5 to 7 stages, each tied to a verifiable buyer signal not a rep activity. Specific exit criteria that gate progression. MEDDPICC fields installed at Qualification and required to advance. Probabilities calibrated from historical close-rate data, not the HubSpot defaults. Stage-velocity reporting that flags drift inside 30 days. Done right, the pipeline is a forecast machine. Done wrong, it's gut-feel optimism wrapped in software.
The recommended stage architecture
For a typical B2B SaaS pipeline with 30 to 90 day cycles. Adjust for your motion, but the principles hold across most enterprise sales.
1. Discovery
Probability: 10%Buyer has had a meaningful conversation. Not just "a call happened" but "buyer's pain has been articulated and INSIDEA understands it."
Exit criteria: Pain identified and documented in deal notes. Pain owner identified (who feels this most acutely). Timeline initial estimate captured.
2. Qualification
Probability: 25%Deal qualifies on the M, E, and C of MEDDPICC: Metrics (the impact buyer expects), Economic Buyer identified, Champion identified.
Exit criteria: MEDDPICC fields populated for Metrics, Economic Buyer, Champion. Decision timeline confirmed. Budget range understood.
3. Demo or Technical Validation
Probability: 40%Buyer has seen the product or capability mapped to their pain. Demo isn't a feature tour, it's a tailored conversation against the Metrics defined at Qualification.
Exit criteria: Demo delivered. Technical fit confirmed. Decision criteria documented. Mutual action plan drafted.
4. Proposal
Probability: 60%Pricing and scope shared in writing. Buyer is comparing INSIDEA against alternatives or against doing nothing. The MEDDPICC P, P, I (Paper Process, Identify Pain, Competition) all matter here.
Exit criteria: Proposal delivered. Paper Process identified (procurement, legal, security review). Competition known.
5. Negotiation
Probability: 75%Buyer wants to buy, terms are being settled. Pricing, scope, T&Cs, security, legal. Champion is actively selling internally.
Exit criteria: Verbal commit captured. Procurement engaged. Redlines under negotiation. Close date locked within 30 days.
6. Verbal / Contract Sent
Probability: 90%Final terms agreed verbally. Contract delivered for signature. Anything not closed within 14 days at this stage is at real risk and gets flagged for revisit.
Exit criteria: Contract sent. Signatory confirmed. Expected close date within 14 days.
7. Closed Won / Closed Lost
Probability: 100% / 0%Terminal stages. Closed Lost requires loss reason captured (pricing, competition, no decision, timing). Loss reason data drives the next quarter's product and marketing priorities.
Why most pipelines don't work
1. Stages describe rep activity, not buyer signals. "Demo Scheduled" means the rep booked a demo. It doesn't mean the buyer is qualified. Reps move deals to that stage as they work, regardless of buyer commitment. Fix: rewrite every stage as a buyer-side condition.
2. No exit criteria. Deals advance because the rep clicks the dropdown, not because anything happened. Reviews where forecast accuracy gets blamed on "sales execution" are usually pipeline-design issues. Fix: HubSpot workflow validation that prevents stage advancement until specific fields are populated.
3. HubSpot default probabilities. 100% Closed Won, 90% Verbal, 50% Proposal, 25% Qualification. These are the same for everyone, which means they reflect no one's reality. Fix: pull your historical close-rate data and recalibrate from your own numbers.
4. No velocity tracking. Deals sit at Demo for 90 days and nobody notices. Fix: calculated property for days-in-stage, alerts on outliers, weekly review of stage drift.
5. Forecast based on weighted pipeline alone. $4M in stage 4 at 60% = $2.4M forecast. That math is wrong. Mature forecasting weights by both stage probability AND time-in-stage AND deal-size correlation with that stage's historical close rate. We install a forecasting layer on top of the pipeline, not just the stage probabilities.
MEDDPICC, the qualification layer
MEDDPICC fields installed on every deal, populated progressively as the deal advances. Required-on-stage rules: a deal can't advance to Qualification without M, E, C populated. Can't advance to Demo without D, D added. Can't advance to Proposal without P, P, I, C all complete.
This forces real qualification. Reps push back because it's more work upfront. The payoff: forecasts inside the next quarter become 2-3x more accurate, AE coaching becomes data-driven (you can see exactly which letter of MEDDPICC the deal is weak on), and deals that get stuck reveal their weakness instead of dying mysteriously.
A Series B SaaS customer's forecast variance dropped from ±35% to ±8% in one quarter after we redesigned the pipeline architecture. Deal velocity (time from creation to close) compressed by 22%. The CRO referenced it as the cleanest forecast call she had run in three years.
FAQ
How many deal stages should I have?
5 to 7 for B2B SaaS. Fewer than 5 and you lose forecast precision; more than 7 and reps stop updating accurately because the bookkeeping cost outweighs the value. Each stage should represent a meaningful change in deal probability based on a verifiable buyer action, not an internal milestone.
What's wrong with HubSpot's default pipeline stages?
They're a starting point, not a model. The default stages (Appointment Scheduled, Qualified to Buy, Presentation Scheduled, Decision Maker Bought-In, Contract Sent, Closed Won, Closed Lost) describe what reps have done, not what the buyer has done. The result: reps move deals forward as they make calls, even when the buyer hasn't signaled intent. Forecasts drift, deals stall, board reviews get awkward. We rebuild stages around buyer signals, not rep activity.
What is MEDDPICC and should I use it in HubSpot?
MEDDPICC stands for Metrics, Economic Buyer, Decision Criteria, Decision Process, Paper Process, Identify Pain, Champion, Competition. A qualification framework that's particularly strong for enterprise B2B. We install MEDDPICC fields on every deal at the Discovery or Qualification stage. They become exit criteria: a deal can't progress past Qualification without specific MEDDPICC fields filled in. Forces real qualification, not gut-feel optimism.
How do I set probability per stage?
Use historical close-rate data, not the HubSpot defaults. Pull every closed deal from the last 12 months, segment by what stage they were in 90 days before closing, calculate the close rate per stage. That's your probability. Most B2B SaaS pipelines have probabilities like Discovery 10%, Qualification 25%, Demo 40%, Proposal 60%, Negotiation 75%, Verbal 90%. The defaults assume a different shape that rarely matches your reality.
What are exit criteria and why do they matter?
Specific conditions that must be met to advance a deal to the next stage. Not 'I had a good call' but 'pricing has been shared in writing AND the economic buyer has been identified AND a mutual action plan exists.' Exit criteria force reps to do qualification work instead of moving deals forward on hope. Implemented as required field validation in HubSpot workflows.
How do I track deal velocity?
Three metrics. (1) Average days in stage: how long deals sit at each stage before moving. (2) Stage conversion rate: what percentage of deals at stage N advance to stage N+1. (3) Time from creation to close-won. Set up a calculated property on the Deal object for days-in-current-stage, plus a HubSpot custom report showing stage-conversion rate over rolling 90 days. Alert when any stage's average duration exceeds 1.5x the historical baseline.
Should I have multiple pipelines?
Yes if you sell distinctly different motions. Common splits: New Business vs Renewals/Expansion (different buyer, different cycle, different stages). Inbound vs Outbound (different early-stage qualification). Self-serve vs Enterprise (different deal sizes, different sales cycles). Don't split by sales rep or by team; that's an organization problem, not a process problem.
How long does it take to redesign a pipeline?
2 to 4 weeks. Week 1 is data analysis: pull closed deals, identify the actual buyer journey from your data. Week 2 is stage design and exit criteria. Weeks 3 to 4 are HubSpot configuration, automation builds, training. The hardest part isn't the configuration; it's getting the team to actually use the new exit criteria instead of falling back to gut-feel staging.
How INSIDEA works
