MMM Case Study: +13% in conversions with the same budget invested

Liked this post? Share it!

Share on facebook
Share on linkedin
Share on twitter
Share on email

The Question:

If I have 14 Campaign Types in my Marketing Mix how much should I spend on each channel to maximize my return?

The Brand:

The company sells shipping for eCommerce businesses.

It basically allows the eCommerce business to buy cheap shippings that deliver a better service for the end client compared to the rest of the shipping suppliers here in Italy.

The Challenge:

As a business born and raised during covid had an incredible boost in growth that made the company more than double its marketing investments.

From investing just on google search, started investing on Facebook, programmatic, bing, youtube, display, performance max, etc.

After multiple conversion lift experiments we noticed that the attribution was broken, 60% of the conversions went into a “Direct” Source.

Result on Google's SERP when searching 'attribution problem'


How can you plan your advertising budget if you can’t know what ROI got in the past?

How Can I better allocate my advertising budget to experience incremental sales?

The solution:

The marketing team leveraged Cassandra MMM tool to evaluate the performance of the past investments discovering what real ROI got each channel

Other than historical attribution, Cassandra analyzed and created diminishing returns curves for all the advertising channels.

These Diminishing returns told the marketing team what was the breaking point at which each investment started to become inefficient.

result on Google's SERP when searching 'cassandra'

After an initial analysis, the marketing team used the Cassandra Budget Allocation tool to double-check their budget distribution hypothesis.

They asked the model to predict what to spend on each channel for the next two weeks and the model told the marketers to reduce the Meta advertising budget and allocate it to google search and google video.

This new allocation would’ve increased their conversions by 5.5% compared to the previous period

The distribution is implemented

The Results

After two weeks we double-checked the results.

We experienced a 5.75% increase in the number of conversions compared to the previous period

Cassandra refreshed its model on the new data and gave a new budget allocation for the following two weeks.

Cassandra suggested:

“Lower your budget by 12% on Google Search ads and reallocate 8% to google discovery & 4% to Meta “

This new change would’ve increased our total number of transactions by 6.6%

The distribution has been implemented.

After another 2 weeks, we double-checked the results and we experienced a + 7.6% in the number of transactions

Summary of the results

In 1 one month and one week, just modifying the budget distribution we’ve been able to obtain:

  • +13.36% in the number of transactions
  • +42.21% in profit
  • +22% in ROAS
  • -7% CAC (thanks to an additional Geo-lift experiment)

How can I implement this solution in my organization?

In Cassandra, we have developed a solution composed of 3 steps:

Step 1: In 2 weeks you can receive a personalized MMM for your company with a money-back guarantee offer if you don’t like the outcome.

Step 2: In 2 weeks we create a data infrastructure able to centralize all your data daily and refresh the model every two weeks.

Generate a dashboard for your company: SEE THE EXAMPLE

Step 3: 2 hours of monthly consultancy to transform model insights into experiments that will drive better ROI over time.

If you want to learn more and verify if this solution could bring value to your company I invite you to click on the following link and book a 30 minutes call with me to see if and how we can help you

Click Here To Book a call with Gabe

More Free resources for you

Reducing CPA by 30% with

How to optimize CPO by identifying marketing budget waste is one of the biggest sellers in the shipping industry in Italy.  It had an

Find out how Cassandra can help you

Click below to book a call with one of Cassandra's founders to learn more about how it works and get the most out of your data