Book a strategy sessionBook a session
Sales pipeline & RevOps engine
Hubspot logo ops

Sales pipeline & RevOps engine

  • sales
  • pipeline
  • crm
  • revops

Run B2B pipeline as a RevOps engine, capture & route in minutes, qualify on PQL/BANT, work stages, forecast off hygiene, then analyse losses and recycle to sharpen the ICP.

Book a strategy session See how it runs
7 Steps in the flow
5 Tools orchestrated
6 Human checkpoints
Weekly Run cadence
forecast-accurate, monthly Closed-won revenue

The objective

What this system is built to do

The pipeline is a conveyor, not a graveyard. Route and qualify fast, give every stage a real exit criterion, forecast off hygiene not hope, and treat every closed-lost as the data that sharpens next quarter's ICP.

Route and qualify every lead fast, work each stage on a real exit criterion, forecast off hygiene, and recycle losses back through qualification while the loss reasons sharpen the ICP. Tools: , , , Calendly. Not doing: advancing deals on happy ears or forecasting off hope instead of next steps.

Cycle & hand-offs: Refine ICP & playbook runs monthly and starts the cycle. Runs on its own rhythm: Capture & route leads (daily); Qualify (PQL / BANT) (daily); Work pipeline stages (daily); Nurture & re-engage (weekly (Mon)); Forecast & manage (weekly (Mon)); Close & loss analysis (weekly (Fri)). Any other link waits for a person. Sign-off steps stay with a person; everything else can run on its own once armed.

How it runs

The steps your team follows

The workflow, exactly as it runs inside . It plays through step by step, or pick any node on the map.

1Capture & route leads
2Qualify (PQL / BANT)
3Work pipeline stages
4Nurture & re-engage
5Forecast & manage
6Close & loss analysis
7Refine ICP & playbook
Start
1

Capture & route leads

Weekly
  1. Catch every inbound + outbound lead into one place in .
  2. Enrich and de-dupe in , score against the ICP, and round-robin assign to the right rep, all in under 5 minutes.
  3. Drop a booking link in the first touch so a hot lead can self-serve a meeting.
  4. Human approval, confirm the routing rules and ICP score thresholds before this batch goes live.
  5. Resolve before you create. Search existing contacts and companies (active *and* archived) on each strong key, work email, canonical domain, before writing anything. One clear match → link to it; no credible match → create, with owner and source set; conflicting evidence → a review item showing the candidates. Never match a company on its name alone, and re-check just before the record is written.
  6. Route on two questions, not one: is the company a fit, and is this person a buyer? Fit company + buyer → meeting with the owner. Fit company + non-buyer → a helpful reply, and the account owner finds the buyer. Not a fit → a self-serve or nurture path, never a rep's calendar.
  7. Weekly leak sweep. Compare the inbox, calendar, call recorder and forms against the CRM for hand-raisers who got no reply, booking or follow-up. Confirm each drop in two systems before contacting anyone, check for an open deal owned by someone else, then draft one recovery note per confirmed drop. Human approval before anything sends.
2

Qualify (PQL / BANT)

Weekly
  1. Score PQL signals and run the deal through the BANT/MEDDIC gate in .
  2. Pull call evidence from to confirm pain, budget and authority, not the rep's optimism.
  3. Advance only on a written exit criterion; hard-disqualify or recycle anything that fails the bar.
  4. Tag the source signal so you can trace which segments actually qualify.
  5. Score the evidence, not the checkbox. For each dimension, pain, impact in the buyer's own numbers, budget, decision process, authority, timing, score 1–5 on what the record actually holds (a quote, a transcript line, an email). A 3 or above must point at something. Set a minimum per stage (default: every dimension ≥2 to enter proposal, ≥3 to commit; tune on your own data).
  6. Score when the evidence lands, straight after the call, from the transcript, not in a Friday batch from memory.
  7. A deal never sits above the stage its evidence supports. Commit-stage deals must answer in writing: *what breaks for them if they do nothing until next quarter?*
3

Work pipeline stages

Weekly
  1. Work each open deal against its stage exit criterion in , next step set, champion confirmed.
  2. Use to inspect deal momentum and surface risk (no next meeting, single-threaded).
  3. Keep every deal's next step and close date honest; no zombie deals.
  4. escalate deals that need pricing or exec air-cover.
  5. After every meeting: pull the notes and transcript from Granola, classify the outcome, update the deal (stage, next step, close date, evidence scores) and draft the follow-up in the owner's voice, committing only to what was actually agreed. A first meeting that clears the qualification bar gets a deal proposed, never auto-created. Human approval before anything sends or is created.
  6. Thread check. Any deal with one engaged contact: name the missing role (economic buyer, evaluator, user, champion) before suggesting a person. If a champion is warm, ask them for the intro, don't go around them on a late-stage deal.
  7. Mutual plan once value is agreed. Co-write with the buyer the steps *they* must take to reach value, owners, dates, approvals, security, legal. Mark unknowns *to confirm with the buyer*; never invent them. Measure progress by what the buyer has done, not what you've sent.
4

Nurture & re-engage

Weekly
  1. Find deals stalled past their expected stage age in .
  2. Run a light re-engage sequence, new value, a fresh signal, a reason to talk now.
  3. If they re-engage on a real trigger, push them back into the working stages; if not, route them to loss analysis.
  4. Don't let stalled deals quietly inflate the pipeline.
  5. Revive only what was once engaged. Three lanes: closed-lost deals, champions who moved to a new company, and conversations that went quiet after real engagement. Contacts who never replied are not dormant pipeline, they go back to prospecting.
  6. Name the loss pattern before drafting, timing or budget cycle, champion left, lost on one capability, price-to-value at that scope, sponsor lacked authority, went quiet after proposal. Reuse the play that worked for that pattern; a new pattern gets added to the library with this account as its first example.
  7. One knock, one follow-up. Default spacing: 7 days for closed-lost, 10 for champion movers, 14 for went-quiet (tune on your own data). No reply → passive nurture until a fresh signal fires. Reopen on new information only, never on elapsed time.
  8. Human approval, every revival draft goes to review with its lane, premise, evidence and next-touch date. Log each batch's outcomes by lane so the next cycle knows which lane earns more slots.
5

Forecast & manage

Weekly
  1. Build the forecast off stage probability and deal hygiene in , with forecast intelligence as the reality check.
  2. Inspect commits: next step set, champion confirmed, close date sane, strip anything that fails.
  3. Track pipeline coverage vs target and call the gap early.
  4. Human approval, review the call before it goes to the leadership number.
  5. Weekly review = a short action list, not a dashboard. Flag stalled (past stage age), regressed, missing activity (no contact, no next step, past close date) and at-risk deals, one action per deal with an owner and a date. Check for a fresh account signal only on flagged deals.
  6. Below-gate exceptions. Every deal sitting above the stage its evidence supports gets a verdict, named and dated: re-qualify, move back, or close lost. "Leave it" is not a verdict. Deals moving backwards is a sign the gates are real.
  7. Hygiene as reviewable operations. Snapshot first; each fix is one record, its current value, proposed value and reason. Sample before approving a batch, re-read each record just before writing, and skip any value that changed since planning. Then fix the source, the form, import or integration that created the defect, and give it an owner.
6

Close & loss analysis

Weekly
  1. Close won/lost in ; make the structured loss reason mandatory.
  2. Mine call recordings for why deals actually died vs the reason logged.
  3. Recycle 'timing / no-budget-now' losses straight back to qualification for re-engagement.
  4. Roll up the loss-reason distribution for the ICP refresh.
  5. Study wins as well as losses. Code every decided deal against one fixed driver set, capability, proof and trust, price-to-value, champion strength, competitor move, timing and budget, our own process, using the buyer's words, quoted. Note where the logged reason and the record disagree.
  6. Interview a sample of decided buyers 2–6 weeks after the decision. The interviewer is never the deal owner; no selling, no defending, no correcting.
  7. Mine recurring objections into six buckets, timing, budget, authority, trust, fit, competition, rank them by frequency, and name where upstream each should have been answered (offer, first touch, website, qualification). Move the answer there. Opt-out language goes straight to the suppression list.
7

Refine ICP & playbook

Weekly
  1. Read the loss-reason distribution and the segments that qualified-and-won.
  2. Cut ICP segments that book but never close; double down on the ones that convert.
  3. Update the qualification bar, the routing rules, and the rep talk-tracks.
  4. Feed the sharper ICP back into capture & routing for the next cycle.
  5. Derive the ICP from revenue, not the deck. Select deals with a populated value in the last 12 months (don't assume a won stage exists or is used consistently), do the sums in code, and cluster into 2–4 archetypes by revenue share, never deal count. Count the distinct companies behind each archetype, one large customer is a concentration to report, not a segment.
  6. Write the ICP as gates, default-deny. Cheapest checks first: structural fit → evidence of the problem → exclusions (competitors, clients, recent contact) → can we reach the buyer → can they pay our floor → is the timing window open. Tier 1 = every gate plus a timing signal; Tier 2 = every gate; Tier 3 = structural fit only (nurture, no paid enrichment). Keep the exclusion log.
  7. Test it before shipping it: the gates should place most of your last ten wins in Tier 1–2 and filter most late losses. If they don't, the ICP is still aspirational.
  8. On each new win, capture the win profile and queue a handful of lookalike accounts not yet in the CRM, with drafts for review. Human approval before any ICP change goes live.
1 / 7

Setup

The stack

The tools this system drives, and the context you bring once. Connect your own accounts, or swap in your team's equivalent.

Tools to connect

HubSpotFree – $90 / mo Powers Capture & route leads Or Pipedrive, Salesforce
Clayfrom $149 / mo Powers Capture & route leads Or Clearbit, Apollo
CalendlyFree – $12 / mo Powers Capture & route leads Or Cal.com, SavvyCal
GongCustom Powers Qualify (PQL / BANT) Or Chorus, Clari
Granola Powers Work pipeline stages

Context to bring

Your documents 4 files
  • ICP & scoring rules How leads are qualified, scored and routed.
  • Stages & exit criteria What has to be true to move a deal forward.
  • CRM access HubSpot / your CRM plus the call data it reads.
  • Playbook & messaging The plays and talk tracks reps run.

Proven in market

Run by 111 teams, verified on real accounts

Sales pipeline & RevOps engine is built and verified by Growth Division, the GTM team behind 130+ companies. Here is what the same play has produced for their clients.

Built a scalable sales system

Clicksit · B2B

Higher conversion across the journey

SeedLegals · Legal Tech

150+ MQLs in the first 6 months

Retain International · B2B Services

Built & verified by Growth Division

What we've learned running it hundreds of times

The small, hard-won details that decide whether it works, from every run.

<5 min speed-to-lead target
2.4× lead→meeting from fast routing
+10pp win rate from a real qualification bar
lead→meeting high confidence
Before 8%
After 19%

Routing leads in under 5 minutes lifted lead→meeting

Speed-to-lead is the cheapest conversion lever you have. The prospect's intent decays in minutes, not hours, route, enrich and assign before it cools, and the meeting rate roughly doubles with no extra spend.

Routing A/Bn≈2,400 inbound leads9 weeks
win rate high confidence
Before 18%
After 28%

An explicit PQL+BANT bar lifted win rate and freed SDR time

Qualification is a filter, not a formality. A written PQL+BANT bar with a real disqualify path raises win rate and hands your reps their time back, a clean 'no' is worth more than a hopeful 'maybe'.

Stage-gate rolloutn≈1,150 deals10 weeks
× coverage medium confidence
Before 2.1×
After 3.4

Loss-reason analysis + recycling lifted pipeline coverage, with a lag

Closed-lost is data, not an ending. Mandatory loss reasons plus a recycle loop rebuild pipeline coverage and sharpen the ICP, but it compounds over a sales cycle, so give it a quarter before you judge it.

Loss-reason + recycle loopn≈640 closed-lost13 weeks
The full lab notebook Every experiment charted, with the raw before/after and the decision rules behind each call. Book a session

Get started

How to run this system

Live in an afternoon. Clone it, connect your tools, set your inputs, and turn it on.

  1. 1

    Add to

    Clone the Sales pipeline & RevOps engine into your workspace, wired end to end.

  2. 2

    Connect your tools

    HubSpot, Clay, Calendly +2 more.

  3. 3

    Set your inputs

    Your ICP, targets and brand rules, once.

  4. 4

    Turn it on

    It runs weekly on autopilot, checking in with you only where it matters.

~40 min / week your time on autopilot · vs 10–14 hrs / week by hand
forecast-accurate, monthly Closed-won revenue, produced for you
All growth systems

Next launch: Week of 3 Nov 2026 · 1 slot left

Run the Sales pipeline & RevOps engine in your workspace.

In 30 minutes we find where your go-to-market is leaking and map the first system we’d launch. You leave with a plan whether or not you work with us.

Book a strategy session Get your free GTM Gap Report