Einstein Lead Scoring
Key Takeaway
Einstein Lead Scoring ranks leads by how closely they match the leads your company converted in the past. When you use lead segments, each segment must meet the data requirements, because segments turn off the global model.
Einstein Lead Scoring uses machine learning to give each lead a score. The score shows how much the lead has in common with leads your company converted in the past. Sales reps use the Lead Score to decide which leads to work first. Reps can also see which field values pushed each score up or down. Because Einstein finds your conversion patterns itself, this approach is simpler, faster, and more accurate than traditional rules-based lead scoring.
How Einstein learns your conversion patterns
Einstein studies your past converted leads, including their custom fields and activity data, to learn your conversion patterns. It then ranks current leads by how closely they match those converted leads. A higher score means a closer match, so reps who focus on high-scoring leads are likely to convert more of them to opportunities.
Einstein builds its model from closed leads over the past 3 years. If your conversions have seasonal patterns, Einstein picks those up too.
Einstein groups similar job titles into a Job Rank and a Department. For example, a lead with the title CEO gets the C-level Job Rank. Grouping many titles into a short list makes patterns easier to find.
Einstein can also use external data that you add to leads through custom lead fields.