Deal health - Beacon

What it is

Modelled. How strong the commercial relationship behind an open deal is, from 0 to 100, where a higher number means a healthier relationship. It answers a different question from whether the deal will close. deal_probability answers will this close? Deal health answers if it closes, will this become a good customer? A deal can score high on one and low on the other, and that gap is the whole reason both exist. It is modelled, not measured. A measured value is reproduced by running the same arithmetic over your data. A modelled value is composed: it weighs several things against each other, it is scored against what is normal for deals like this one, and it sharpens as your closed-deal history builds. Everything below follows from that difference.

How it is calculated

Modelled. Five dimensions are each scored 0 to 100 and combined at fixed weights.

Dimension	Weight	What it reads
Multi-threading	0.30	How many people are involved on the buyer’s side, how senior they are, how many functions they cover, and how often the buyer starts the conversation
Executive sponsor	0.25	Whether a sponsor is named and confirmed, whether an executive-to-executive meeting has happened, and how recently the sponsor was active
Reciprocity	0.20	The balance of buyer-initiated to seller-initiated activity, how fast the seller responds, and days since the last two-way exchange
Mutual success plan	0.15	Whether a plan exists, has a timeline and milestones, has been reviewed with the buyer, and whether those milestones are realistic
Technical qualification	0.10	Whether the technical evaluation is complete, the evaluator is engaged, and any security or compliance review is done

The weights sum to 1.00. Multi-threading carries the most because stakeholder breadth is the strongest single predictor of both closing and of what happens afterwards. Reciprocity is there because a buyer who initiates contact has a different relationship with the purchase than one who only responds. A worked example, with two dimensions deliberately empty:

An enterprise deal at stage 4. Multi-threading 40, executive sponsor 0 (none named), reciprocity 70, mutual success plan 0 (missing), technical qualification 60. That gives 12 + 0 + 14 + 0 + 6 = 32.

Tier	Range
Healthy	70-100
Developing	55-69
At risk	40-54
Critical	0-39

Every dimension is scored against what is normal for this deal’s segment and stage, not against a fixed bar. A multi-threading score of 60 can be at risk for an enterprise deal at stage 4, where the norm is 80 or above, and healthy for a smaller deal at stage 2, where the norm is 45. Those norms are refreshed quarterly from your own closed deals. How complete the picture is travels with the score. Where the underlying data is thin, confidence is reduced and the score is flagged — it is a flag on confidence, not a change to the weights. Below 40% completeness no score is shown at all; you get health assessment pending instead.

Where it comes from

Source	What it supplies	Required
Your CRM — contact object	Who is on the deal, their seniority and function, when each was last active, and which activities the buyer started	Yes
Your CRM — activity log	Meetings, emails and calls, split by who initiated them; seller response times; days since the last two-way exchange	Yes
Your CRM — deal fields for sponsor, success plan and technical evaluation	Whether a sponsor is named and confirmed, whether a plan is complete, whether the evaluation is done	Yes
Your own closed deals, by segment and stage	The norms every dimension is scored against	Yes

Three of the five dimensions depend on CRM fields most companies do not already have. Executive sponsor, mutual success plan and technical evaluation are usually custom fields, and they have to exist and be filled before the score means anything. If your sales motion has no mutual success plan step, that dimension will score low consistently and correctly — the fix is the process, not the field. The score starts at stage 3. Before that you get a partial score carrying a flag. Segment norms need at least ten closed deals in that segment and stage; below that Beacon uses an estimate drawn from your other segments and says so.

How fresh it is

Recalculated on every qualifying CRM event — a contact added, a meeting logged, a sponsor or plan field updated — and again in a full daily pass. Updating a sponsor or plan field yourself triggers an immediate recalculation rather than waiting for the next cycle. Every response carries two timestamps. as_of is the moment the value describes. computed_at is when Beacon last worked it out. Read both rather than assuming a cadence.

Currency and rounding

Neither applies. The score has no unit and holds no money value. Deal size does not enter it — a small deal and a large one with the same relationship pattern score the same.

What changes it

Adding a stakeholder. Naming a sponsor. A buyer-initiated message arriving, or fourteen days passing without one at stage 3 or later. A slow seller response. Completing the success plan or the technical evaluation. And a quarterly benchmark refresh, with nothing changing at the deal at all. Because every dimension is scored against the segment-and-stage norm, a change in what is normal moves the score. This is the least intuitive property of the value and the main reason to read the two timestamps. What does not change it: deal size, stage advancement in itself, and close probability. The weights are not self-adjusting. A change of more than three points to any dimension weight needs a named owner’s approval, weights are not retuned more than once every ninety days, and the agent that operates this value cannot change the weights, the thresholds or the stage minimums on its own.

What it is not

Not deal probability. That is whether the deal closes. This is what you get if it does. A deal can be very likely to close and unhealthy, and that combination is worth knowing about before the close rather than after.
Not deal velocity percentile. That is how fast the deal is moving. This is how well it is built.
Not one of the five dimensions. Each is also 0 to 100, so a multi-threading score of 60 reads like a deal health of 60 and is not. Only the combined score is published on the deal.
Not health score. That one is measured, it is about an account after the sale, and it is about progression rather than relationship quality. Deal health at the close does seed the early view of the account; it does not become it.
Not a record of who your sponsor is. The score reads whether a sponsor is named, confirmed and recently active. The name itself is never published, on any door.

​What it is

​How it is calculated

​Where it comes from

​How fresh it is

​Currency and rounding

​What changes it

​What it is not