In many organizations, 50% of sales energy is spent on prospects who never convert. What if you could predict which leads are likely to have a need for what you are selling? This is what MadKudu’s customer fit model is here to help with.
What is it?
MadKudu’s customer fit models automatically research your leads and identify those who are a good fit to buy from you.
MadKudu augments your leads with firmographic, demographic, and technographic data (eg. company size, industry, location, capital value, job title) and finds from historical conversions which lead profiles are most likely to end up becoming a customer.
MadKudu labels your leads with segments simple to understand and to act upon. The possible values are:
- very good
Leads with a “very good” customer fit are usually seen to convert about 10 times more than leads with a “low” customer fit.
NB: if your pricing is built upon multiple tiers or contains an Ent. plan, the customer fit is a prediction of conversion for your highest tier or Ent. plan.
How is it used?
Here are some of the most common use cases.
Faster lead routing
By identifying which leads have potential (customer fit is “good” or “very good”), your sales team can contact the top leads faster, and give them more attention. Top performing sales team implements a SLA that takes into account customer fit segment. For example, a typical SLA is:
- leads with a “good” or “very good” customer fit should be contacted within 4 hours.
- leads with a “medium” customer fit should be contacted with 48 hours.
Increased sales efficiency
You can use customer fit to identify leads who are not worth pursuing. Those are leads who aren’t a fit for your product (“low” customer fit). At the extreme of that are the junk leads (yes, I am looking at you “email@example.com”). The faster and the better you identify low quality leads, the more time the sales team can spend selling to high quality leads.
Better marketing perfomance
Grading your incoming leads is extremely helpful to rapidly determine which marketing channels bring high quality leads and which ones bring you just noise. Using the MadKudu customer fit, you don’t need to wait weeks to find out how good the leads of a marketing campaign are.
Customer fit is a great tool to rapidly extract high potential leads from any list. For example, find out how many people have a “good” or “very good” customer fit in the attendee list of an event to determine how valuable it can be. Or extracts the leads with most potential from this prospect list that you’ve obtained.
Keep out junk leads from getting into your CRM (ie. Salesforce, Hubspot).
How is it computed?
For MadKudu Startup plan, the customer fit is an industry-specific model that scores every new lead based on aggregated trends.
For MadKudu Growth and Pro plans, the customer fit model is trained against your existing paying customer database.
It isolates signals that determine the likelihood of a leads to convert based on traits including:
- Company Size
- Company Industry
- Enterprise Technologies used (Marketo, Salesforce, Taleo, Adobe Omniture…)
- Retargetting Technologies used
- Technologies (amount, type…)
- Company Geo-economics (GDP, language…)
- Company Tags (B2B, SaaS…)
- Company financials (funding, growth, market value…)
- Job title (seniority, department)
Using its historical the training, the model will then score any new identified lead or user based on those signals.
For MadKudu Pro and Ent API plans, you can also send us a training data set via CSV using the following template
For MadKudu Pro plan, the model is reinforced on an ongoing basis based off of CRM statuses.
Leads that are strongly rejected (or unqualified) and leads that convert to SALs/SQLs (sales accepted or qualified) constantly define new data points for the model to increase its accuracy.
What are the default signals
- personal email address
- fortune 500
- firmographics profile
- potentially spam
- employee count
- number of employees
- team email
- test email
- market cap
- capital raised
- web traffic volume
- based in a country where gdp per capita
- enterprise tech found on website
- belongs to industry where revenue per employee
- number of tech found on website
- twitter followers
- company founding year