Customer Fit Insights - See how variables influence the conversion rate
The Customer Fit model is pretty much the foundation for success when using MadKudu. Once you build it, you can reuse the score in a few other places (at no cost with practically zero effort) to accelerate and optimize your funnel.
The TLDR -> more opportunities to make money.
So let's say you've integrated your CRM, you've seen the results first hand...but what is it really based on? This support article aims to demystify that question and to help you understand your data at a deeper level, using the functionality provided to customers via the MadKudu App.
Why is access to this data cool? You can see the traits and their subcategories that are indicators of your ICP (Ideal Customer Profile) that are:
Core - your bread and butter traits, typically convert pretty well
Niche - leads with these traits convert well, you just don't have that many
Anti - traits that are indicative of someone who won't convert
That said, MadKudu looks at ALL of these traits and subcategories, and statistically determines an overall score so you don't have to. Using the insights, you may learn more about your leads and you may find gems of opportunities - either of the traits of leads to pursue or leads to avoid.
But above all, remember we want the model to do the number crunching and the analysis so your team can trust the output and focus on identifying opportunity and generating more revenue.
I. Access your App
- Go to the Predictions Tab
- Click on the Model Tile
- Scroll down to the Customer Fit Insights section of the page (note the earlier parts of the page are currently under development and will be released soon!)
II. See all the Traits that comprise your score
- Either type into the Search box what trait you are looking for e.g. Company Size, Funding, Role, Country etc. or, scroll down and preview the list of ~95 - when you find something you want to dive into, simply click on it.
III. Understand the Page Layout - 3 Main Components
- Breakdown of the trait population and conversions into subcategories
- Subcategories reflected by overall Lift
- The data behind the graphs
III.1 Breakdown of the trait population and conversions into subcategories
- This is the overall population of leads, in this image - Blue has approx. 68%, Red has approx. 13%
- This represents the subset of leads who converted and the percentage of conversions per category. For example, of the population who converted, the percentage of the Blue conversions is 71% and the percentage of the Red conversions is 6%
- This is the legend to decode the graphs. Blue is the United States, Red is France.
1. Core ICP Trait Example
Looking at the % population bar for the United States & at the % conversion, you will notice that the values are pretty close, 68% of the population of leads is from the United States, of the overall population who convert, 71% of those are from the United States.
When the bars are pretty close in terms of size, this is typically indicative of a core trait.
A core trait is probably not an eye-opener to you as this is most of your conversions. This is what you expect, it is predictable. This is what you want the model to predict - those who convert predictably based on the traits of those who have historically converted.
2. Anti ICP Trait Example
Looking at the % population bar for the France and at the % conversion, you will notice that the values are noticeably different, 13% of the population of leads is from France, of the overall population who convert, 6% of those are from France.
When the bars differ in terms of size, getting noticeably smaller from the % population to the % conversion (or left to right) this is typically indicative of an anti-ICP trait.
Similar to a core trait, an anti trait is probably not an eye-opener to you - as this is probably who you already do not target - whilst there may be some conversions, it's an even bigger time consumer.
Who to avoid? It can be predictable - though, not everyone has the ability(or time, or energy or tools) to drill down into the numbers this way, this type of analysis may reveal some new insights. And likewise, you want the model to be able to predict which leads will not be the best use of your time(and energy and ability).
3. Niche ICP Trait Example
Next, move up the chart and look at the purple bar - purple is our friends in Poland.
Converse to an anti trait, if the size of the bar gets larger from left to right, this is an indicator of a niche ICP.
A niche ICP is when there is a small population of leads, but comparatively, they convert pretty consistently. When you see a niche ICP, ask yourself, can you generate more leads of that type?...because our friends in Poland...convert pretty well, we may have just overlooked them so far.
III.2 Subcategories reflected by overall Lift
To understand this chart you need to understand what Lift is. Lift is the difference when comparing one category's conversion rate with that of the average conversion of the entire population(e.g. all the categories for that specific trait).
Solely looking at this graph, you may be tempted to think that Poland converts better than the United States. Poland doesn't. In english, this graph is telling you:
"When I consider Poland's conversion rate, with the average conversion rate for the entire population of the Company Country trait, Poland converts better than the average conversion rate of the entire population."
Lift is a statistical term that helps you understand a category against an average for a total population. What you also want to know is, if Poland has more actual conversions than the United States...keep reading.
III.3 The data behind the graphs
We know from the previous section that Poland has a higher lift than the United States - but how do we calculate and how do we interpret it beyond just the lift value.
1. The average conversion rate for the entire population = 5.04
2. United States category only Conversion Rate = 5.28
3. United States Lift Factor = 0.05
4. Poland category only Conversion Rate = 15
5. Poland Lift Factor = 1.98
United States Lift Factor
5.28 - 5.04 / 5.04 = 0.05
Poland Lift Factor
15 - 5.04 / 5.04 = 1.98
Ok - now what? Well, whilst the category may be the best converting compared to it's trait-mates - is it all you should focus on? No. Let's see that table once more:
6. is the actual number of conversions in the US
7. is the actual number of conversion in Poland
If I am equally concerned with making money as I am with lift - 6 & 7 tell us where the money is being spent and therefore made - the United States.
Ultimately, lift is a great data point - but to validate if your highest lift is your highest converter - you need to look at the data manually. As they say, the devil is in the details. Ask your friendly Customer Success team if you have any questions on interpreting your data.