MadKudu Docs


Likelihood to buy

In most B2B organizations, lead conversion rates hardly reach the one digit range. What if you could predict which leads are likely to convert based on their behavior? This is what MadKudu’s likelihood to buy model does.

What is it?

MadKudu’s likelihood to buy models automatically learns from historical patterns to uncover specific behaviors that separate leads who were on the path to conversion from others.

MadKudu continuously scores all your active leads based on their behavior (in-app activity, marketing & sales interactions…) to determine which are on the verge of closing.

MadKudu labels your leads with segments simple to understand and to act upon. The possible values, explained below, are:

  • low
  • medium
  • high
  • very high

Ready to buy Moments

MadKudu creates a “ready to buy” moment as soon as a free user become highly likely to convert.

This moment is then used to trigger an email aimed at converting the user.

How are the “ready to buy” moments computed:

  1. Data Preparation
  2. Feature construction and selection
  3. Model training
  4. Export to your marketing automation platform


Likelihood to buy segments

When ranking users by probability of conversion, we create the following groups:

  • Top 1% = “Very High”
  • 2-10% = “High”
  • 11-30% = “Medium”
  • rest = “Low”

Power Curve

The power curve is a way of looking at the cumulative amount of total conversions each segments accounts for.