Revenue AI Strategy
Open the platform
All posts
L5 Maturityrevops-data 6 min read

Next-Best-Action Engine: How to Rank Sales Tasks by $ Value

Learn how to build an Expected Value (EV) engine that gives your AEs a ranked daily list of high-impact sales actions. Stop guessing, start closing.

Run the playbook

Next-best-action engine for AEs (L5)

For every open opp, the system recommends and ranks the next action by expected value. Managers manage to the plays, not to dials.

Why this matters

The average Account Executive (AE) spends 30% of their morning "triaging." They scan their inbox, scroll through Opportunity lists, and try to remember which prospect mentioned a competitor in a Fathom transcript from three days ago. This cognitive load is a silent killer of productivity. When a rep guesses what to do next, they gravitate toward "productive procrastination"—updating CRM fields or emailing the "nice" champion who isn't the decision-maker.

Implementing a Next-Best-Action (NBA) engine moves your team from a reactive state to a mathematical certainty. By ranking every possible sales activity by its Expected Value (EV), you ensure that your $150k OTE talent is focused on the $5,000 activity, not the $50 task. Companies moving to an EV-based motion typically see a 15-20% increase in win rates and a total elimination of "stalled" pipeline in the mid-funnel.

How it works: Identifying and Scoring Revenue Momentum

Step 1: Define the Action Universe

Revenue Ops must audit the last 12 months of CRM history to find the "turning points." Don't guess; look at the delta between closed-won and closed-lost deals. You are looking for 8-10 high-impact "NBA Actions."

  • Multi-threading: Adding a 3rd persona (e.g., a Finance lead) to a deal.
  • Executive Alignment: Getting your VP of Product on a call with their CTO.
  • Mutual Action Plan (MAP) Update: Sending a revised timeline via a tool like Momentum.io or a shared workspace.
  • Stalled Deal Re-engagement: Triggered when a deal hasn't had an external meeting in 10 days.

The Logic: A "Multi-thread" action is triggered if Stage = Discovery AND Contact Count < 3.

Step 2: Calculate Expected Value (EV) per Action

This is where the magic happens. You need to assign a dollar value to every action based on its "Win Rate Delta."

The Formula: (Probability Increase x Average Contract Value) = EV

  • Example: Your ACV is $50,000.
  • Historical data shows deals with "Executive Alignment" win at a 35% rate, while deals without it win at 20%.
  • The Delta is 15%.
  • Action EV: 0.15 * $50,000 = $7,500.

If a rep has 10 minutes before their next call, do they follow up on a $500 EV email or prepare for a $7,500 EV executive intro? The math makes the choice for them.

Step 3: Build the CRM Automation Engine

Using Salesforce Flow Builder or HubSpot Workflows, build the triggers to surface these actions.

When a trigger is met, the system creates a specialized "NBA Record" or a high-priority Task for the owner. Ensure these records include the EV_Score and a clear Description (e.g., "Schedule a call with the CFO to discuss ROI; mention the cost-savings found in the Granola notes from Tuesday").

Step 4: Deploy the Ranked Daily Queue UI

Stop using standard Task lists. Build a "Command Center" dashboard and set it as the default CRM home page for all AEs.

  • Sort Order: Descending by EV_Score.
  • The Rule: AEs must clear the top 5 actions before touching their inbox.
  • Integration: Use Clay to auto-enrich the NBA task with recent news about the prospect's company to make the outreach more relevant and faster to execute.

Tools you need

  • CRM: Salesforce (with Einstein NBA) or HubSpot.
  • Intelligence/Notes: Fathom or Granola to identify key signals (like a competitor mention) that trigger an NBA.
  • Deal Rooms: Momentum.io or Dock to track Mutual Action Plan updates.
  • Automation: Clay for automated research to support the "Action" (e.g., finding the CFO's LinkedIn for a multi-threading task).

KPIs to track

  • Expected Value per Rep Day: The total sum of EV completed by an AE. Target $20k+/day for enterprise reps.
  • Win Rate by NBA-Action: Are deals where the "Executive Alignment" action was completed actually winning at the predicted rate?
  • NBA Execution Velocity: The time it takes from an NBA being generated to it being marked "Complete."

Common pitfalls

  • The "Nag" Trap: Don't include CRM hygiene (e.g., "Fix close date"). That isn't an NBA; that’s a chore. Only include actions that move the needle.
  • Task Fatigue: Limit the engine to 10 active NBAs per rep. If they have 50 "Next Best Actions," they have zero.
  • Math Perfectionism: Don't wait for a data scientist to build a machine learning model. Start with static weights based on basic historical averages. 80% accuracy today is better than 99% accuracy in six months.

When to graduate to the next level

Once your team is consistently clearing their daily EV queue, you can move to L6: Dynamic EV. This involves using Claude Code or custom Python scripts to analyze your data warehouse and update weights in real-time based on current market volatility or seasonal trends. At that point, your NBA engine isn't just a list—it’s a live map of your company’s path to revenue.

nbaevforecasting

Ready to ship it? Open the playbook

Next-best-action engine for AEs (L5)

Step-by-step instructions, the tools to use, and the KPIs to watch — already wired into the Revenue AI Strategy workspace.

Open playbook