Introducing Predicted ECR
What is Predicted ECR?
Predicted ECR is our latest Machine Learning model designed to forecast how advocacy is likely to evolve over the next 12 months within your existing customer base and what you can do to improve and exceed those predictions.
You can use Predicted ECR to see:
- Which of your fans are likely to keep advocating
- Which of your fans are likely to stop advocating
- Who your future advocates are likely to be
Target these segments with unique experiences to improve and exceed those predictions.
Build your long term strategy with confidence by understanding your customers advocacy behaviour now and into the future.
How is it calculated?
We use historical patterns to predict customer lapsing, purchase and referral behaviour in the next 12 months. This gives us personalised propensities for each customer, enabling us to calculate how the make-up of your base might change. The prediction takes into account a customers recency, frequency and value (RFV)s for both orders and referrals, as well as other factors, in order to make the prediction as accurate as possible.
How do I see my Predicted Advocacy Shift?
For those on our Optimise, Advanced or Ultimate package you will now be able to see a section in the Advocacy Intelligence > Customer Revenue.
Scroll down the page (where you can see the breakdown of your current customers Extended Customer Revenue) and you will find a visualisation of your Predicted Advocacy Shift.
The Sankey chart is a visual representation of what we expect to happen in a year's time to your existing customers.Visualising the movement between segments is a powerful way of showing what is likely to happen to the existing segments based on what you have been doing, and allows for a quick understanding of where the most movement is and what areas need to be focussed on to improve your outcomes for the year.
NB. This is only for your existing customers, we don’t have the data to predict new customers.
Why are you only predicting for existing customers?
For customers who you will acquire in the next 12 months, we don’t have the data to make a prediction. You will see those customers expand your current and predicted base over time as they are acquired.
How do I find out who is predicted to shift?
The insights shows segments of customers to target and clicking the ‘Take Action’ will take you through to a download of who is in these segments.
We have three key segments :
- Future Advocates - These are the ones most likely to go on to advocate this year, and should be given the opportunity to with targeted calls to action
- Likely to lapse - These are the customers we predict will no longer be purchasing or referring. Depending on the objectives of the client, these should be customers targeted with reactivation campaigns, either purchase or referral to keep them engaged. We would recommend they experiment with what works best but putting a referral spin on it means if they are not in a position of want or need, they can still generate value and remain engaged via referral.
- Future High Advocates - These are the most valuable advocates and should be given a great experience to maintain that advocacy, identify them in your system and keep them happy.
What are the data requirements?
To build the models required to make these predictions and produce these visualisations and data we have some minimum data requirements. For Predicted ECR we need:
- At least 100k customers (distinct emails) in the data you have provided to MentionMe (either through the referral program or historic order uploads)
- 3 years (no gaps) of order data
- 90 days of referral
- Currently we cannot make predictions for subscription businesses, but if you’d like to help us develop this please reach out to your CSM today.
FAQ
- What package is it?
- Access to our Predicted ECR feature requires you to be on our Optimise, Advanced or Ultimate package.
- How do I provide feedback?
- Contact your CSM today to send feedback or reach out to our product team directly product@mention-me.com
