A sales process is the operating system reps run on. Done right, it turns every opportunity into a predictable sequence of actions, every stage into a decision point with specific exit criteria, and every forecast into a conversation grounded in evidence instead of hope. Done wrong, it becomes a bureaucratic layer reps work around, forecasting becomes a gut feel exercise, and new reps take 12 months to ramp instead of six. This guide walks you through the full build: mapping how buyers actually buy, defining stages and exit criteria, building playbooks per stage, wiring forecast categories, instrumenting the CRM, piloting with one team, and reviewing the process quarterly so it evolves with the business.
Time: 4 hours
Most sales processes are built around how reps work: discovery, demo, proposal, close. The resulting process reflects the rep's to-do list, not the buyer's decision journey, and reps end up advancing deals through stages the buyer has not actually reached. Flip the frame. Pull the 30 closed-won deals and walk every one backward through the lens of the buyer: when did the buyer first recognize a problem, when did they involve the second stakeholder, when did finance get looped in, when did they shortlist vendors, when did they commit to buy, when did signature happen. The sequence that emerges is the buyer journey. Build the sales process to mirror the buyer journey, with a stage boundary at every major buyer decision point. Reps advance a deal only when the buyer has crossed the decision point, not when the rep has sent the next deliverable. This single reframe is what separates sales processes that forecast accurately from the ones that live on hope.
Four to six stages is the right range. Fewer than four and the process cannot distinguish early from late pipeline. More than six and reps spend more time moving deals than working them. Pick names tied to buyer decisions: "Problem Confirmed," "Economic Buyer Engaged," "Solution Validated," "Procurement Started," "Verbal Commit," "Closed Won." Each name describes something the buyer has done, not something the rep has done. Pair each stage name with a one-sentence definition so there is no ambiguity about what the stage means. Avoid overlap. "Qualified Lead" and "Discovery Scheduled" are the same thing in different words; a process with both will produce arguments about which stage a deal belongs in. Share the stage definitions with marketing, sales, and finance so everyone works from the same vocabulary. Vocabulary mismatches cause more forecast disputes than data problems do.
Exit criteria are the specific evidence required to advance a deal from one stage to the next. The common failure mode is writing exit criteria that are rep opinions ("buyer is bought in") rather than verifiable facts ("buyer has stated the pain in writing on a logged call"). Rewrite every criterion as something a manager can audit. For "Problem Confirmed," exit criteria might be: logged discovery call with named problem statement, confirmed timeline, identified business trigger. For "Economic Buyer Engaged," criteria might be: economic buyer attended a call, documented decision authority, confirmed budget range. Each criterion has to be verifiable from CRM data without asking the rep for context. If a manager has to call the rep to verify a stage is correct, the exit criteria are too vague. Store the criteria on the stage definition so they appear on the opportunity record. Make the stage advancement workflow require the criteria be checked off before the stage can be changed.
A stage without a playbook is a stage reps improvise. Build a playbook per stage with required activities, recommended activities, and the templates reps should use. For "Problem Confirmed," required activities might be: discovery call completed, discovery notes logged, next-step email sent within 24 hours. Recommended activities: LinkedIn connection with primary contact, follow-up research on the business trigger. Templates: discovery call questions, next-step email template, meeting notes template. The playbook sits inside the CRM on the opportunity record so reps can see it live, not in a wiki they visit once at onboarding. Each playbook should be short, no more than a page per stage, because long playbooks get skimmed and ignored. The playbook is the operational companion to the exit criteria. Exit criteria tell reps what the stage means; the playbook tells them how to get there.
Forecast categories give finance a view of the pipeline that is independent of stage. Common categories: Pipeline (deal exists but not forecasted), Best Case (upside), Commit (committed to close in-quarter), Closed (closed won or closed lost). Map each stage to a default forecast category, then let reps override within reason. For example, "Problem Confirmed" maps to Pipeline by default; "Solution Validated" to Pipeline or Best Case; "Verbal Commit" to Commit. The mapping should match the historical conversion rates from each stage to closed-won. If 40 percent of deals in Solution Validated close in the current quarter, Solution Validated's default category should be Pipeline, not Commit. Overriding should require a reason captured in a field. The combination of stage and forecast category gives sales leadership two independent views: process position (stage) and expected outcome (category). Both views matter for different conversations.
A sales process that lives on good intentions is a process that erodes within one quarter. Instrument the CRM to enforce the process. Make the exit-criteria check-offs required at stage advancement. Add validation rules that block stage changes until required fields are populated. Build picklists for close reasons, next steps, and competitive mentions so the data is analyzable later. Create reports that surface deals missing playbook activities so managers can coach rather than audit. Avoid overloading reps with required fields; every required field is a tax on rep productivity, so require only fields that drive decisions. For most teams, 8 to 12 required fields per opportunity is the right balance. Teams that require 30+ fields see rep adoption drop and data quality decline because reps start copy-pasting or ticking checkboxes to pass validation. Lean instrumentation beats heavy instrumentation.
A process without a review cadence atrophies. Build a weekly 1:1 between manager and rep (30 minutes, focused on deal review), a weekly pipeline council for the sales leadership team (60 minutes, focused on coverage and risk), and a monthly business review (90 minutes, focused on process health). Each meeting has a specific agenda tied to the process: the 1:1 reviews deals at risk and deals needing escalation; the pipeline council reviews coverage against the plan and stage conversion against expectations; the business review covers process health (stage conversion drift, time-in-stage trends, win rate by segment). The cadence turns the process from a static artifact into a living operating system. Without the cadence, deals age in stages they should have exited, stage conversion drifts without anyone noticing, and the forecast becomes a monthly argument. Store meeting agendas in the team wiki so attendance has a reason, not just a slot on the calendar.
Process health is measurable. Build a dashboard with three to five metrics: stage conversion rate (deals that enter a stage and advance within the expected window), time in stage (median days a deal spends in each stage), win rate by stage (deals that advance from each stage to closed-won), and pipeline age (deals aging past expected close date). Watch for drift. A sudden drop in stage conversion from Problem Confirmed to Economic Buyer Engaged signals the discovery playbook has stopped working, the market has shifted, or new reps are missing a key activity. A sudden rise in time-in-stage for Solution Validated signals the proof-of-concept framework is broken or the competitive landscape has moved. The dashboard is scanned every week in the pipeline council and dug into every month in the business review. Metrics without context are just numbers; metrics tied to specific process actions produce decisions.
Before rolling the full process to the organization, pilot it with one team for 30 days. Pick a team that has a manager willing to invest the time and reps who are open to feedback. Have every opportunity on the team run through the new process: stage definitions, exit criteria, playbooks, forecast categories, pipeline reviews. Track three things through the pilot: adoption rate (percentage of deals following the process), forecast accuracy (actuals versus predicted), and rep sentiment (short weekly survey). Expect to revise. Playbooks that look clean on paper surface friction when real deals move through them. Exit criteria that seem verifiable turn out to be ambiguous when a manager tries to audit a borderline deal. Revise mid-pilot based on the data, roll out to the next team in week five, and layer in the full organization over the following six to eight weeks. Phased rollout outperforms big-bang rollout.
The sales process succeeds or fails on manager behavior. If managers do not run the process in 1:1s, do not use the exit criteria to audit stage changes, do not treat the pipeline review as a decision forum, reps notice within one week and quietly stop using the process themselves. Train managers before reps. The manager training covers: how to run a weekly 1:1 against the process, how to audit stage changes using the CRM, how to coach a deal through stage advancement, how to handle a rep pushing back on exit criteria. Spend four hours on manager training, not 30 minutes. Pair the training with weekly manager calibration for the first eight weeks where managers compare notes on how they are applying the process. Manager calibration is where consistency across teams comes from; without it, every team applies the process slightly differently and the organizational data loses coherence.
The sales process is a living document. Every quarter, pull the process end to end and ask: do the stages still match the buyer journey, do the exit criteria still predict advancement, do the playbooks still reflect current competitive positioning, do the forecast category mappings still match actual conversion. Revise where the gap is material. Keep a changelog with every revision, the date, and the rationale. Share revisions team-wide with context on why the change is being made, because reps who see the process evolve with evidence trust it more than reps who find out from a manager that a stage got renamed. Treat the process the way a product team treats a shipping product: measured, iterated, versioned, never finished. The programs that go stale do not fail loudly; they fail quietly by producing worse forecasts, slower deals, and shorter rep tenures until someone notices the process has not been updated in two years.
Four to six is the right range for most B2B sales organizations. Fewer and the process cannot distinguish early from late pipeline. More and reps spend more time moving deals through stages than working them. Transactional SMB sales can land at four stages; complex enterprise sales usually need six. Pick based on the natural buyer decision points in your market, not on process complexity.
Exit criteria are the specific evidence required to advance an opportunity from one stage to the next. Good criteria are verifiable facts a manager can audit from CRM data without calling the rep: a logged discovery call with named problem statement, a documented decision authority, a committed timeline. Bad criteria are rep opinions like buyer is bought in that cannot be audited.
Map each stage to a default forecast category (Pipeline, Best Case, Commit, Closed) based on historical conversion from that stage to closed-won. Early stages default to Pipeline; late stages like Verbal Commit default to Commit. Reps can override within reason, with a required reason field. The combination gives leadership two independent views: process position (stage) and expected outcome (category).
Quarterly for structural revisions (stage changes, exit criteria updates, playbook refresh). Monthly for minor adjustments based on pipeline review signals. The process is a living document, not a kickoff deliverable. Teams that revise quarterly produce more accurate forecasts and faster ramp times; teams that treat the process as fixed see it decay within a year.
The stages and exit criteria should be the same across teams selling into the same market, but segment-specific playbooks are often necessary. An enterprise team selling into Fortune 500 accounts and a mid-market team selling into growth companies share the same stages but need different discovery questions, proof points, and procurement playbooks. Shared stages, segmented playbooks.
For a team of three, yes, with heavy manager discipline. For anything larger, no. The process depends on stage audit, exit-criteria check-off, pipeline review, and health metrics, all of which require CRM data. Strkr delivers the stage model, exit criteria, playbook surfacing, forecast categories, and dashboards the process needs to live. Without a CRM, the process becomes a slide deck nobody references.
Strkr pins the sales process to the opportunity record, surfaces exit criteria and playbooks in context, and feeds pipeline data back into stage conversion dashboards. Build the process once, review it every quarter, see the forecast tighten.
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.
