing needs to perform…consistently, especially if you’re a top grossing gaming publisher spending $1m+ per month on app marketing.
Many UA marketers can only dream of budgets that big and many don’t have a need to invest that amount nor the expertise to do it effectively; of course no two apps are the same. Appsumer works with some businesses who do. We work with top-grossing and top ranking apps across gaming, travel and consumer internet. We get to see what’s driving the biggest impact, and where the biggest mobile apps are focussing to give them the confidence to scale and accelerate their user acquisition.
As a result we’ve identified the 3 key metrics the big spenders focus on when it comes to scalable user acquisition:
Cohort based Return-on-ad-spend (ROAS)
User acquisition is ultimately about the cost of acquiring a customer and the lifetime value of revenue returns, which we define as ROAS. The significance of looking at this with a defined a cohort window is to ensure you’re comparing the performance apples for apples as well as having a clear view before making a decision. Successful businesses are continually improving campaigns and the apps themselves. This frequency means the effects of those changes aren’t really noticeable. This interference needs to be stripped out to get a clear picture of what’s worth investing more in. Otherwise you might optimise to a source that looks like it has a high ROAS but find it doesn’t’ correlate to the revenue from that cohorted cost.
Cohorted analysis was described as a game-changer for mobile app businesses over 4 years ago by Adjust in Venturebeat (1), and we’re surprised to find that many e-commerce businesses aren’t looking at things this way.
Cohorted data also allows better forecasting. When you know how long it’ll take for sufficient revenue to generate to pay back your advertising investment, you can phase your future ad spend with greater confidence. This is music to the ears of your friends in Finance and avoids the awkward conversation of explaining why your increased spend for the last few days still isn’t yielding higher revenue.
The big spenders are taking this further and looking at ROAS based on predictive LTV as well as defining a short cohort window of ROAS to optimise against which correlates with this.
Conversion rate from impression
Achieving high ROAS isn’t enough if your campaign isn’t delivering the reach and scale. The conversion rate from impressions vs your bid ultimately dictates your CPM. And when it comes to volume CPM is king. To accelerate reach cost effectively, brands need to recognise this and optimise for it. The conversion rate of course depends on how you buy the media; if it’s a on a cost-per-install basis then your install rate is as important as your click-through rate.
Simply, it’s about making your ads as relevant as possible to the audience you’re showing it to, and most importantly in the context of your competitors. Facebook provides some transparency to this with what they call a ‘relevance score’. High relevance score ads typically can win reach whilst still bidding low.
Increasingly apparent and increasingly important are fraud metrics. App install fraud specialists, Machine, found that almost one third of the 22.4m app installs they analysed between January and May 2018 resulted from attribution theft (2), just one of many types of fraud that top-grossing apps are continuously checking against in their attempts to minimise its impact on their budgets and their business.
High-spending apps have more to lose in money terms from fraud and often have more advanced protection in place, but a proportionately higher percentage of installs for more modest spenders are fraudulent; budget they can sometimes ill-afford to waste.
Being able to identify fraudulent sources and seeing it against dimensions at a granular level helps user acquisition teams exclude more bad traffic sources sooner. It also allows budget to be re-allocated down at the sub-publisher level which can be significantly skewing aggregated results and can lead to poor decision-making.
Something else we can learn from the ‘big spenders’ is their ability to access these key metrics on-demand as well as at the most granular detail and be able to optimise for them with agility. Some have invested significant resource building in-house analytics tools for this purpose. Many though choose ‘buy’ over ‘build’ finding tools with the flexibility that let them unify and visualise these insights in the ways they need to accelerate their operation, and their business.
This article was originally published on MobileMarketer.