InMobi inDecode - How to Acquire & Retain High LTV Users

Editor's Notes

• #5: 1 - [Connect the idea of user lifecycle and lifetime value] Some truths we know about mobile users, they may click on your ad they may install your app they may open the app everyday they may even buy once BUT the very next day, they may also disengage & uninstall AND every user behaves differently 2 - [Define quality users as those with high lifetime value] Every marketer’s dream, users that try -> engage -> buy -> keep buying -> often & a lot -> tell their friends. In short - quality users quality users = valuable, loyal users with longevity = users with high lifetime value 3 – [Example from the gaming world to illustrate the point] 750 new games are released every day Less than one third of all users will pay for an app Less than 3% of mobile gamers make an in-game Purchase But only 0.15% of mobile gamers account for 50% of in-game revenue
• #6: 1 - [Distinguish actual vs predictive LTV] For existing users - we can measure lifetime value based on actual behavior For new users - we look at how they behave soon after installing the app and approximate their future behavior. In other words, we analyze a user’s in-app install and post-install events to predict the potential dollars that a user will contribute across their lifetime, i.e. their lifetime value. 2 - [Explain the value of LTV based acquisition & optimization, and sharing post-install data] Tracking that post-install user behavior provides insight into the user segments that are more likely to be higher quality, and passing that to channel partners means those cohorts can then be more precisely targeted. The key is understanding the behavior signals that are most likely to indicate your quality users, in other words, figuring out the most important proxies to predict LTV.
• #7: LTV can be leveraged for both, Install = UA – drive quality users through discovery, optimize all elements of your acquisition campaign and adapt your spend Post-Install = UE/UR - set additional campaign goals that drive users further down the lifecycle towards engagement, re-engagement and retention So today’s smart marketing strategy has 2 equally essential parts, identifying the best characteristics of an entire population to drive acquisition of quality users. Here you start by having a few data points on a large number of users. optimizing based on individual users, leveraging all the install and post install data available to drive optimal user behavior. Here you analyze a huge number of data points on a small number of users. Both parts have to be seamless. Optimize on the basis of one unified user journey funnel as opposed to two perceived independent funnels (awareness/install vs. post-install). Because getting more relevant users to click-through the ad and install the app can translate into higher retention downstream, and by proxy, high LTV users. Key question:  how do you choose whether to acquire new users or re-engage existing users? => must be capable of measuring and optimizing to individual user Lifetime Value How do you know whether to spend on Installs or Reengagement? “Easy.” ROAS = Total Revenue per Day [ / Week / Month ] divided by UA Cost [ ave by time ] => maybe “Our new users are buying more than our older users” => maybe you need to more effectively engage your existing users [ while you can ] Best answer = UA ROI <vs> RE ROI UA ROI UA Cost * Ave Lifetime Value * factor for Optimizing Future Incremental Revenue * longer-term time value of money RE ROI RE Cost * REmaining LTV * short-term factor for time value of money 
 UA Cost high vs RE Cost low Ave LTV vs REmaining LTV => how big a factor is user age / existing purchase history? => how good are you at identifying / predicting / incenting “whales”
• #8: -CPQ vs RoAS -Evolving towards Cost per state-change. Retargeting in this context is different from affiliate traffic drives, more CRM.
• #10: Optimization is a continuous process, so the data integration and flow needs to be, automated, always-on process. It is also highly time-sensitive, so the flow needs to be real-time or near real-time, elements like recency & frequency are critical. Post-install events need to be leveled, so that optimization can happen at every level, not just at purchase, thus reducing spend waste at every step in the user journey, ultimately pushing up RoAS Trusted partners like InMobi safeguard your data, give you comprehensive control over it, and have clear data protection policies and systemic processes to enable them. Ultimately, you remain firmly in control of your data, sharing it only to unlock higher value