One of the advantages of real-time media buying is to access a potentially unlimited pool of supply. However, at the same time, this is also one of its main challenges as it requires to constantly maximize media buying efficiency by identifying and focusing on the best-performing campaign set-up. Strong of our experience of running or helping our clients run a myriad of programmatic campaigns through our platform DataLift, here are seven tips and levers you can act on when optimizing the performance of mobile RTB campaigns.
1. Inventory Type
The first lever you can act on when optimizing RTB campaigns is selecting or avoiding the type of inventory you bid on. Here are three subsets of optimizing options:
1. Mobile Web vs Apps
Generally, apps tend to perform better than mobile web as they offer a more engaging environment for users. However, it is always worth testing mobile web placements nonetheless as they increases the pool of inventory. Make sure to carefully select the publishers (see below) and the formats to ensure that the ad sizes are adequate.
2. Whitelisting and Blacklisting
When starting a programmatic media buying campaign, it is important to minimize the learning curve by quickly discovering the best-performing inventory placements and focusing your buying efforts on those publishers. You should however also dedicate a small part of your budget to exploring new inventory pockets that might help scale the campaign volumes in the future.
The opposite goes for blacklisting: take out the worst-performing placements from the inventory you’re targeting.
3. PMPs – Private Marketplaces
When a set of placements works extremely well (far above average), it can be worth setting up a private marketplace (PMP) with the exchange in order to access the inventory through a private auction mechanism. This will command a higher floor price, but in exchange you will gain priority access to the inventory.
2. Audience Targeting
The second key lever you have at your disposal to make the most of your programmatic campaigns is audience targeting. Finding the right audience can help you avoid wasting undesired impressions and can be done through the use of first-, second-, and third-party data.
- First-party data: data that you own about your users. See here for a more comprehensive account of what can be done with first-party data.
- Second-party data: data coming from the media buying activities, such as device type, OS, etc.
- Third-party data: data coming from third-party providers, typically data management platforms (DMP). This data can help you enrich the information you already have from the bid requests and help you make better bidding decisions.
3. Dayparting and Day of the Week
Selecting which time of the day and which day of the week to focus your efforts on in order to follow usage patterns can pay off, especially on mobile. You can identify users behaviors in order to serve them the impression at the time they are the most keen to click and download the app.
For example, it could mean identifying at what time your target users are commuting in a given city or country and increasing campaign caps at this time, or focusing efforts on the weekends.
Another advantage is that it can be helpful to avoid click and install fraud.
In the below example, you can see that eCPIs (effective CPIs) tend to drop from Wednesdays onwards with a low point during weekends. Why Wednesdays? Hard to say, but it’s important to make sure to take it into account those trends to optimize the campaign.
4. Geo Targeting & Geo-Fencing
Another way to optimize RTB campaigns is by leveraging mobile capacities such as geo-targeting and geo-fencing; for instance targeting a specific ad creative to a specific geographic area.
For example, for a travel app it could mean using different creatives depending on where you advertise. For example you could use a sunnier creative to target Illinois than you would for Florida.
To take a concrete example: one of our clients, Louder Rewards, has an app which they skinned for the various football teams across the US. In order to reach the right users with the right app, we used specific creatives to geo-fence the home stadium of these teams.
5. Creative Variety and Refresh
Make sure you have at your disposal a range of various creative sizes, types, and designs that can target and reach the right audience. It is also highly recommended to refresh creatives at least once a month, as with user fatigue user performance tends to decline with time.
In the below example, on December 12th, the eCPI decreased significantly (from $18 to $3) following a creative refresh our programmatic team recommended for the advertiser.
6. Frequency Capping
In order to avoid user fatigue, a great lever is to set frequency capping for the campaigns, ie the number of times the same ad can be shown to the same user within a specific period of time. Naturally, the level of frequency capping should depend on how advanced in time the campaign is.
For a brand new ad campaign with freshly-designed creatives and broad targeting, frequency capping can be low as the totality of the vast majority users are seeing the ads for the first time. As the campaign matures, the frequency cap should generally be gradually increased.
For a retargeting campaign, advertisers should also account for user fatigue right from the beginning of the campaign as users already know the advertised app or product.
7. Algorithms and Bidding Strategy
Many DSPs out there (DataLift included) offer a bidding algorithm to optimize campaigns and work towards a specific CTR or CPI goal. However, it is important to understand that you should never exclusively rely on algorithms for the success of your campaigns. Humans are still needed to set up efficient campaign targeting and no bidding algorithm can make up for a poor initial campaign set up!
Humans are also essential in the creative design and optimization process. For instance, they are still often quicker in understanding why a creative performed better than another. Human judgement can often save a lot of time and money and you should never rely on machines without any human interaction.
What are your tips to optimize mobile RTB campaigns? Let us know in the comments!