Infer Terminology
Understand the main Infer terminology so you understand what you're looking at when you look at your Infer profile
Basic Infer terminology to understand how each of the features work and how to understand concepts inside the tool and from the documentation.
Conversion (eg "lead to conversion") - This refers to when a lead (potential customer) "converts" to an "opportunity" (a serious chance at selling the product).
Win (eg "lead to win") - This usually refers to when a lead (potential customer) buys the product the sales team is trying to get them to buy. This is as opposed to a mere "conversion" (see above).
Target - The target of a predictive model. Could be the creation of an Opportunity, a Closed Won Deal, etc...
PMP - Infer’s Prospect Management platform enables segmentation using thousands of internal and external data signals, as well as predictive lead scores, which provide the most complete view of each prospect and help marketers and sales professionals determine their next-best programs and tasks.
QBR - Quarterly Business Review. Infer teams analyzes and meets with each customer to understand if a model refresh is needed.
Signal/Feature - This is the same thing as a "signal". Sales usually uses the word "signal" as it's easier for customers to understand, while Eng usually uses the word "feature" as that's the word used in the machine-learning literature. For example, one signal/feature is "Website: Hosting Provider Name is 1&1 Internet AG".
Converted Rate - This is one of the signals indicatives to understand if they’re good or bad. It allows us to understand the percentage of entities from the total that were converted.
Lift : Also about the signals, this is the conversion rate of the signal compared to the average conversion rate
Coverage: The percentage of all entities that contain this signal.
Converted Coverage: The percentage of all converted entities that contain this signal.
Behavioral model: Infer’s Behavior models incorporate information from Marketing Automation activity tables into a temporally relevant score. The behavior model can look at Leads or Contacts and predict the likelihood to convert to either an Opportunity or a Closed Won within a 3-week time period. The Lead or Contact is then scored with a 0-100 score and then generally groups into four groups [1-4].
Fit Model: A Fit model is used to provide a predictive score for entities like leads, contacts, opportunities and accounts. For example, leads are typically sparsely populated with data on creation, so Infer uses a limited number of fields at the outset of scoring in order to provide the most accurate score based on information that we can find about the Lead.
Smart Signals: This is one of Infer features that allows the customer to access some of the signals used on the models, incorporating them on your salesforce. Examples: Fortune 1000, Global 500, City, Competitors, Company ID, Name, Country, # of Employees and more.
Glance: Glance is an Infer feature designed to provide additional insights into why a Lead, Contact, or Account scored the way that it did. Glance also saves reps effort by surfacing key company information.
Did this answer your question?