Winning by Design * Engineering Spec · WS4

SPICED Pulse: The Per-Rep Action Agent

The dashboard diagnoses. This agent acts. It reads one rep's SPICED gaps from the data model, pulls the matching manager playbook section, hands the rep a ready-to-run prompt plus a weekly reinforce and improve pair, and follows up on the Tuesday 1:1 cadence. It is the arc of the flywheel that turns a diagnosis into rep behavior change, and it is how we move the lagging number. This document is the spec an engineer builds against: inputs, outputs, rules, and the closed conditions.

Reads SPICED gaps Reads the manager playbook Reads recent Grain calls Delivered at the Tuesday 1:1 Manager can override
Section 1 · Inputs

What the agent reads.

Three sources, all already produced by the WS1 pipeline. The agent adds no new data collection. It consumes the scored data model, treats the manager playbook as a queryable knowledge base, and reads the rep's most recent Grain calls for fresh evidence.

InputSourceShape the agent reads
SPICED gaps The scored data model (per rep, per deal) Per dimension: Completion (yes or no) and Quality (1 to 5), a state (weak, solid, strong), and a gap boolean. The revenue-linked headline is Completion; the coaching signal is Quality. Impact and Critical Event are the completion binaries the scorer detects.
Manager playbook WbD method library, indexed as a knowledge base Each play keyed to a SPICED dimension and framed in REKS, retrievable by gap. The agent queries it by the rep's specific gap, so a weak Impact returns the Impact-quantification play, not the whole library.
Recent Grain calls Grain, segmented and speaker-resolved, joined on the HubSpot deal id The last few genuine sales-cycle calls for this rep, with the verbatim customer line, the rep's actual response, and the deep link to the call moment. This is the evidence the agent quotes back, never invented.
Grounding contract

Every gap the agent names traces to a scored dimension in the data model, and every quote it shows traces to a verbatim Grain segment with a deep link. The agent produces no claim that is not backed by one of these three inputs. If the evidence is not present, the agent stays silent on that dimension rather than manufacturing a coaching moment.

Section 2 · Outputs

What the agent hands the rep.

Two outputs per cycle. First, for each open gap, the matching playbook section plus a ready-to-run prompt the rep pastes into their own AI. Second, a weekly reinforce and improve pair: one win to repeat, one growth moment to work, each as a four-part card grounded in the rep's own calls.

Output A: gap to play to prompt

For each gap, the agent emits a triple. The gap names the dimension, the play is the retrieved method, and the prompt is executable by the rep on their next real deal.

1
The gap. Named from the data model. Example: Impact is incomplete on the CBRE deal. Quality is 2 of 5. The rep surfaced pain but never quantified the business consequence.
2
The play. The Impact-quantification method retrieved from the playbook, framed in REKS, with the one behavior to change on the next call.
3
The prompt. A ready-to-run prompt the rep pastes into their own AI to draft the impact question for this specific account before the next call.
Ready-to-run prompt handed to the repYou are my SPICED discovery coach for the CBRE deal. The customer told me their only payment paths are physical check or ACH, and that a lost payment "comes back on me." I never pushed that pain to a quantified business consequence. Draft three impact questions I can ask on our next call that turn that risk into a number the buyer owns: what a slipped payment on a 25,000 dollar buyout costs in lost revenue or rework, and how often the near-miss happens. Keep each question short and buyer-owned. Then give me the one follow-up that pins a dated critical event to it.

Output B: the weekly reinforce and improve pair

One win, one growth moment. Development framing, always paired, so the rep reads it as coaching and not a grade. Each is a four-part card: the customer line, the rep's actual response, the verdict and why, and the next action.

Reinforce · the win to repeat
Customer said
"If the money was lost, it's going to come back on me."
Rep said
"Who else feels that risk when it happens, and what does your finance lead do about it today?"
Verdict and why
Strong Engagement move. The rep widened the deal past a single champion and surfaced the finance stakeholder, which raises Decision Process quality.
Next action
Repeat this stakeholder-widening question on the next two discovery calls. Keep doing what worked.
Improve · the growth moment
Customer said
"The only way we can process money is physical check or ACH. We don't do credit card payments."
Rep said
"I'm so glad you said that. We have a demo just on Momentus payments."
Verdict and why
Impact left on the table. The rep jumped to a feature before quantifying the consequence. The buyer named a risk they own and it went unmeasured.
Next action
Run the ready-to-run prompt above, then ask one quantified impact question on the next CBRE call.
Why exactly one of each

One win and one growth moment per week is deliverable inside a single 1:1 and is small enough that the rep can actually change one behavior before the next cycle. More than one growth moment at a time is where coaching turns into a grade and the rep stops trusting it. The pairing is the mechanism: the win earns the right to name the gap.

Section 3 · Cadence

Weekly, anchored to the Tuesday 1:1.

The agent runs on the manager's existing ritual, so it needs no new habit to survive. The manager delivers the pair and the play inside the Tuesday 1:1 they already hold, and the manager can override before anything reaches the rep. The rep runs the prompt on their next real call. The next scored call closes the loop.

Monday
Agent assembles
Reads the week's scored calls, retrieves the plays for open gaps, drafts the reinforce and improve pair, and stages it to the manager.
Monday to Tuesday AM
Manager reviews
Manager reads the staged pair, edits or overrides any card, and suppresses false positives before the rep ever sees them.
Tuesday 1:1
Manager delivers
The pair and the play are delivered in the existing 1:1. The rep leaves with one win to repeat and one behavior to change.
Tuesday to next call
Rep acts
The rep runs the ready-to-run prompt, prepares the impact question, and asks it on the next real deal call.
Next scored call
Agent re-scores
The next call is scored, the dimension is re-checked, and the week-over-week delta feeds Monday's assembly.
Why anchored, and not a new ritual

A coaching tool that needs a new weekly habit dies. This agent attaches to the Monday pipeline review and the Tuesday 1:1 the manager already runs. The one required manager act is small: review the pair, deliver it. Everything upstream is assembled for them, and everything downstream is measured for them.

Section 4 · Escalation, false-positive suppression, manager override

The rules that keep the rep trusting it.

The first wrong call the rep catches is the moment they stop trusting the agent. These three rule sets keep the human manager in the loop and give the rep a visible path to correct the agent.

Rule set A

False-positive suppression

  • 1Evidence gate. No card ships without a verbatim Grain line and a scored dimension behind it. Missing evidence suppresses the card.
  • 2Call-type gate. Only genuine sales-cycle calls feed the agent. Customer-success, status, and renewal-ops calls are routed out before scoring.
  • 3Confidence floor. A dimension scored at low model confidence is held for manager review, never auto-delivered to the rep.
  • 4Repeat-gap only. A gap is delivered as a growth moment once it appears across two calls, so a single off day does not become a coaching card.
Rule set B

Manager override

  • 1Pre-delivery gate. Every pair is staged to the manager before the rep sees it. Nothing reaches the rep unreviewed.
  • 2Edit or kill. The manager can rewrite a verdict, swap the play, or discard the card entirely.
  • 3Mute a dimension. The manager can mute a dimension for a rep for a defined window when context makes it a non-issue.
  • 4Override is signal. Every override is logged and feeds back to tune the scorer, so the agent gets fewer false positives over time.
Rule set C

Escalation

  • 1Stalled gap. A gap open for three consecutive weeks with no closing delta escalates from a rep card to a manager coaching flag.
  • 2Deal risk. A weak-SPICED gap on a large open deal escalates to the Deal Pulse view as a forecast risk, visible to leadership.
  • 3Rep dispute. A rep can flag a card as wrong in one click. The flag routes to the manager, pauses that dimension, and is logged.
  • 4No behavior change. When re-scoring shows no movement after the play was delivered, the agent escalates to the manager for a live coaching session rather than resending the same card.
The trust rule, stated plainly

The rep always has a one-click way to say the agent got it wrong, and that flag always reaches a human manager who can override. The rep owns the relationship with the agent, and the manager owns the final say. The agent never has the last word on a rep.

Section 5 · How it knows a gap closed

Re-score the next call, track the delta.

The agent does not take the rep's word or the manager's word that a gap closed. It re-scores the next genuine sales-cycle call on the same dimension and measures the week-over-week delta. The closed condition is explicit and measured, so the loop has a real exit.

The closed condition

  • ICompletion flips to yes. For Impact and Critical Event, the dimension moves from incomplete to complete: a quantified impact tied to a buyer-owned metric, or a real dated critical event with a consequence.
  • QQuality clears the bar. The dimension is re-scored above 3 of 5 on the next call, up from its opening score.
  • The delta holds. The improvement appears on the next scored call and holds on the one after, so a single good call does not falsely close the gap.

What happens on close

  • 1The growth moment is retired and becomes a reinforce card: the win to repeat.
  • 2The week-over-week delta is written to the rep's trend, framed as growth.
  • 3The manager sees the before and after in the pipeline review and can clear it.
  • 4The agent selects the next open gap by revenue impact and starts the next triple.
Worked example, from the WS2 slice

CBRE, rep John Haire, Impact opens at 2.0 and incomplete. The agent hands the Impact-quantification play and the ready prompt. Exit condition: the next call shows an asked impact question and a quantified consequence, Impact re-scored above 3, Completion flipped to yes. The manager, Josh Klatz, sees the before and after in the pipeline review and clears it. The gap is closed by measurement, not by assertion.

Section 6 · The one leading metric it moves

Impact plus Critical Event completion, tied to the lagging number.

The agent moves one leading metric: the rate of Impact and Critical Event completion across the rep's open deals. That metric is chosen because it is the pair most reps miss and the pair most correlated to revenue. Tying the leading behavior change to the lagging revenue number is how the agent proves it earned its cost.

LayerThe metricWhy this one
Leading
the agent moves this
Rate of Impact and Critical Event completion across the rep's open sales-cycle deals, week over week. Impact, Critical Event, and Decision Process are the most common team gap. Completion is the revenue-linked binary the scorer already detects, so the agent moves a number the pipeline already measures.
Lagging
the business cares about this
Win rate and sales-cycle time, read on the same deals in HubSpot. These are what renewal and leadership buy on. The agent is unbuyable at renewal unless the leading behavior change is shown to move these.

The proof design that links them

1
Before and after per rep. Baseline each rep's completion rate, win rate, and cycle time before the agent starts, then track the same three after the agent is delivering weekly.
2
Holdout where the team is large enough. A set of reps runs the agent, a matched set does not, and the completion-rate and win-rate gap between them is the agent's measured lift.
3
Deal-level link. Deals where completion flipped to yes are tracked to their win or loss and cycle length, so the correlation the WS1 definition already claims is proven on our own data, not borrowed from the study.
The reference correlation the proof is measured against

The WS1 product definition anchors the claim: sellers at full SPICED completion average plus 10 percent win rate and plus 29 percent quota attainment, and completion of Impact and the Critical Event carries the largest ARR lift of the dimensions. The agent's job is to reproduce that link on WbD's own pipeline, so its value is provable at renewal, not asserted.

Section 7 · Guardrails

Development, held by the rep and their manager.

The agent is a development tool. The rep owns it. It is held between the rep and their manager, and it is measured as growth. Read the other way, it becomes surveillance, and a surveillance tool gets gamed and abandoned. These guardrails are load-bearing for adoption.

What the agent is

  • 1The rep owns it. The rep sees their own gaps first, as growth, with a trend that goes up.
  • 2Held by the pair. The relationship is rep and manager. Cards are delivered inside the 1:1, in context.
  • 3Reinforce first. Every cycle leads with a win to repeat before it names a growth moment.
  • 4Correctable. The rep can flag any card as wrong, and a human manager decides.

What the agent is not

  • 1Not a ranking feed to leadership. The per-rep cards are for the rep and their manager, not a leaderboard.
  • 2Not an autopilot. Nothing reaches the rep without manager review. The agent never has the last word.
  • 3Not a grade. The score is a coaching signal and a growth trend, never a performance verdict.
  • 4Not a data grab. Raw recordings never leave the environment. The agent reads derived, scored evidence only.
The one line that governs the build

Build every surface so a rep would choose to open it. The moment the agent reads as the manager watching, the rep games the calls and the leading metric stops meaning anything. Development framing is the design constraint, and it is what makes the lagging number real.