A pipeline is only useful if every stage reflects real buyer behavior and every deal moves on evidence, not optimism. This guide walks you through the full build: mapping the buyer journey, defining exit criteria, setting coverage targets, and locking in the review cadence that keeps forecast error inside ten percent.
Time: 90 minutes
Before you create a single stage, sit with three recent closed-won customers and three closed-lost deals. Walk through the moments where buyer behavior visibly changed: first meaningful conversation, scoping, proof, procurement, signature. Those shifts become your stage boundaries. Avoid the common trap of mirroring internal activities like "demo scheduled" or "proposal sent." Those describe your workflow, not the buyer. Pipelines built on seller actions decay fast because reps game them. Pipelines built on buyer evidence hold up under scrutiny and feed cleaner forecasts. Write the journey on paper first, then translate it into stages. Six is a healthy ceiling for most B2B motions. Transactional segments can collapse to four. Enterprise motions sometimes need seven, never more.
A stage without exit criteria is a wish. Every stage needs a short checklist of observable facts that must be true before a deal advances. Think "economic buyer identified by name and title," not "feels engaged." Keep each list to three or four items. More than that and reps either lie or stop updating the record. Fewer than three and the gates are too loose to filter noise. Store these criteria in your CRM as required fields on the opportunity, not as a wiki page nobody reads. When a rep tries to drag a deal forward with blank fields, the system blocks it. That friction is a feature. It is the single highest-leverage change you can make to forecast accuracy.
Each stage carries two numbers: a win probability and a forecast category. Probability drives weighted pipeline math. Forecast category tells leaders which deals belong in Commit, Best Case, Pipeline, or Omit. Do not invent these percentages. Pull them from your own closed-won rate at each historical stage. If you have no history, borrow benchmark percentages for your motion and segment, then recalibrate after your first full quarter. Forecast categories are judgment overlays on top of probability. A late-stage deal with weak sponsor coverage belongs in Best Case, not Commit, no matter what the probability says. Train your managers to use both signals together. Probability alone produces smooth-looking garbage.
Decide what every opportunity must carry: amount, close date, next step, decision maker, decision date, source, competitor, and loss reason if applicable. Make the first six required at creation. Loss reason becomes required only on closed-lost. Keep the field list ruthlessly short. Every extra field is a tax on rep time and a surface for garbage data. Use picklists over free text wherever a finite set of answers exists. Picklists are forecastable. Free text is not. Resist the urge to add vanity fields that only one executive looks at twice a year. If a field does not feed a report, a routing rule, or a required decision, it does not belong on the opportunity.
Coverage is the ratio of open pipeline to the quota you need to hit in a given period. The healthy range depends on your win rate. A twenty-five percent win rate implies four-times coverage at the start of the period. A fifteen percent win rate implies closer to six. Calculate your actual conversion rate from Stage One to closed-won across the last four quarters, invert it, and that is your minimum coverage. Then set your early-period target twenty percent above that minimum to leave room for slippage. Publish the number. Reps and managers need a single, visible target that tells them whether to be prospecting harder or defending what they have.
Pipelines rot without maintenance. Automate the maintenance so reps cannot avoid it. Flag any opportunity with a close date in the past, any deal with no activity in fourteen days, any deal missing required fields, and any deal where the next-step field is older than seven days. These flags should surface in a daily rep view and a weekly manager view, not in an email nobody opens. The point is to make hygiene visible, not punitive. Reps who see their own stale deals will fix them before a manager asks. Hygiene rules also protect your forecast: a deal with a thirty-day-old next step is almost always slipping, whether or not the rep has admitted it.
The pipeline review is where theory meets reality. Hold it weekly, same time, same agenda, no exceptions. Walk every rep through their top five deals and every deal above a dollar threshold you set based on average deal size. For each deal, ask three questions: what has to be true for this to close, what evidence do you have that it is true, and what are you doing this week to prove or disprove it. Do not accept "they said they would get back to me." That is not a next step. A next step is a confirmed calendar event with a named person on a specific date. Managers who enforce this standard see forecast error drop by half within two quarters.
Once the pipeline is live, you need four numbers in front of leadership every week: average days per stage, stage-to-stage conversion rate, overall win rate, and average deal size. Together they describe pipeline velocity. If velocity drops, something upstream is broken: lead quality, qualification discipline, or stage criteria enforcement. Instrument these in dashboards tied to live CRM data, not spreadsheets maintained by hand. Spreadsheets drift. Dashboards do not. Watch especially for the stage where deals pile up longest; that is almost always where your real friction lives, and it rarely matches where reps say they are stuck.
A pipeline is not a static artifact. Buyer behavior shifts, product changes, segments mature, and competitors move. Every quarter, pull the last ninety days of closed-won and closed-lost data and ask three questions: did deals actually move through stages in the sequence you designed, did exit criteria predict wins, and did any stage become a graveyard where deals go to stall. If a stage shows a conversion rate below ten percent, either the gate is too loose or the stage should not exist. If a stage shows above eighty percent, it may be redundant. Treat the pipeline the way a product team treats a shipping product: measured, iterated, and never finished.
Most B2B motions perform best with five or six stages. Transactional segments with sub-thirty-day cycles can run four. Enterprise motions with multi-stakeholder procurement sometimes need seven. More than seven creates maintenance overhead without improving forecast accuracy, and typically signals that seller activities have been mistaken for buyer commitments.
Use one pipeline per sales motion, not one per rep. If you sell to SMB and enterprise with different sales cycles and buying committees, build two pipelines. If you sell new business and expansion with meaningfully different stages, build two. Fragmenting further than that makes cross-rep benchmarking impossible and bloats the admin surface.
Treat the first ninety days as a stabilization window and resist changes unless something is clearly broken. After that, review quarterly. Change stages only when you have at least a full quarter of data showing a stage either traps deals or passes everything through. Avoid changing stage definitions mid-quarter; it breaks historical reporting and confuses the team.
Allow skipping forward with manager approval and a logged reason. Never allow silent skipping. A deal that jumps from early qualification to late-stage negotiation is either a genuine fast-track or a rep hiding missing diligence. The approval requirement surfaces the difference. Skipping backward should be routine and encouraged, since it reflects honest reassessment.
Invert your win rate from first stage to closed-won, then add twenty percent as a buffer. A twenty-percent win rate implies five-times minimum coverage, so target six at the start of the period. The specific number matters less than publishing it, tracking it weekly, and treating it as a leading indicator for prospecting intensity rather than a vanity metric.
Weighted pipeline is useful as one input, not as the forecast itself. Use probability math to spot gaps in coverage and to compare reps on a normalized basis. Build the actual commit forecast from manager-reviewed deals with named evidence. Pure probability math smooths over the judgment calls that distinguish accurate forecasters from optimistic ones.
Strkr gives you the stage logic, required fields, hygiene automation, and review views you need to run this playbook without stitching tools together. Build once, review weekly, forecast with confidence.
No sales call, no migration consultant, no four-month implementation. Enter your card, get 14 days of the full Pro tier, cancel any time before day 14 with zero charge. Spin up a workspace, import your CSV, and have something useful before lunch.
