Reducing user acquisition costs with marketing mix modeling | Pecan AI
Introducing Predictifier

Dive in and see how your business can drive success with AI predictions

Campaign Optimization
Gaming & Apps
Lifetime Value

Reducing user acquisition costs with marketing mix modeling

Delivered in just 3 weeks, Pecan’s MMM offered granular budget insights and simulation tools that helped the team reduce its most popular app’s customer acquisition cost by 10% in the U.S.

↓ 10%
reduction in cost per acquisition
3 weeks
turnaround for a complete MMM model
↓ 7%
less decrease in users acquired, in comparison to other countries with similar budget reductions
Industry
Mobile app publisher
Company Size
600M+ downloads; tens of millions of monthly users; over 5M monthly subscribers
Solution
Optimizing marketing spending
Platform Use Case
Marketing mix modeling (MMM)
Data Stack
Google BigQuery

The app developer behind popular image enhancement apps faced challenges in measuring marketing impact due to recent changes in mobile data privacy standards — especially Apple’s SKAdNetwork. Seeking a fast, future-proof answer, their UA team partnered with Pecan on a marketing mix modeling (MMM) solution. Delivered in just 3 weeks, Pecan’s MMM offered granular budget insights and simulation tools that helped the team reduce its most popular app’s customer acquisition cost by 10% in the U.S., where the app gains 45% of its new users.

Challenge

Finding future-proof sources of certainty in a fast-changing mobile app ecosystem

This mobile app developer builds popular image enhancement apps. The company has achieved impressive success, with hundreds of millions of app downloads worldwide. 

An effective marketing engine has fueled this success. But then Apple’s iOS 14.5 changed the game for mobile app advertising. With a customer base skewed toward U.S. iOS users, this company lost granular insights about much of its user base. 

Without that vital information, the company’s user acquisition team needed a way to better understand the effectiveness of their campaigns and strategy — fast. 


Solution

Responding rapidly to the changing mobile ecosystem with Pecan’s SaaS MMM solution

The UA team sought to accurately understand their campaigns’ effectiveness, including the incremental effects of multiple channels. Determining those effects is tough without complete attribution data. 

MMM is an excellent way to gain a holistic understanding of the true incremental contribution of all marketing channels, though it can be complex to implement. The UA team recognized that working with an MMM SaaS provider would dramatically accelerate results, so they partnered with Pecan to adopt an MMM solution quickly. 

Attribution and measurement methods that rely on user tracking data today fail to provide enough information. That makes MMM a critical part of a complete marketing measurement toolkit. With a software solution, this team fast-tracked significant improvement in marketing efficiency.

The UA team provided Pecan with 3 years of historical data from two of its apps, including details on geolocation, OS, spend, and paying subscribers. 

Pecan’s MMM solution was ready in just 3 weeks. The Pecan team’s experience and the platform’s automated data connection and preparation capabilities accelerated the process.

Results

Delivering fresh insights into cost per acquisition and channel effectiveness

The UA team’s primary goal in using Pecan’s MMM was to reduce customer acquisition cost (CAC) by gaining a better understanding of marketing channels’ incremental impact and allocating spend most efficiently. The model predicts the impact of each channel on the number of Day 7 paying subscribers. 

Compared with the previous business-as-usual approach, decisions guided by Pecan’s MMM helped the UA team decrease the CAC for its most popular app by 10% within its most significant geographic area. At the same time, a similar app in the same geographic location and platform saw a 27% increase in CAC. That increase reflects the difficulty of acquiring users at a reasonable cost in today’s ecosystem.

As a whole, Pecan’s MMM helped reduce the cost of user acquisition significantly. The UA team is also now able to make much more confident cross-channel budget decisions, knowing in advance the results they’re likely to see.

That notable improvement has been made possible by the deep insights into marketing effects provided by Pecan’s MMM solution. The MMM model is refreshed weekly to ensure the insights remain current and actionable. The model is highly accurate (wMAPE of 6.8% and R2 of 0.75). The marketing team can make decisions about budget allocation frequently and rapidly to maximize results. 

With the platform’s easy-to-interpret dashboards, the UA team can examine budget decomposition over time, including budgetary and non-budgetary components. The team can also deeply analyze channel saturation and scrutinize anomalies in channel behavior. 

All this information is available for each of the company’s apps, ad platforms, countries, and channels, empowering detailed, future-driven decision-making about budget allocation.

The UA team can also use Pecan’s built-in simulation tools to test different budget scenarios to minimize CAC. Optimizing every marketing dollar’s impact lets them spend smarter and more efficiently with MMM.

Pecan’s MMM captures and accounts for the full range of this app developer’s marketing efforts, even where reliable data is sparse or inaccessible. With such detailed insights and Pecan’s tools, this UA team is ideally positioned to make data-driven decisions about their marketing efforts, even as the mobile app ecosystem evolves.

Contents

Accelerate your team’s journey to AI success.

Most data analysts using Pecan successfully build their first machine learning model in under a day.  Ready to bring powerful machine learning and its business benefits to your organization?