Forecast accuracyUnweighted pipeline rule

Why is my HubSpot forecast not weighted by stage probability?

Why a single missing stage probability breaks the whole forecast

Weighted-pipeline forecasting is the single most common forecasting model in B2B sales. The math is straightforward:

Weighted forecast = Σ (Deal amount × Stage probability)

A New deal worth $50k in a stage with 10% probability contributes $5k. A Negotiation deal worth $50k in a stage with 80% probability contributes $40k. The math captures the reality that late-stage deals are much more likely to close than early-stage ones.

But this all depends on every stage having a probability. If one stage is missing it, HubSpot has two choices: treat the missing probability as 0% (excluding the deal entirely) or as 100% (including it at full value). Both are wrong, and which one your portal does depends on configuration.

The result: deals in the misconfigured stage are either invisible to the forecast or hugely overstated. Either way, your number is off — sometimes by hundreds of thousands of dollars — for reasons no rep can explain.

What pipeline-level configuration actually drives the math

The check is at the pipeline level, not the deal level:

  1. For each pipeline, walk through every stage.
  2. For each stage, check whether metadata.probability is set.
  3. Flag pipelines where any non-closed stage has no probability.

Closed Won is by definition 100%. Closed Lost is 0%. Every stage between them needs an explicit probability for the math to work.

Why drift quietly breaks pipelines that started right

The pipeline configuration is one of those settings that gets right at setup, then drifts:

A new stage gets added without probability. A sales-ops change to add a "Procurement Review" stage between Negotiation and Contract Sent. Whoever added it forgot to set the probability. Three months later, deals are flowing through it. The forecast quietly under- or over-counts them.

A pipeline gets cloned and only partially updated. A new pipeline cloned from sales for partnerships. Stage names changed to fit partnership flow, but probabilities not adjusted. They're now using sales probabilities for a totally different conversion pattern.

A "stage probability override" is set and forgotten. HubSpot lets you set a per-deal probability override that wins over the stage default. Useful for one-off cases. Easy to forget you set it. Three months later, that deal is in your forecast at 50% even though its stage says 80%.

The audit is straightforward to run once. Catching the drift is the hard part — and HubSpot doesn't surface "any stage in any pipeline is missing a probability" as a built-in alert. You'd need to script it, schedule it, and then actually look at the report when it runs. Most teams skip this step entirely and find out their forecast was wrong only when finance asks why CRM-pipeline doesn't match the close report.

The manual HubSpot recipe

Settings → Pipelines → eyeball every stage. The check is fast; remembering to re-run it after every pipeline change is the discipline that breaks down.

HubSpot recipe~10 minutes to audit · per pipeline
  1. Open Settings → Objects → Deals → PipelinesClick the gear icon → Objects → Deals → Pipelines. This shows every pipeline (sales, partnerships, renewals, etc.).
  2. Click 'Edit stages' on each pipelineFor each pipeline, open the stage editor. The probability field appears on the right side of each stage row.
  3. Check every non-closed stage has a probabilityScan down the list. Closed Won is by definition 100%. Closed Lost is 0%. Every stage between them needs an explicit probability — even if it's 0.
  4. Flag any blank or null probability stagesDocument which stages are missing. Don't fix immediately — talk to sales-ops first because changing probabilities mid-quarter shifts the forecast number.
  5. Repeat for every pipeline5-10 minutes per pipeline. For organizations with 4+ pipelines, this is a 30-45 minute audit.
  6. Re-run after any pipeline changeStage added, stage renamed, pipeline cloned — any change risks introducing a missing probability. Audit becomes part of the change process or it doesn't get done.

What Bloated does instead

The Unweighted pipeline rule

Forecast-breaking pipeline drift, caught the day it happens.

Bloated checks every stage in every pipeline via the HubSpot API on every scan — flagging the moment a new stage gets added without a probability, or a clone misses an update. Forecast math integrity becomes automatic. Pair with the suggested action: assign sales-ops to set the probability, or escalate to RevOps for forecast model review.

Reads: metadata.probability · HubSpot pipeline configuration
3stages
Unweighted pipelineField: metadata.probability · HubSpot pipeline configuration
S
Sales pipeline – Procurement Review
Probability: not setAdded 3mo ago
Set probability
P
Partnerships pipeline – Pilot stage
Probability: not setCloned from Sales, not updated
Set probability
R
Renewals pipeline – Awaiting signature
Probability: 0% (likely wrong)Excluded from forecast
Verify
Run a free scan7 days free · no card · cancel any time