Unlocking Sales Success with Model Builder: A Practical list of Sales Use Cases
This will be a series of posts detailing Predictive Use Cases that can be built with Data cloud’s Model Builder. The first article in this series is aimed at Sales Use Cases, which can improve productivity of sales teams and help achieve their targets faster 🎯.
Bonus: A ready-reference of this blog is available as a pdf in our revamped Help article section. Here is the direct link.
The State of Sales teams in today’s world
Sales teams today face more pressure than ever before. High volume of leads (quite possibly of low quality), tougher economic environment with the wallets tightened, and ever-changing customer expectations. And let’s not forget the constant worry about customer churn and the burden of manual administrative tasks. All of these lead sellers to feel the pressure of meeting targets or unfortunately, missing them. The pressure then cascaded to the Sales leaders who are answerable on their team’s performance, gaps on forecast accuracy,
Just look at some of the statistics from a recent Salesforce report. They reinforce what sales teams face in today’s world. So where does this leave sales teams and what’s the solution. Let’s talk on the power of Predictive AI.
Power of Predictive AI for Sales
Predictive AI has been a powerful helping hand to solve these challenges via predictive scores that can be seamlessly incorporated into the business processes. Predictive models helps sales reps see which leads are most likely to convert, which deals are on the brink of closing, and which customers are at risk of churning. That’s the magic of Predictive AI. It can provide you with insights that transform how sales reps approach sales. After all, Sales is an art and Sales reps need their tools to sell more, sell faster and sell better.
Data cloud + Model Builder = Sales Insights
How to get started on building Predictive use cases for Sales teams. In companies where all of the myriad sources of data viz. CRM metadata, engagement data from different channels are already unified, harmonized in Data cloud, all you need to do is get started with building these on the No-Code Model builder.
Do read our blog Build an AI Model With Clicks In Data Cloud for a comprehensive introduction. We have a lot more published, check out the resources sections at the end of this article.
With your models built, you should be able to embed the predictions wherever you want: Object pages as a LWC component, or use in flows, prompt builder, and so on. Here is an illustrative LWC component detailing out the prediction, the top factors and the recommendations to improve your outcome.
Specific challenges of Sales teams
Before diving into use cases, let’s first understand some key challenges faced by sales teams. Among their many concerns, they are most obsessed about low deal win rates, low contract renewals, and high churn rates. For sales teams, it all boils down to one thing—hitting their targets.
However, obstacles like the inability to prioritize hot leads, prolonged deal cycles, and inaccurate forecasts have a direct impact on these outcomes. These issues not only hinder sales performance but also create frustration, as failing to meet targets often results in lower performance ratings. And let’s be honest—who wants that?
A Practical list of Sales Use Cases:
To tackle these challenges, let’s explore how predictive AI can provide solutions through a variety of tailored predictive models:
#1 Increase lead conversion
#2 Improve win probability
#3 Decrease time to close
#4 Predict expected revenue
#5 Decrease churn probability
#6 Improve sales rep performance
#7 Increase contract renewals
#8 Increase quote acceptance
#9 Increase repeat business
#10 Increase lifetime value
Impact of B2B vs B2C on the use cases
- Lead scoring, Churn and some of the other use cases listed here are applicable for both B2B and B2C.
- If its B2B, then collect data on the account – demographics, firmographics, and geo data and their engagement
- If its B2C, then collect data on the customer / individual – demographics, geo data and the customer’s engagement.
Structure of the use case
The use case covers some context on some key components:
- Recommended Model Builder setup:
- Possible data sources that can help you get started
- Recommended Model type
- Objective function
- Note: The setup and data sources outlined here are recommendations designed to provide a starting point and inspire ideas for your business. Predictive models can be formulated in various ways with different variables, as there’s no one-size-fits-all approach to predictive modeling. Start with this suggested setup, and as you progress, consider exploring additional variables tailored to your business and specific use case.
- How to use the model outputs in business: This section is to call out how the predictions can be used in the flow of work, define business processes based on the scores
- Outcomes driven with this model: This section calls out the outcomes that can be achieved with the specific model.
Now let’s dive into the individual use cases.
#1 Increase lead conversion with a Predictive Lead Scoring model
Use the lead scoring model to automatically score and rank leads based on their likelihood to convert, and ensure that sales teams focus their efforts on the most promising leads, resulting in increased conversion rates.
Recommended Model Builder setup
To get started with building a predictive lead scoring model using Data Cloud’s Model Builder, you’ll need some key inputs for historical data:
- Gather information on lead demographics, their job title, experience, role in the deal cycle (buyer, user, decision-maker), geographic details and so on.
- Add the Account’s firmographic data on Industry, No of employees, Revenue ranges, etc. and its geographic details as well (applicable for only B2B)
- Compile the interaction history across channels, such as campaign opens, CTRs; website visits, session duration, content downloads
- If applicable, add the Product usage data such as recency, frequency as DAU or MAU, specific features usage and so on.
- If you are able to source, get the Social Media data like interactions on your content and the Third-party intent data
With this data in hand, you can set up a binary classification model aimed at maximizing the likelihood that a lead converts to a customer. This approach will help you prioritize high-potential leads and focus your sales efforts where they matter most.
A prep step before you build the model
If you haven’t yet prepared your training dataset, start by compiling data that includes the available attributes along with the target variable (or goal) into a single DMO. Illustrating an example for lead scoring, create a representative training DMO containing lead attributes, account attributes, engagement metrics, and the target variable (or goal). This is what we will use for the model building.
In this case, the target variable is IsLeadConvertedToCustomer (highlighted in purple along with a 🎯 icon). This field is tagged is Yes, if the lead successfully converted to a customer, encompassing both won and lost deals, as well as existing and churned customers. For disqualified or dropped leads, the field is marked No.
✅ Model type: Binary Classification
🎯 Goal: Maximize IsLeadConvertedToCustomer = Yes
How to use the model outputs in business
Once you’ve built the models and generated predictions on live data, the next step is to incorporate them into your workflow. Here are some key actions to take once your predictions are ready:
- Prioritize top priority leads based on the prediction
- Route these hot leads to sales teams
- Personalize outreach for them
- Optimize resource allocation to these high potential leads
Outcomes driven with this model
By implementing a robust lead scoring model with the Model builder, businesses can achieve below outcomes:
⬆️ Increased lead conversion rate
⬆️ Increased sales efficiency
⬆️ Accelerated sales cycle
⬆️ Aligned marketing and sales teams
#2 Improve win probability with a Predicted likelihood of deal wins or closure model
Use the model to evaluate and prioritize sales opportunities in the pipeline, based on the likelihood of closing, enabling sales reps to focus on the highest potential and accelerate sales cycle. Avoid missed opportunities and focus on the right ones.
Recommended Model Builder Setup
For this model, you need the historical data on deals or opportunities foremost:
- Gather opportunity details like deal size, amount, discount offered, product details, and other attributes which influence win rates.
- Create some variables on historical deal win/loss data
- Gather information on lead demographics, their job title, experience, role in the deal cycle (buyer, user, decision-maker), geographic details
- Add Account firmographic data on Industry, No of employees, Revenue ranges, etc. (applicable for only B2B)
- Collect customer interaction history, such as meetings, calls, emails and create metric based variables
- Include variables on the stage in the deal cycle, time spent in each stage, etc.
- Factor in product interest and competitor for enhanced accuracy.
With this data, you can set up a binary classification model aimed at maximizing the likelihood that a deal closes. This approach will help you prioritize high-potential opportunities and focus your sales efforts where they matter most.
✅ Model type: Binary Classification
🎯 Goal: Maximize IsDealClosed = Yes
How to use the model outputs in business
Here are some key actions you can take once your model is up and running:
- Prioritize highest potential deals vs at risk deals
- Tailor sales strategies based on the scores
- Allocate resources effectively
- Monitor and adjust tactics
Outcomes driven with this model
By implementing this model with the Model builder, businesses can achieve below outcomes:
⬆️ Increased deal win rates.
⬆️ Accelerated sales cycle
⬆️ More accurate sales forecasting
⬆️ Improved sales productivity
#3 Decrease time to close with a Predictive time-to-close model
Use the predictive time-to-close model to evaluate and prioritize sales opportunities in the pipeline based on their predicted time to close. This enables sales managers and teams to streamline their forecasting processes and focus on deals that can be closed quickly, resulting in decreased time to close deals.
Recommended Model Builder Setup
To predict time to close, you’ll need data mainly on the deals and the stages:
- Gather opportunity details like deal size, amount, discount offered, product details, etc
- Create some variables on historical deal win/loss data
- Include variables on the stage in the deal cycle, time spent in each stage, etc. This is important to understand the velocity in each stage.
- Gather information on lead demographics, their job title, experience, role in the deal cycle (buyer, user, decision-maker), geographic details
- Add Account firmographic data on Industry, No of employees, Revenue ranges, etc. (applicable for only B2B)
- Collect customer interaction history, such as meetings, calls, emails and create metric based variables
- Factor in product interest and competitor for enhanced accuracy.
With this data in hand, you can set up a regression model aimed at minimizing the number of days to close or the close date. This approach will help you identify and prioritize opportunities that can be closed quickly, streamlining your sales process.
✅ Model type: Regression
🎯 Goal: Minimize DaystoClose
How to use the model outputs in business
Here are some key actions you can take once your model is up and running:
- Prioritize quick wins vs at-risk deals
- Streamline processes
- Optimize follow-ups
- Allocate resources
Outcomes driven with this model
By implementing this model with the Model builder, businesses can achieve below outcomes:
⬆️ Accelerated sales cycle
⬆️ Enhanced sales forecasting
⬆️ Increased deal win rates.
⬆️ Improved sales productivity
#4 Predict expected revenue with a Predictive revenue model
Use the revenue prediction model to prioritize sales opportunities in the pipeline based on their potential revenue, enabling sales reps to focus on deals with the highest financial impact, resulting in increased revenue and more accurate revenue forecasting
Recommended Model Builder Setup
To predict revenue, you’ll need data mainly on the deals
- Gather opportunity details like deal size, amount, discount offered, product details, etc
- Create some variables on historical deal win/loss data
- Include variables on the stage in the deal cycle, time spent in each stage, etc. This is important to understand the velocity in each stage.
- Gather information on lead demographics, their job title, experience, role in the deal cycle (buyer, user, decision-maker), geographic details
- Add Account firmographic data on Industry, No of employees, Revenue ranges, etc. (applicable for only B2B)
- Collect customer interaction history, such as meetings, calls, emails and create metric based variables
- Factor in product interest and competitor for enhanced accuracy.
With this data in hand, you can set up a regression model aimed at maximizing deal revenue. This approach will help you identify and prioritize high-revenue opportunities, enhancing your sales strategy.
✅ Model type: Regression
🎯 Goal: Maximize DealRevenue
How to use the model outputs in business
Here are some key actions you can take once your model is up and running:
- Prioritize High-value opportunities
- Improve revenue forecasting
- Optimize sales strategies
- Allocate resources
Outcomes driven with this model
By implementing this model with the Model builder, businesses can achieve below outcomes:
⬆️ Enhanced sales forecasting
⬆️ Increased deal win rates.
⬆️ Accelerated sales cycle
⬆️ Improved sales productivity
#5 Decrease churn probability with a Churn Prediction Model
Use the churn prediction model to identify existing customers who are at risk of leaving. This allows sales teams to take proactive measures to retain them through targeted offers or personalized engagement, resulting in improved customer retention rates and reduced churn.
Recommended Model Builder Setup
To get started with building a churn prediction model using Data Cloud’s Model Builder, you’ll need some key data inputs:
- Gather information on lead demographics, their job title, experience, role in the deal cycle (buyer, user, decision-maker), geographic details
- Gather account firmographics like age, Industry, Segment, etc and demographics like location (applicable for only B2B)
- Compile Contact or Account’s recent interaction history across channels (support tickets / cases, emails, calls, meetings)
- Compile purchase history contract renewal, and customer satisfaction scores related variables
- Include recent product usage data based variables
- Add deal summary on number of open/closed deals, amounts, discounts given
With this data in hand, you can set up a regression model aimed at minimizing customer churn. This approach will help you identify at-risk customers and implement effective retention strategies.
✅ Model type: Binary Classification
🎯 Goal: Minimize IsCustomerChurned = Yes
How to use the model outputs in business
Here are some key actions you can take once your model is up and running:
- Identify at-risk customers
- Personalize retention strategies
- Offer targeted incentives
- Monitor and adjust engagement tactics
Outcomes driven with this model
By implementing this model, businesses can achieve below outcomes:
⬆️ Improved customer retention rates
⬆️ Enhanced revenue stability
⬆️ Increased customer lifetime value
⬆️ Reduced churn-related operational costs
#6 Improve sales rep performance with Sales Rep Performance Model
Use the sales rep performance model to analyze the performance of individual sales reps, identifying strengths, areas for improvement, and best practices that can be shared across the team, resulting in enhanced overall team performance and productivity
Recommended Model Builder Setup
To get started with building a sales rep performance model, you need data on the sales reps and their past performances:
- Gather sales rep demographics, experience, role, and other attributes
- Compile deal details, including past deal closure rates and average deal size as a measure of what they have handled in the past.
- Include data on the bonuses they earned, targets met.
- Collect data on their customer engagement data to know the efforts they put in.
- Include customer feedback
With this data in hand, you can set up a binary classification model aimed at maximizing sales rep performance. This approach will help you identify top performers and areas where reps can improve.
✅ Model type: Binary Classification
🎯 Goal: Maximize SalesRepPerformanceRating = High
How to use the model outputs in business
Here are some key actions you can take once your model is up and running:
- Identify Top and bottom performers
- Highlight areas for improvement
- Tailor training and development programs
Outcomes driven with this model
By implementing this model, businesses can achieve below outcomes:
⬆️ Improved individual sales productivity
⬆️ Accelerated Revenue Growth
⬆️ Enhanced quota attainment rates
⬆️ Improved Resource Efficiency
#7 Increase contract renewals with Contract renewal prediction model
Use the contract renewal prediction model to predict which contracts are most likely to be renewed and which are at risk. This enables targeted renewal strategies resulting in increased contract renewal rates.
Recommended Model Builder Setup
To get started with building a contract renewal prediction model, you’ll need contract details first and foremost.
- Compile contact / account’s contract history like previous contract duration, renewal dates, amounts, discounts,
- Gather information on lead demographics, their job title, experience, role in the deal cycle (buyer, user, decision-maker), geographic details
- Gather account firmographics like age, Industry, Segment, etc, demographics like location, etc (applicable for only B2B)
- Compile Contact or Account’s recent interaction history across channels (support tickets / cases, emails, calls, meetings)
- Compile revenue metrics like LTV, recent feedback / satisfaction scores
- Also recent product usage data based variables
With this data in hand, you can set up a binary classification model aimed at minimizing contract renewal risk. This approach will help you identify contracts likely to be renewed and those at risk of non-renewal.
✅ Model type: Binary Classification
🎯 Goal: Maximize IsContractRenewed = True
How to use the model outputs in business
Here are some key actions you can take once your model is up and running:
- Identify Contracts at Risk: Focus on contracts likely to be renewed and those at risk of non-renewal.
- Personalize Renewal Strategies: Tailor your renewal efforts to address specific customer needs.
- Offer Targeted Incentives: Provide personalized offers to encourage contract renewals.
- Monitor and Adjust Engagement Tactics: Continuously refine your strategies based on customer responses.
Outcomes driven with this model
By implementing this model, businesses can achieve below outcomes:
⬆️ Improved customer retention rates
⬆️ Enhanced revenue stability
⬆️ Increased customer lifetime value
⬆️ Improved operational efficiency
#8 Increase quote acceptance with Quote acceptance prediction model
Use the quote acceptance prediction to streamline the quote-to-cash process by predicting bottlenecks and automating routine tasks, reducing the time it takes to close deals, resulting in a more efficient and effective sales process
To get started with building a quote acceptance prediction model using Data Cloud’s Model Builder, you’ll need some key data inputs:
- Gather quote details like amount, payment terms, discount offered, product details, revision history, etc
- Include variables on the stage in the deal cycle, time spent in each stage, etc. This is important to understand the velocity in each stage.
- Gather information on lead demographics, their job title, experience, role in the deal cycle (buyer, user, decision-maker), geographic details
- Add Account firmographic data on Industry, No of employees, Revenue ranges, etc. (applicable for only B2B)
- Collect customer interaction history, such as meetings, calls, emails and create metric based variables
- Factor in product interest and competitor for enhanced accuracy.
With this data in hand, you can set up a binary classification model aimed at maximizing quote acceptance. This approach will help you identify process bottlenecks and automate routine tasks.
✅ Model type: Binary Classification
🎯 Goal: Maximize QuoteAccepted = True
How to use the model outputs in business
Here are some key actions you can take once your model is up and running:
- Identify Process Bottlenecks: Pinpoint areas in the quote-to-cash process that slow down deal closure.
- Automate Routine Tasks: Streamline repetitive tasks to free up time for more strategic activities.
Outcomes driven with this model
By implementing this model, businesses can achieve below outcomes:
⬆️ Higher quote acceptance rates
⬆️ Accelerated sales cycle
⬆️ Accurate sales forecasting
⬆️ Improved productivity
#9 Increase repeat business with Repeat Purchase Prediction Model
Use the repeat purchase prediction model to identify customers who are likely to make repeat purchases. This enables sales teams to focus on these customers with targeted offers and personalized engagement, resulting in increased repeat business and customer loyalty.
Recommended Model Builder Setup
To get started with building a repeat purchase prediction model using Data Cloud’s Model Builder, you’ll need some key data inputs:
- Gather purchase details like average order value (AOV), frequency of purchases, time since the last purchase, product categories, and discounts offered for contact / account as applicable
- Include variables on contact / customer revenue metrics like lifetime value (LTV), total orders placed.
- Gather information on lead demographics, their job title, experience, role in the deal cycle (buyer, user, decision-maker), geographic details
- Add Account firmographic data such as industry, geographic location, number of employees, and customer age (applicable for only B2B)
- Compile customer interaction history, such as meetings, calls, emails, etc
- Incorporate customer feedback metrics like NPS and competitor activity data for enhanced predictive accuracy.
With this data in hand, you can set up a binary classification model aimed at maximizing repeat purchases. This approach will help you identify high-value repeat customers and tailor your engagement strategies.
✅ Model type: Binary Classification
🎯 Goal: Maximize RepeatPurchaseCustomer = True
How to use the model outputs in business
Here are some key actions you can take once your model is up and running:
- Identify High-value repeat customers
- Target personalized campaigns
- Allocate resources for re-engagement
Outcomes driven with this model
By implementing this model, businesses can achieve below outcomes:
⬆️ Increased repeat purchase rates
⬆️ Enhanced customer loyalty and retention
⬆️ Increased customer lifetime value
⬆️ Improved personalized marketing strategies
#10 Increase lifetime value
Use the customer lifetime value prediction model to predict future value of high-value customers, enabling sales and marketing teams to strategize their focus on this segment, resulting in increased revenue and long-term customer relationships
Recommended Model Builder Setup
To get started with the model, you’ll need some key data inputs on the customers.
- Gather purchase history details like average order value (AOV), purchase frequency, time since the last purchase, and product categories for contact / account.
- Include variables on revenue metrics such as past lifetime value (LTV), total orders, gross revenue, and historical spending patterns for contact / account
- Gather information on lead demographics, their job title, experience, role in the deal cycle (buyer, user, decision-maker), geographic details
- Add account firmographic data like industry, geographic location, customer age, and demographic attributes. (applicable for only B2B)
- Compile customer interaction history, including meetings, calls, emails, and engagement levels.
- Factor in customer feedback metrics like NPS, churn likelihood, and competitor influences to refine predictions.
With this data in hand, you can set up a regression model aimed at accurately predicting customer lifetime value (CLV). This approach will help you identify high-value customers and strategize accordingly.
✅ Model type: Regression
🎯 Goal: Maximize CustomerCLV
How to use the model outputs in business
Here are some key actions you can take once your model is up and running:
- Identify high-value customers and personalize marketing and sales strategies
- Enhance customer engagement
- Optimize cross-selling and upselling opportunities
Outcomes driven with this model
By implementing this model, businesses can achieve below outcomes:
⬆️ Accurate customer lifetime value predictions
⬆️ Focused retention and acquisition strategies
⬆️ Improved prioritization of high-value customers
⬆️ Higher ROI on marketing and sales efforts
Conclusion
In this article, we explored practical use cases that can be built with Model Builder to turn sales challenges into opportunities, enabling your team to make more informed, data-driven decisions. By exploring these use cases and building models tailored to your needs, you can unlock faster sales cycles, higher win rates and consistently hit your sales targets with ease.
Onward to selling more, smarter, faster!!!
Special thanks to Bobby Brill for his review
For more resources on Model Builder
- Blogs:
- Salesforce Help articles:
- YouTube: Model Builder: Ask Me Anything with Salesforce Developers
- Podcast: What are the key features of Salesforce’s Model Builder?
Written by
