We are at the cusp of a new wave of apps. Established brick and mortar brands are looking at what Starbucks and McDonalds are achieving through their mobile strategies and finally prioritizing their own app development. Traditional entertainment and lifestyle brands are seeing the value of stickers and emoji as a new front in their digital mix. One thing all of these brands bring with them is existing agency relationships. And these agencies have to adjust how they think about campaigns to deliver the performance their clients demand. From our experience working with agencies, here are some areas to start.
A typical digital ad agency office
Mobile = Performance
Mobile app acquisition is performance-driven by nature. The ad tech in the mobile industry was modeled after performance platforms of the desktop era, with significant improvements including the increased focus on measuring engagement through post-install metrics.
It’s not that agencies are not accustomed to KPIs and benchmarks. They are. They establish deep and complex metrics to evaluate media spend. But they look at budgets in mirror-reverse terms compared to a performance marketer. A $50,000 buy for mobile, for example, might be expected to generate 25,000 installs. The campaign might start with an even spend over the flight dates. The agency will optimize to the KPIs as they go.
A performance marketer flips this on its head. In performance, we buy installs one at a time, for $2.00 each. We also know that if we want to buy $50,000 worth of installs, we need to make sure that for every 10 installs we generate, 5 register and 2 make a purchase. We know that our ultimate goal is a $10 eCPA on purchase. The reverse in thinking reflects the change in the risk equation. A traditional media spend risks losing budget; a performance campaign risks money up front, as we are only paid on results. To mitigate that risk, we optimize from the moment we start sending impressions. We optimize up, not down from a daily cap.
Flight Dates and Even Spend
Another area where we have found that agencies and performance marketers differ is in the notions of flight dates and even spend. This is due, in part, because agencies have typically centered on branding while the performance industry was born around leads and sales.
Even spend is a component of branding. It controls the frequency and regularity of the ad placements. It also ensures that there are no dark periods for the campaign due to running out of budget.
From the performance way of thinking, if an advertiser is only paying for engaged installs of their mobile app at an ROI positive rate, why wouldn’t they scale? Who wouldn’t want profitable installs at the largest scale and fastest rate possible? Working with agencies, performance marketers have to adjust their daily spends to smooth out the budget across the period of the flight dates. But agencies need to understand one of the challenges this presents for the performance marketer. When you are buying a branding campaign on a CPM, once you pause the campaign, it is paused. There is no more spend. When a performance marketer hits pause, the attribution window (which can last as long as 30 days) is still ongoing, so conversions will continue to roll in. This makes smoothing budgets a more difficult exercise.
Tags and Fraud
One question that comes up often with agencies is our ability to place tags. These are prevalent with agencies because they protect against the kind of fraud that affects them, largely fraud related to ad viewability. This makes perfect sense for branding. The budget is spent on eyeballs, and if the ad is below the fold, the advertiser is paying for an ad no one saw. But it is important to remember that fraud is always where the dollar is. In this kind of traditional display, the dollar is on the impression. That’s where the fraud lives.
In performance, the dollar is on the install. One, this means we don’t have to worry about ad viewability. If the ad is hidden, no one will install the app, no money will change hands, and no fraud will have taken place. We do have to worry about fraudulent installs, and it would behoove agencies to look at tools performance advertisers have developed over the years to prevent this. For example, Adperio has developed proprietary tools that monitor every conversion in real time. We also have a Best Practices team analyzing patterns for red flags that are better discovered through human analysis, such as mean time to install distribution.
Whatever the approach, knowing where and what to watch for is critical.
Pricing and Discounting
Agencies come with discounting. We can work with that, but it is important to understand this pricing structure has potential consequences for the campaign. In the traditional model, the discounting is applied to the price of the inventory being bought. In performance, discounting affects the CPI of the campaign, which can decrease opportunities to scale or lose out altogether in a competitive marketplace.
Another potential consequence is creating an uneven playing field. If a performance company brings quality and scale to a campaign, they will fall under the higher discount rate. Smaller partners who do not add as much value or volume will have higher effective rates, which goes against our intuition of how an efficient marketplace should operate.
Creative partnerships should explore ways, like dynamic CPIs based on a combination of quality and scale, to keep discounting practices in place while solving these issues to the benefit of the agencies’ clients.