top of page

How to Set Up Salesforce Einstein Lead and Opportunity Scoring


Flat-style illustration showing a business team analyzing CRM data with the help of an AI assistant, symbolizing the setup and impact of Salesforce Einstein Lead and Opportunity Scoring.

Ever feel like your sales team is playing darts blindfolded when it comes to prioritizing leads and deals?


You're not alone. Most sales teams struggle with the same question: Which leads should we call first? Which opportunities deserve our best reps' attention?


Salesforce Einstein Lead and Opportunity Scoring takes the guesswork out of this equation. Instead of relying on gut feelings or random follow-ups, Einstein uses your actual CRM data to predict which leads will convert and which deals will close.


Here's the thing though – setting it up right isn't just about clicking a few buttons. There's a method to making Einstein work for your specific business. Let's walk through exactly how to implement Salesforce Einstein scoring so your team can focus on what matters most.


What Exactly Is Einstein Lead and Opportunity Scoring?

Think of Einstein as your data-driven sales advisor.


Einstein Lead Scoring analyzes your historical lead data to predict conversion likelihood. It looks at patterns like industry, company size, lead source, and behavior to assign each lead a score from 1-99.


Einstein Opportunity Scoring does the same thing but for deals in your pipeline. It examines factors like deal size, stage duration, activity levels, and past win/loss patterns to score each opportunity.


The beauty? These scores update automatically as new data comes in. Your team gets real-time intelligence without manual number-crunching.


Before You Start: The Prerequisites Checklist


Don't jump straight into setup. Einstein needs the right foundation to work properly.


Edition Requirements

Einstein scoring comes with these Salesforce editions:


  • Performance and Unlimited: Included free

  • Enterprise: Available as $50/user/month add-on

  • Lower editions: Not available (time to upgrade)


Data Volume Requirements

Einstein is hungry for data. Here's what you need:


For Lead Scoring:

  • At least 1,000 leads created in the last 6 months

  • Minimum 120 of those leads converted to Account + Contact

  • Clean, consistent data across key fields


For Opportunity Scoring:

  • 200 closed-won opportunities in the last 24 months

  • 200 closed-lost opportunities in the same timeframe

  • Each deal should have at least 2 days of history


Don't have enough data yet? Einstein will use a global model trained on anonymized data from other Salesforce customers. Once your data volume grows, it automatically switches to your org-specific model.


Data Quality Matters


Garbage in, garbage out. Before enabling Einstein:


  • Clean up duplicate records

  • Fill in missing values for key fields (Company, Amount, Industry)

  • Remove placeholder or test data

  • Standardize formats (especially for picklist values)


Step-by-Step: Setting Up Einstein Lead Scoring


Ready to get your hands dirty? Here's the exact process.


Step 1: Enable the Feature


Navigate to Setup in your Salesforce org. Type "Einstein Lead Scoring" in the Quick Find box. Click Enable Einstein Lead Scoring.


You'll see two options:

  • Default Settings: Scores all leads together using all available fields

  • Custom Settings: Lets you define segments and field selections

For most orgs, custom settings give better results.


Step 2: Choose Your Conversion Milestone


Decide what "conversion" means for your business:

  • Lead to Account + Contact: For companies focused on generating qualified contacts

  • Lead to Opportunity: For sales teams that measure success by pipeline creation

This choice affects how Einstein defines success, so think carefully about your sales process.


Step 3: Set Up Lead Segments (Optional but Recommended)


If your leads come from different sources or target different markets, segmentation helps. Einstein can create up to 35 segments based on criteria like:


  • Geographic region (Domestic vs. International)

  • Lead source (Website vs. Trade Show vs. Referral)

  • Product line or industry focus


Each segment gets its own scoring model. This prevents Einstein from comparing apples to oranges.


Step 4: Select Fields for the Model


By default, Einstein includes all lead fields. But not every field is predictive.


Include these types of fields:

  • Demographic data (Company, Industry, Title)

  • Behavioral data (Email opens, website visits)

  • Source information (Lead Source, Campaign)


Exclude these types of fields:

  • System-generated IDs

  • Internal notes or comments

  • Fields with mostly blank values

  • Circular fields (like "Reason for Loss")


When in doubt, include the field. You can always refine later.


Step 5: Review and Launch


Einstein shows you a summary of your choices. Double-check your segments and field selections, then click Score Leads.


The initial model building takes up to 48 hours. You can monitor progress on the Einstein Lead Scoring setup page.


Step-by-Step: Setting Up Einstein Opportunity Scoring


Opportunity scoring follows a similar but slightly different process.


Step 1: Access the Setup

In Lightning Setup, search for "Assisted Setup" and select it. Look for Opportunity Scoring in the Einstein Sales section.


Step 2: Define Your Data Scope

Choose whether Einstein should analyze:

  • All opportunities: Simple but might include irrelevant deals

  • Filtered opportunities: Exclude test deals, partnerships, or other non-standard opportunities

If you filter, define clear criteria. For example, you might exclude opportunities with $0 amount or specific record types.


Step 3: Field Selection


Like lead scoring, you can include or exclude custom opportunity fields. Focus on fields that actually influence deal outcomes:


Good fields to include:

  • Deal amount and probability

  • Account information (size, industry)

  • Sales activity levels

  • Competitive information

  • Product or solution details


Fields to exclude:

  • Internal tracking fields

  • Created dates and system stamps

  • Fields with inconsistent data


Step 4: Launch and Monitor


Review your settings and click Start. Again, model building takes up to 48 hours.

Once complete, Einstein begins scoring all opportunities based on your historical win/loss patterns.


Making Scores Visible to Your Team


Scores don't help if your team can't see them. Here's how to surface Einstein insights where your reps work.


Add Scores to Record Pages


For Lead Scoring:

  • Edit your Lead page layouts in App Builder

  • Add the Einstein Lead Scoring component

  • Position it prominently (top of the page works well)


For Opportunity Scoring:

  • Salesforce automatically adds the Opportunity Score field to default layouts

  • For custom layouts, manually add the Opportunity Score field


Update List Views

This is where the magic happens. Sales reps live in list views, so make scores prominent:

  • Add Lead Score and Opportunity Score columns to all relevant list views

  • Create new views filtered by score ranges (e.g., "High-Score Leads 80+")

  • Set up default sorting by score (highest first)


Build Score-Based Reports and Dashboards


Create reports that show:

  • Conversion rates by score band

  • Pipeline velocity for high-score vs. low-score deals

  • Rep performance using scored leads/opportunities

These reports help validate Einstein's accuracy and demonstrate ROI.


Best Practices for Long-Term Success


Einstein isn't a set-it-and-forget-it tool. Follow these practices to maximize value:


Monitor Model Performance

Check Einstein's accuracy regularly:

  • Review conversion rates by score range

  • Validate that high-score records actually perform better

  • Look for unexpected patterns or anomalies

Use Salesforce's built-in Einstein analytics dashboards for easy monitoring.


Refine Field Selection Over Time

As your business evolves, so should your scoring models:

  • Add new fields that become predictive

  • Remove fields that lose relevance

  • Adjust segments based on changing market conditions

Einstein retrains models monthly, so changes appear relatively quickly.


Drive User Adoption

The best scoring model is worthless if your team ignores it:

  • Train reps on what scores mean and how to use them

  • Incorporate scores into sales processes and workflows

  • Create incentives for acting on high-score records

  • Share success stories where Einstein helped close deals


Integrate with Other Tools

Einstein scores work best when integrated into your broader sales tech stack:

  • Use scores in email marketing campaigns

  • Set up automated workflows based on score thresholds

  • Include scores in forecasting and pipeline reviews


Common Implementation Mistakes to Avoid


We've seen these pitfalls trip up many organizations:


Mistake 1: Rushing the Setup Take time to clean your data and think through field selections. A hasty setup leads to poor scores.


Mistake 2: Ignoring Data Quality Einstein amplifies whatever patterns exist in your data. If your data is messy, your scores will be too.


Mistake 3: Over-Segmenting Too many segments can spread your data too thin. Each segment needs enough historical records to build a robust model.


Mistake 4: Not Training Users Scores are meaningless without context. Invest in proper user training and change management.


Mistake 5: Expecting Immediate Perfection Einstein gets smarter over time. Give it a few months to learn your patterns before making major adjustments.


Measuring ROI and Success


How do you know if Einstein scoring is working? Track these metrics:


Lead Scoring Success Metrics:

  • Increase in lead conversion rates

  • Faster response times to high-score leads

  • More qualified opportunities in pipeline

  • Better lead source attribution


Opportunity Scoring Success Metrics:

  • Improved win rates for high-score deals

  • Shorter sales cycles

  • More accurate forecasting

  • Better resource allocation


Most organizations see measurable improvements within 3-6 months of proper implementation.


Your Next Steps


Salesforce Einstein Lead and Opportunity Scoring can transform how your sales team prioritizes its work. But success depends on thoughtful implementation, not just feature activation.


Start with clean data, choose your settings carefully, and commit to ongoing optimization. When done right, Einstein scoring becomes one of your most valuable sales tools.


Need help getting Einstein scoring right the first time? At Implementology, we've guided dozens of companies through successful Einstein implementations. We know the pitfalls to avoid and the shortcuts that actually work.


Ready to let data drive your sales prioritization? Let's talk about how Salesforce Einstein Lead and Opportunity Scoring can work for your specific business needs.


 
 
 

Comments


bottom of page