All the definitions of Breadcrumbs' core terms.
| APPLIES TO: |
| Free Plan |
| Pro and Enterprise Plan |
In this article, you'll find all the key terms and concepts to get the best out of Breadcrumbs.
Start with the foundation and the key terms to familiarize yourself with Breadcrumbs.
Learn all the terms you need to know when creating a scoring model and when analyzing your results.
FOUNDATION
In this section, you’ll find all the terms you need to know to familiarize yourself with Breadcrumbs.
Account
Your Breadcrumbs account is your centralized place where all the workspaces live. There can be one Breadcrumbs account per subscription and one account owner. At the account level, account owners can manage their profile, plan, and billing data and assign user permissions to specific workspaces while other users can view and manage the workspaces they have access to.
Account Owner
The account owner is the main user of a Breadcrumbs account. As an account owner, you can manage your profile plan and billing data and assign user permissions to specific workspaces. If you need to transfer ownership of your Breadcrumbs account, reach out to support, and we'll be happy to do it for you.
Activity
Activity refers to behavioral data such as email clicks, page visits, and trial sign-ups. When using Reveal, you can measure Activity impact by surfacing the events that occurred most frequently for contacts in your ICP. You can use these events, which are indicators of higher-value contacts, to create laser-focused activity models.
Co-Dynamic Distribution
At Breadcrumbs, we use an alpha-numeric score to give you an in-depth understanding of the potential of different contacts. This methodology is called co-dynamic distribution and was initially introduced to the world by Eloqua.
By default, it uses a 4x4 grid to represent the combined fit grade and activity grade of a contact. The top left of the distribution would include A1 representing your best fit and most active contacts. The bottom right would include D4 representing your worst fit and least active contacts.
Copilot
Breadcrumbs Copilot is a powerful tool that analyzes your data and generates a suggested model in just a few clicks. You remain in control and can make changes at any point. Copilot leverages the same always-on ML analysis of Reveal to build your data-driven scoring model, suggesting a list of high-impact attributes and actions.
Explore
Explore is a powerful analytics tool available at the workspace level. It allows you to view, filter, segment, and sort contacts and accounts on your objective and scoring lists based on attributes across any connected data source. It also unlocks powerful insights to compare scoring outputs side-by-side for any scoring model.
Learn more about how to use Explore to its full potential.
Fit
Fit refers to firmographic data such as Industry, Job Title, and Company Revenue. When using Reveal, you can measure Fit impact by surfacing the contact properties with the highest positive lift rate. You can use these properties to define your ICP and create laser-focused fit models.
Frequency
Frequency refers to how often a contact performs an action. In the context of lead scoring, the higher the frequency of certain actions (like viewing a demo page), the higher the likelihood that a contact is ready to convert (be it purchase, upsell, or cross-sell.)
Frequency is selected at the activity category level, and it will impact your final score grade.
Objective list
Your objective list is the list you are analyzing your scoring list against. It contains the contacts used to define success (i.e., paying customers.) ML will cross-reference your objective list with your scoring list to surface the properties that are the best predictors of revenue.
Primary Data Source
Your primary source is the main data source Breadcrumbs will use when analyzing and scoring your contacts. This data source can be your CRM, or Marketing Automation Platform.
Once your scoring model is active, the system will route back the information to your primary source so that you can set up triggers and workflow and act on your scoring results.
Recency
Recency refers to how recently a contact performed an action. In the context of lead scoring, an action taken today doesn’t have the same value as one taken two months ago. The more recent an action was, the higher the likelihood a contact is ready to convert (be it purchase, upsell, or cross-sell.) Conversely, a lack of recency for usage-related activities like product log in can result in a higher churn risk.
Recency is selected at the activity category level, and it will impact your final score grade.
Reveal
Reveal is a powerful engine that lives in your Breadcrumbs account. It analyzes your existing marketing, sales, and product data to highlight what attributes (Fit) and actions (Activity) are the best predictors of revenue.
Use Breadcrumbs Reveal to make data-driven decisions around the attributes that make up your ICP based on your best customers (or any customer segments) and the actions that predict a higher buying intent (or any conversions, such as upsell or cross-sell.)
Scoring list
Your scoring list is the list that contains the contacts you want Breadcrumbs to analyze. If you are on a Free, Pro, or Enterprise plan, this list will be used for your Reveal analysis and in all the scoring models you create in the workspace you're currently on.
Scoring Model
A scoring model is a tool used to prioritize contacts based on a set of rules built on specific criteria. You can use a scoring model to identify contacts that are ready to buy, find which contacts are perfect for upsell/cross-sell opportunities, or those at risk of churn.
In the context of Breadcrumbs, a scoring model has a Fit side and an Activity side. Once your model is live, the scoring information is routed back to your primary data source. You can have only one scoring model active at a given time for each workspace.
You can still experiment with multiple scoring models (and hypotheses) by setting your models to testing mode (learn more about Breadcrumbs A/B testing capabilities.)
Secondary Data Source
After adding your Primary data source, you can add additional data sources to enrich your primary source with other information. For example, your contact data may live inside your CRM, but your product data may lie within another tool. For a complete list of the secondary data sources supported, check here.
Time Decay
Time plays a meaningful (yet often neglected) role in lead scoring, and it’s never linear. Time decay in lead scoring refers to the gradual decrease in the scores value of a scoring model.
When using Breadcrumbs, your activity grades decrease over time at a customizable rate, and your contact’s score grade changes accordingly.
Users Permissions
The account owner of a Breadcrumbs account can grant permissions over one or more workspaces to multiple users from their Settings page. Users will have access to view and manage the workspaces they have access to, but won't be able to manage the account's profile plan, billing data, or assign user permissions.
Workspace
Workspaces are where your scoring models live. Since each workspace can have only one active scoring model at a given time, you can create multiple workspaces depending on whether your scoring operations are targeted at different ICPs, goals, product marketing motions, products, or geographies in parallel.
Each workspace has one primary data source and multiple secondary data sources that are independent between different workspaces. You can link your connected data sources to multiple workspaces.
SCORING MODEL CREATION
In this section, you’ll find all the essential terms you need to know when creating your scoring model in Breadcrumbs.
Absolute Decay
The absolute decay is a percentage you set up at the activity category level.
The Breadcrumbs algorithm allows you to give the correct credit to an action depending on how closer in time to the conversion that action happened.
The absolute decay indicates at which percentage this credit decreases and works together with the interval rate and maximum and minimum frequency to determine the score related to a given category.
For example, if the absolute decay is set at 50% and the interval rate is set at 3 days, an activity that happened 3 days ago gets half as much credit as an activity that happened yesterday.
Activity Category
Activity categories live in your Activity model and score contacts based on high-value actions, behaviors, or events, such as but not limited to, pricing page visits, email opens, and clicks.
You can customize each category to include weight, max frequency, min frequency, absolute decay, and interval rate.
Activity Model
The activity model represents one of the two dimensions of your scoring model. After defining your fit model, you can select a number of activity categories that score contacts based on high-value actions, behaviors, or events, such as pricing page visits, email opens, and clicks.
You can add several activity categories to your activity model and set different priorities for each activity category by adjusting their weights up to 100%. There are cases where you may want to score over 100% (while keeping the threshold at 100%.)
Learn how to build your Activity model in 10 steps.
Contact Routing
If you are on a Pro or Enterprise plan, Breadcrumbs lets you activate your scoring model and route back the score grades to your primary data source for further targeting and segmentation.
You can also set up rule-based criteria based on your scoring model that assign contacts to different branches of your customer journey based on the contact’s score grade. You can ensure they follow the same sequence of steps or assign them to certain activities. These features let you create more advanced journey flows and further leverage your scoring model.
For example, you can immediately assign your A1 contacts to your sales team or trigger a nurture campaign for B3 contacts.
Fit Category
Fit categories live in your Fit model and score contacts based on how closely they resemble your Ideal Customer Profile (ICP,) including demographic and firmographic data such as Job Title, Industry, Company Size, and Total Revenue.
You can customize each fit category by mapping the fields from your data sources and assigning specific values to each field. Additionally, you can select multiple fields to determine the scoring logic, such as assigning 20 points if ARR > 1M or Funding > 5M.
Fit Model
The fit model represents one of the two dimensions of your scoring model. Before defining your activity model, you can select a number of fit categories that score contacts based on how closely they resemble your Ideal Customer Profile (ICP,) including demographic and firmographic data such as Job Title, Industry, Company Size, and Total Revenue.
You can add several fit categories to your fit model and set different priorities for each fit category by adjusting their weights up to 100%. In some cases, you may want to score over 100% (while keeping the threshold at 100%.)
Learn how to build your Fit model in 10 steps.
Interval Rate
The interval rate is a timeframe you set up at the activity category level.
The Breadcrumbs algorithm allows you to give the correct credit to an action depending on how closer in time to the conversion that action happened.
The interval rate indicates the timeframe after which this credit decreases and works together with the absolute decay and maximum and minimum frequency to determine the score related to a given category.
For example, if the absolute decay is set at 50% and the interval rate is set at 3 days, an activity that happened 3 days ago gets half as much credit as an activity that happened yesterday.
Matching Rules
When setting up a fit category or activity category, you need to specify one or more rules you want to score your contacts against. There are several options, depending on the category you choose. For example:
Fit categories: the system will look at a MATCH/RANGE/DATE within the fields you select and assign a score based on the values that match. You can choose to include or exclude a specific field and assign a negative score to any of the fields. If no match is found, the score will be 0.
Activity categories: the system will look at the activity KEY, TYPE, and EVENT within the fields you select and assign a score based on the values that match. You can choose to include or exclude a specific field and assign a negative score to any of the fields. If no match is found, the score will be 0.
Maximum Frequency
The maximum frequency is set at the activity category level and indicates the maximum number of times an activity can happen and be relevant to the score. If an activity happens more times than the maximum frequency selected, the higher possible score is given.
Let's consider that the activity "pricing page visits" has a maximum frequency of 3. The highest possible intent is defined when "a user visits the pricing page more than 3 times." This means that if the activity happens 3, 4, or 5 times, it's always considered the highest possible intent without increasing in value after that number.
The maximum frequency works together with the interval rate, absolute decay, and minimum frequency to determine the score related to a given category.
Minimum Frequency
The minimum frequency is set at the activity category level and indicates the minimum number of times an activity can happen and be relevant to the score.
Let's consider that the activity "website page visits" has a minimum frequency of 10. This means that only users with at least 10 website page visits will be scored.
The minimum frequency works together with the interval rate, absolute decay, and maximum frequency to determine the score related to a given category.
Overlapping Categories
When creating your fit model or activity model, the traditional approach is to use a 100-point system (or 100%.) Sometimes it makes sense to overrule this approach.
For example, when you aim to double down on your ICP, you can use scoring categories that sum beyond 100, where two or more categories have predictive value but are mutually exclusive. This allows you to achieve a higher hit rate as you’ll look at more contributing categories.
Rocket Category
A rocket category is an additional category that you add on top of your 100-point system because you know that when that high-value activity occurs, your lead is hot, and you want to boost its score.
For example, your lead answers a DM on LinkedIn. This action signals that the lead is ready to talk to you, and you want to give it a higher priority in your scoring model.
Slider Bars
After creating a scoring model in Breadcrumbs, the score grade combines fit grades and activity grades, creating a co-dynamic quadrant of 16 different buckets.
Typically, a high-performing scoring model has 5-15% of scored contacts in the top-left quadrant. By using slider bars when creating your scoring model, you can easily adjust weighting for grades, fine-tune the threshold for each score bucket, and see how it will affect your overall distribution in real-time.
This allows you to understand how your contacts are distributed in terms of fit and activity grades and improve your model accordingly. Additionally, you get a curated list of your best contacts to follow up with to meet your monthly targets.
Weight
Weight refers to the percentage each fit category or activity category carries for the overall fit model or activity model, respectively.
To decide what your category weight should be, you'd need to answer this question:
The traditional approach to scoring suggests that any model should reach 100% (or a 100-point threshold.) We chose to overcome that limit and give you the ability to exceed 100% when creating your fit model and your activity model to add more contributing categories while keeping the threshold at 100%.
By allowing more contributing categories, you don’t miss out on relevant signals based on an arbitrary scale of 100, even if the threshold is still 100, and it still lets you act on the score with triggers, workflows, and processes.
If you want to learn more about this topic, read our article on why the 100-point rule of scoring is wrong.
DATA ANALYSIS
In this section, you’ll find all the essential terms you need to know when analyzing your scoring model results in Breadcrumbs.
Activity Grade
In a co-dynamic system, an activity grade is a number from one to four that represents the sum of the weights of your activity model. One being the most active, four the least. Since your activity model’s threshold is 100%, a 1% to 100% scoring scale defines the activity grade.
By default, if the sum of your activity weights is between 1% and 30%, it’s a grade of 4; if it’s between 31% and 60%, it’s a grade of 3; if it’s between 61% and 80%, it’s a grade of 2; and finally, if it’s between 81% and 100%, it’s a grade of 1.
You can easily fine-tune the threshold for each score bucket and see how it will affect your overall distribution in real-time through customizable slider bars that you can set up when creating your scoring model.
Category Best Match
The category best match is the category that has the highest hit rate among all the categories that make up your scoring model. You will have one best category for your fit model and one for your activity model.
Cohort
Generally speaking, a cohort is a group of people who share similar characteristics. In the context of lead scoring, a cohort is a group of contacts that have similar characteristics in terms of fit or activity.
Conversion
A conversion is an event in which a contact takes an action (such as making a purchase or signing up for a service). Tracking conversions helps you measure the effectiveness of your lead-scoring program by showing how many leads become customers.
Conversions, however, do not only refer to customer acquisition. Depending on your objectives, a conversion can be an upsell or cross-sell event as well. By understanding conversion rates, you can customize lead scoring models to focus more on high-value contacts and prioritize them accordingly.
Fit Grade
In a co-dynamic system, a fit grade is a letter from A to D that represents the sum of the weights of your fit model. A being the best fit, D the least. Since your fit model’s threshold is 100%, a 1% to 100% scoring scale defines the fit grade.
By default, if the sum of your activity weights is between 1% and 30%, it’s a grade of D; if it’s between 31% and 60%, it’s a grade of C; if it’s between 61% and 80%, it’s a grade of B; and finally, if it’s between 81% and 100%, it’s a grade of A.
You can easily fine-tune the threshold for each score bucket and see how it will affect your overall distribution in real-time through customizable slider bars that you can set up when creating your scoring model.
Grade Description
The grade description is a detailed explanation of your scored contact's final grade. It includes all the scored categories that make up a contact’s grade both from your fit and activity model.
For example, if your fit model targeted Job Title, Industry, and Total Revenue, a contact’s grade description will include which of those categories are true for that specific contact and how much they weighed in their total score grade.
Lift
At its core, lift is directional data. It helps you validate your assumptions that certain properties of your contacts (that is, firmographics, demographics, actions, or behavior) increase (or decrease) the chance of a conversion.
It is calculated as objective list fill rate/scoring list fill rate.
A positive lift indicates that the property correlates with a conversion, while a negative lift indicates anti-correlation. Furthermore, a low lift (positive or negative) indicates a poor correlation with a conversion.
For example, if 50% of your leads live in the US and 50% of your leads live in Europe, but 80% of your customers live in the US and only 20% in Europe, this means that a US lead has a higher likelihood to convert and will have a higher lift.
Score Grade
The score grade is your final co-dynamic score. It is the combination of Fit Grade (A-B-C-D) and Activity Grade (1-2-3-4.)
For example, a contact in your ICP who is highly engaged is A1, while a contact who is not in your ICP and not engaged is a D4.