Deal Velocity Tracker: When to Push, Park or Kill
A three-signal framework that gives reps a concrete, repeatable verdict on any deal in pipeline: Push (with a named next action), Park (with re-entry conditions), or Kill (with CRM disposition instruction). Built around days-in-stage benchmarks, last meaningful buyer action, and stakeholder engagement trend.
How to use it
Run every open opportunity through the three signals once a week, or any time a deal feels stuck. Match your signal combination to the decision matrix to get your verdict, then follow the word-for-word next step for that outcome. Use the failure mode section as a self-check before you act.
What's inside
- Three measurable signals with clear definitions and what counts as evidence for each
- Separate threshold numbers for SMB and enterprise sales cycles
- One-page decision matrix mapping signal combinations to Push, Park or Kill verdicts
- Word-for-word scripts for each verdict outcome including re-engagement and kill messages
- CRM disposition instructions for parked and killed deals
- Re-entry conditions: the exact criteria a parked deal must meet before it moves again
- Named failure mode 1: parking deals that should be killed (with the tell-tale signs)
- Named failure mode 2: pushing deals that should be parked (with the tell-tale signs)
- A weekly pipeline review cadence note showing when to run the tracker
What This Does
Most pipeline reviews produce opinions. This framework produces verdicts. Run any deal through three measurable signals, match the combination to the matrix, and you get one of three outputs: Push, Park, or Kill, each with a specific next step you can execute today.
No signal on its own is enough. A deal that is overdue on days-in-stage might still have strong buyer engagement. A deal with recent buyer activity might still have a disengaging champion. The verdict comes from reading all three together.
The Three Signals
Signal 1: Days in Stage vs. Benchmark
How long has this deal sat in its current pipeline stage, measured against your cycle norms?
SMB benchmark (average deal under £25k, cycle under 60 days)
| Stage | Yellow (slowing) | Red (stalled) |
|---|---|---|
| Discovery | 8+ days | 15+ days |
| Proposal sent | 10+ days | 20+ days |
| Negotiation / legal | 12+ days | 25+ days |
Enterprise benchmark (average deal over £25k, cycle over 60 days)
| Stage | Yellow (slowing) | Red (stalled) |
|---|---|---|
| Discovery | 15+ days | 30+ days |
| Proposal sent | 20+ days | 40+ days |
| Negotiation / legal | 20+ days | 45+ days |
If you do not have your own benchmarks yet, use these. Once you have 20+ closed deals in your CRM, replace these numbers with your own median stage durations.
Score this signal: Green (within benchmark), Yellow (at or past yellow threshold), Red (at or past red threshold).
Signal 2: Last Meaningful Buyer Action
Not your last action. The buyer's last action. A meaningful buyer action is any of the following:
- Responded to a message or call with substance (not "got your voicemail")
- Attended a meeting or demo
- Shared internal information (budget, org chart, evaluation criteria, legal terms)
- Introduced you to another stakeholder
- Requested a document, proposal, or reference
Not meaningful: opening an email, a one-word acknowledgement, a calendar invitation they have not confirmed.
Score this signal: Green (meaningful buyer action within the last 7 days for SMB / 14 days for enterprise), Yellow (8-14 days for SMB / 15-28 days for enterprise), Red (15+ days for SMB / 29+ days for enterprise).
Signal 3: Stakeholder Engagement Trend
Is the number of engaged stakeholders growing, holding steady, or shrinking? Compare this week to the last three touchpoints.
- Green: New stakeholders added, or existing ones increasing meeting attendance and responsiveness.
- Yellow: Same stakeholders, engagement frequency dropping. Champion is slower to reply than before.
- Red: Champion has gone quiet, meetings cancelled without rescheduling, or you have been redirected to someone more junior.
The Decision Matrix
Match your three signal scores to the row below. Where you have a mix not listed exactly, default to the most negative signal combination above it.
| Signal 1 (Days) | Signal 2 (Buyer Action) | Signal 3 (Stakeholder Trend) | Verdict |
|---|---|---|---|
| Green | Green | Green | Push |
| Green | Green | Yellow | Push |
| Yellow | Green | Green | Push |
| Yellow | Yellow | Green | Park |
| Yellow | Green | Yellow | Park |
| Green | Yellow | Yellow | Park |
| Yellow | Yellow | Yellow | Park |
| Red | Any | Any | Kill |
| Any | Red | Red | Kill |
| Yellow | Red | Red | Kill |
Rule of thumb: Two Reds anywhere in the three signals is a Kill. A single Red with two Yellows is a Kill. Any combination where no signal is Red but at least two are Yellow is a Park.
The Verdicts: What to Do Next
Push
The deal has enough momentum to justify active effort this week.
Named next action: Identify the single next commitment you need from the buyer to advance the stage. One action, one owner, one deadline.
Script:
"Hi [Name], good to reconnect. Based on where we are, the logical next step is [specific action, e.g. a 30-minute call with your IT lead to walk through the integration requirements]. I have [date] at [time] or [date] at [time] available. Which works better for you?"
Do not push for two things at once. If you cannot name a single next buyer commitment, the deal is not as advanced as the stage suggests.
Park
The deal is not dead, but it cannot move on your effort alone. Something external needs to change before you invest more time.
Conditions to re-enter (all three must be true):
- The buyer initiates contact, or confirms a specific event has occurred (budget approved, project re-prioritised, incumbent contract expiring).
- They agree to a meeting with a set agenda and confirmed attendees within 14 days.
- You can name a specific decision date they have committed to.
CRM instruction: Move to a "Parked" or "On Hold" stage. Set a 30-day (SMB) or 60-day (enterprise) follow-up task. Add a note: "Parked [date]. Re-entry requires: [list your three conditions]." Do not count this deal in your forecast.
Script to send:
"Hi [Name], I want to be straight with you. We have been trying to find momentum here for a few weeks and I do not think the timing is right on your side at the moment. I am going to step back and stop filling your inbox. When [the specific trigger, e.g. the Q3 budget is confirmed / the new CTO is in post], I would genuinely like to pick this back up. Would that make sense?"
Kill
This deal is consuming forecast space and pipeline attention it does not deserve.
Exit criteria:
- No meaningful buyer action in 15+ days (SMB) or 29+ days (enterprise) despite two follow-up attempts across different channels.
- Champion has disengaged and you have no alternate contact.
- The buyer has explicitly or implicitly chosen another path (competitor selected, project shelved, internal build confirmed).
CRM instruction: Move to Closed Lost. Disposition field: choose the most accurate reason from your CRM picklist, then add a manual note: "Killed [date] per velocity tracker. Signal combination: [your three scores]. Last buyer contact: [date and what it was]." This data makes your benchmarks more accurate over time.
Script (final outreach before killing, sent once):
"Hi [Name], I have tried to reach you a few times over the past few weeks without luck. I am going to close this off on my side so I stop taking up space in your inbox. If the situation changes and [the problem you were solving] becomes a priority again, you know where I am. Wishing you well."
Send it. Then kill the deal. Do not wait for a reply before updating the CRM.
Failure Modes
Failure Mode 1: Parking deals that should be killed
The tell: You park a deal because the conversation was friendly, you like the contact, or the deal is large enough to distort your forecast if it disappears. The buyer has given you no meaningful signal of intent, only vague future interest ("definitely want to revisit this in Q3").
Why it matters: Parked deals that should be dead inflate pipeline, corrupt your close rate data, and give you false comfort in your number.
The fix: Apply the re-entry conditions strictly. If the buyer cannot or will not confirm a specific trigger event and a meeting, it is a Kill, not a Park.
Failure Mode 2: Pushing deals that should be parked
The tell: You keep sending follow-ups, adding value-add content, proposing new angles, because you believe if you just find the right message the deal will unlock. The buyer is politely non-committal. Your activity log is full. Their action log is empty.
Why it matters: You are subsidising the buyer's indecision with your time. Deals in this state rarely close. When they do, the price and terms are usually worse because the buyer knows you need it more than they do.
The fix: Check Signal 2 before you send anything. If the last meaningful buyer action is more than 10 days ago (SMB) or 20 days ago (enterprise), you are pushing a parked deal. Stop and send the Park script instead.
Weekly Cadence
Run this tracker on every open opportunity once per week, during your own pipeline review before any manager call. It takes under two minutes per deal once you know the signals. The goal is never to surprise yourself, or your manager, about where a deal actually stands.