Lorenzo Rossi – From his time in Free2Move
There are many reasons why you may need to begin to run marketing and advertising campaigns on multiple channels: you might be assigned a large budget and have to scale rapidly, or you might be advertising a niche app and find that the duopoly (FB and Google) are not efficient for scaling for your specific situation.
Regardless of your reason for deciding to grow a multi-channel strategy, it can be difficult to know where to start. Managing the logistics of running the daily operations of multiple platforms, tracking results, comparing performance, and knowing when to scale can be a challenge. But through my experiences, I’ve seen what managing multiple channels really looks like, and for that reason, I wanted to share what I’ve learned recently developing Free2Move’s multi-channel strategy.
We decided to build a multi-channel approach for Free2Move because the app is a bit niche. It’s currently not available worldwide, only in specific urban environments, and is addressed to people who are using or might want to use shared mobility services. Because of this, we ran FB campaigns with restricted geolocation and additional interest or lookalike targeting, and it made it basically impossible to scale the budgets without growing the cost per acquisition due to low reach and saturation. That’s why we decided to take an “all-you-can-test” approach and introduce new channels.
During the past two years we’ve run campaigns – in addition to FB and Google – on Snapchat, Twitter, LinkedIn, Reddit, Yahoo Gemini, Influencer Marketing, Outbrain, and also Programmatic and Affiliate Networks, Preloaded, plus some OOH campaigns. In an attempt to keep this article from turning into a book, below I’ve shared what I think are the 5 most helpful things to keep in mind when beginning a multi-channel mobile marketing and advertising strategy.
1. How will you organize channel management?
When you run multiple channels, it is essential to have the support of a good BI team (and tool) that is capable of consolidating the costs from all the platforms you are using and connecting them to your MMP’s download and conversion data. Visualizing the results allows you to do specific and cohorted analysis and understand each platform’s performance. It is important to break down the data per country, gender, device, age group, and whatever other targeting options that might be relevant, to see how the different channels perform for each of these target subgroups.
With a bit of automation, managing campaigns can be greatly simplified, so if you organize your team well, one person should be able to manage multiple channels. If the number of countries and channels exceeds the capacity of the team, another option is to contract an external agency. It’s not my favorite option however, as I always prefer to run campaigns in-house. This is because you have better control of the results, react better, avoid agency fees, and because you know your product better than anyone else.
2. How much should you spend when testing a new channel?
The right spend always depends on the channel you are testing. On a platform like FB, Google, or Twitter, I start with the idea of spending up to 4K € during the testing phase (which is around 2 weeks). That means spending at least 250€/day, in order to give the algorithm enough impressions to optimize the campaigns towards the goals. That being said, if after a few hundred euros spent, I see that it hasn’t been working, I try to change the approach, either in terms of the creative, message or targeting.
For other platforms or networks, the testing budget can vary from 5 to 10K, and for programmatic, it can be up to 20K. But of course, it doesn’t mean that you have to spend the entire budget if it doesn’t bring any results, these are just testing budget estimates.
3. How long should you test a new channel?
This again varies, but in general, when we test a new channel we try to give it 10 to 15 days to see not only installs, but also the development of the cohort. If you are testing Programmatic the learning phase might be up to six weeks and it can be more expensive. Pre-installed channels are similar and have longer testing phases.
In general, it’s good to keep in mind that understanding a new channel takes a bit of time, and you should avoid the temptation to give up after 2 days, because it’s impossible to get significant results in such a short time. Especially for platforms like Twitter, Snapchat, or other social native channels, you might also want to test different concepts and angles before giving up, because we often tend to start testing with concepts/creatives from previously used channels, but each platform has its own characteristics and it might take some experimentation to get the message right.
4. Is channel specialization important?
Speaking from my own experience, I think it is good to be specialized in one or two channels, while also having a good grasp of the others. This allows all UA managers to be on the same level, and grants the opportunity for your team to rotate between platforms to avoid getting stagnant or stuck with the same channel forever. Rotating specialists allows them to learn new things and often generates new solutions to challenges – which for me is the best part.
5. But above all!
Don’t make assumptions about channels before you try them. Saying “this channel cannot work for my business” is giving up before the race has started. Each platform has its own characteristics, its own audience, and everything can vary from country to country, for different devices, genders, age groups, etc. Don’t get tired searching before you find the sweet spot. Managing multi-channel strategies is challenging, but worth it.