Revenue intelligence, defined in plain language.

How to read this page

Most of the terms below are in common use across B2B sales and are defined here the way the field uses them, including where a term is weaker than it sounds. A few are ours. Those are labelled, because a vendor inventing vocabulary and then using it as though it were established is one of the easier ways to be misled by a sales page — including this one.

Category

Revenue intelligence

Revenue intelligence is the practice of using data from across a company’s revenue systems — CRM records, email, calendars, calls, support tickets and delivery tools — to judge the real state of a pipeline, rather than relying on what has been manually entered into the CRM.

The category exists because CRM stage is a claim, not evidence: it reflects what a person typed, and when they last remembered to type it. Revenue intelligence tools differ mainly in which evidence they read and whether they show you the evidence behind a conclusion.

Conversation intelligence

Conversation intelligence is the analysis of recorded sales calls and meetings — transcription, topic detection, talk-time ratios and keyword tracking — to coach reps and surface what was said in a deal.

It is a neighbouring category to revenue intelligence rather than a subset of it. Conversation intelligence answers "what was said on the call"; revenue intelligence answers "what is happening to the deal", of which calls are one input.

Sales engagement platform

A sales engagement platform sequences and automates outbound contact — email cadences, call tasks and follow-up reminders — to help reps execute a high volume of consistent outreach.

Engagement platforms are built to create activity. Revenue intelligence is built to interpret it. They are commonly confused because both read email and calendar data.

CRM (customer relationship management)

A CRM is the system of record for accounts, contacts, deals and their stages. It stores what people in the business have recorded about a customer relationship.

A CRM records decisions; it does not verify them. Every metric below is only as good as the last time someone updated the underlying record.

Pipeline metrics

Pipeline coverage ratio

Pipeline coverage ratio is total open pipeline value divided by the quota or target for the same period. A ratio of 3x means there is three times as much open pipeline as the number the team has to hit.

The ratio is only meaningful if the pipeline in the numerator is real. Coverage calculated over stale or unqualified deals is the most common way a team is reassured by a number that is telling it nothing.

Sales velocity

Sales velocity estimates how much revenue a pipeline produces per unit of time, usually as (number of opportunities × average deal value × win rate) ÷ average sales cycle length.

It is a diagnostic rather than a forecast: it is most useful for seeing which of its four inputs changed, not for predicting a specific quarter.

Win rate

Win rate is the proportion of opportunities that end in a closed-won outcome, measured over a defined set of deals — usually those that reached a given stage or closed within a period.

Win rate is highly sensitive to which deals are counted. A team that deletes or reclassifies dead deals will report a higher win rate than one that lets them close as lost, without selling any differently.

Sales cycle length

Sales cycle length is the elapsed time between a defined pipeline entry point — often opportunity creation or first meeting — and a closed outcome.

Averages hide the tail. A median alongside the average usually tells a more honest story, because a handful of very long deals can move the mean substantially.

Stage conversion rate

Stage conversion rate is the proportion of deals that move from one pipeline stage to the next, used to find the point in a sales process where deals most often stop.

It measures movement between recorded stages, so it inherits every delay and omission in how stages are updated.

Slipped deal

A slipped deal is an opportunity whose expected close date has been moved into a later period, usually after it failed to close in the period it was committed for.

Repeated slippage on the same deal is one of the clearest signals available in a CRM, because it is recorded automatically as a side effect of the date being changed.

Deal risk

Deal risk is the likelihood that an open opportunity will not close as forecast, inferred from signals such as buyer-side silence, a single point of contact, repeated close-date changes, or stalled progression through stages.

Risk assessment is only as trustworthy as the evidence it is drawn from. A risk score computed over a pipeline with little recorded activity is measuring the CRM’s emptiness, not the deal.

Qualification

MEDDIC

MEDDIC is a B2B sales qualification framework covering Metrics, Economic buyer, Decision criteria, Decision process, Identify pain and Champion. Each letter is a piece of information a seller is expected to establish before treating a deal as qualified.

Extended variants add Competition (MEDDPICC also adds Paper process). MEDDIC is a checklist for what you should know about a deal — it does not itself tell you whether what you recorded is still true.

BANT

BANT is a qualification framework covering Budget, Authority, Need and Timeline. It originated at IBM and is the most widely taught qualification shorthand in B2B sales.

BANT assumes a single authority figure and a defined budget, which fits fewer modern B2B purchases than it used to — most now involve a buying committee and a budget assembled during the deal.

Buying committee

A buying committee is the group of people at a prospective customer who together decide on a purchase — typically an economic buyer, one or more users, and reviewers from functions such as security, legal, IT or procurement.

Multithreading

Multithreading is maintaining active relationships with several people inside a prospective customer, rather than running the whole deal through one contact.

The opposite condition, a single-threaded deal, is a well-known risk: the deal depends entirely on one person remaining engaged, employed and internally persuasive.

Champion

A champion is a person inside the prospective customer who wants the purchase to happen and advocates for it internally when the seller is not in the room.

A champion is not the same as a friendly contact. The distinction is whether they spend their own credibility on the deal.

Forecasting

Sales forecast

A sales forecast is a stated expectation of how much revenue will close in a defined period, produced from the open pipeline plus judgement about which deals will land.

Weighted pipeline

Weighted pipeline multiplies each open deal’s value by a probability — usually attached to its stage — and sums the result to produce an expected revenue figure.

Stage-based weighting assumes the stage is accurate and that historical stage probabilities still hold. Both assumptions fail quietly, which is why weighted pipeline and actual results often diverge without anyone noticing until the period ends.

Forecast categories

Forecast categories are the buckets a deal is placed in to express confidence — commonly Commit, Best Case, Pipeline and Omitted (sometimes Closed). Commit is the portion a leader is prepared to be held to.

Categories are a judgement recorded by a person, which makes them useful as a record of what was believed and unreliable as evidence of what is true.

Bottom-up vs top-down forecast

A bottom-up forecast is built by summing individual deal judgements. A top-down forecast starts from a target or historical trend and works back. Most organisations produce both and reconcile the gap.

The gap between the two is usually more informative than either figure, because it shows where leadership expectation and deal-level reality disagree.

Quota attainment

Quota attainment is closed revenue as a percentage of the target assigned for the same period, measured for an individual, a team or the whole organisation.

Governance

EU AI Act

The EU AI Act (Regulation (EU) 2024/1689) is European legislation that regulates artificial-intelligence systems by risk level, imposing obligations that scale from transparency duties for limited-risk systems up to substantial requirements for high-risk uses.

Annex III lists high-risk areas including employment and worker management, which is why AI applied to sales-team performance is treated very differently from AI applied to deals and accounts.

NIST AI Risk Management Framework

The NIST AI Risk Management Framework is a voluntary United States framework, published by the National Institute of Standards and Technology, that organises AI risk management into four functions: Govern, Map, Measure and Manage.

It is not law and confers no certification. US buyers commonly use it as the structure for a vendor AI review in the way EU buyers use the AI Act.

Model AI Governance Framework (Singapore)

Singapore’s Model AI Governance Framework is voluntary guidance published by the IMDA and PDPC setting out how organisations should deploy AI responsibly, covering internal governance, human involvement in decisions, operations management and stakeholder communication.

Human in the loop

Human in the loop describes a design where a person reviews or approves an automated system’s output before it takes effect, rather than the system acting on its own.

The term is often used loosely. It is only meaningful if the human can actually see the basis for the output and is able to overrule it in practice, not merely in principle.

Explainability

Explainability is the degree to which the reasoning behind an automated output can be presented to a person in terms they can evaluate and challenge.

Explainability is distinct from interpretability. A system can produce a plausible-sounding rationale that is not the actual cause of its output, which is why the useful test is whether the stated evidence can be checked at source.

SalesSynq terms

Synq ScoreOur term

Synq Score is SalesSynq’s term — not an industry standard — for a deal-level score expressing how well the evidence behind a deal supports its current position, alongside the evidence used to produce it.

It is a SalesSynq name for a SalesSynq output. Where you see it used as though it were a general metric, that is us, and you should read it as a product term rather than a category one.

AbstentionOur term

Abstention is the behaviour of withholding a score or answer when the available evidence is too thin to support one, and saying what is missing instead of producing a number anyway.

The general concept — a model declining to answer under uncertainty — is well established in machine learning. Our use of the word as a product-facing promise is ours, and it is checkable: the product should tell you what would make the answer trustworthy again.

Evidence reliabilityOur term

Evidence reliability is SalesSynq’s term for how much confidence a conclusion’s underlying sources justify — accounting for whether a source is connected, currently reporting, and covering the period in question.

The distinction it exists to make is between "this deal is quiet" and "your data is quiet". Those look identical in most tools, and they call for opposite responses.

Read-only posture

A read-only integration reads records from a connected system without writing to them, so the connected system’s data cannot be altered by the tool reading it.

This is a general integration term rather than a SalesSynq coinage. In SalesSynq it is the default: writing anything to a work tool is a separate permission that is off until an administrator enables it.