# Your Pipeline Review Looks at Deals. It Should Be Looking at Relationships.

> Most pipeline reviews grade deal stage and ignore contact depth. That's why late-stage surprises keep happening.

By Greg McCallum, founder of PeerLab. Published 2026-11-09 in Pipeline & Forecasting. Source: https://www.peerlab.ai/blog/multi-contact-pipeline-review-what-youre-missing/

The late-stage slip is one of the most reliably embarrassing events in a sales organisation. A deal sits at Proposal or Negotiation for three consecutive weeks, the rep says it's tracking, and then on the fourth week it pushes to next quarter or dies entirely. Everyone in the room is surprised. No one should be. The information that would have flagged it was available. You just weren't asking for it.

The problem is structural. The standard pipeline review is built around three fields: stage, amount, and close date. Those three fields describe what a rep believes. They say almost nothing about what a buyer is doing. And yet most revenue teams run their entire forecast cadence off them, week after week, wondering why the surprises keep coming from deals that looked fine on the board.

## What the standard review actually measures

Pull up a typical pipeline review template. You will find columns for account name, opportunity value, stage, close date, next step, and maybe a "confidence" percentage that the rep has either invented or inflated. If you are running a more sophisticated cadence, you might have a MEDDIC or MEDDPICC field somewhere, partially filled.

What you will almost certainly not find is any structured view of contact depth: how many people inside the account have been meaningfully engaged, at what seniority, and how recently.

Stage is a rep's opinion about where they are in a process. It is frequently optimistic and structurally backward-looking, because stage gates are typically triggered by what the rep has done, not by what the buyer has confirmed. "Proposal Sent" is not a buyer action. It is a rep action. The buyer may have forwarded it to legal, filed it, or used it to benchmark a competitor. You do not know, because you are not tracking the relationships that would tell you.

## The structural blind spot

Here is a concrete scenario. A deal is sitting at £180,000, Stage 4 of 5, close date end of this quarter. The rep's next step is "awaiting sign-off from procurement." In the review, that deal gets a nod. It looks like pipeline. It is probably counted in the forecast.

Now ask a different set of questions. How many people inside that account have had a substantive conversation with anyone on your team in the last 30 days? What is the most senior person who has been directly engaged? Has anyone spoken to the economic buyer, not just been CCd on an email to them? Is there a coach inside the account who has told you, explicitly, what the internal process looks like?

If the answer to most of those is "just the champion" or "not sure," that £180,000 is not a Stage 4 deal. It is a single relationship wearing a Stage 4 badge. And single-threaded deals at late stage are where slippage lives.

The research on this is consistent. Multi-stakeholder deals close at higher rates and with shorter late-stage cycles when multiple contacts have been meaningfully engaged before the final approval stage. Meaningful engagement means a two-way exchange with substance, not a CC on a proposal email. When you interrogate your own historical data on deals that slipped versus deals that closed, contact breadth is almost always one of the separating variables. Most teams have never looked.

## Tearing down the review format

Take a standard Stage 4-5 deal section in a weekly review. It typically runs like this:

| Deal | Value | Stage | Close Date | Next Step |
|---|---|---|---|---|
| Acme Corp | £180k | 4/5 | 31 Mar | Awaiting procurement |
| BetaCo | £95k | 5/5 | 28 Feb | Contract review |
| Gamma Ltd | £220k | 4/5 | 31 Mar | Final presentation |

Three deals. One minute each, maybe two for Gamma because of the size. The rep confirms the next step. The manager asks if anything has changed. The rep says no. The deals stay in the forecast.

Here is what that section should look like:

| Deal | Value | Stage | Contacts Engaged (30d) | Most Senior Engaged | Champion Confirmed? | Coach Confirmed? |
|---|---|---|---|---|---|---|
| Acme Corp | £180k | 4/5 | 1 | Mid-level procurement | No | No |
| BetaCo | £95k | 5/5 | 3 | CFO (direct call) | Yes | Yes |
| Gamma Ltd | £220k | 4/5 | 2 | VP Ops | Yes | No |

That table changes the conversation immediately. BetaCo looks like a real deal. Acme, despite being the same stage, is a single thread into procurement with no confirmed champion and no coach. If it closes this quarter, it will be a surprise in the good direction. The forecast should reflect that. Gamma has decent contact breadth but no internal coach, which means your team does not know how decisions actually get made inside that account.

None of this requires the rep to have done anything differently. It requires the manager to ask a different set of questions in a structured way, every week, so that the absence of information is itself visible.

## Replacing one section of your review cadence

You do not need to rebuild your entire review. Add one pass to your late-stage deals, specifically Stage 3 and above, that checks four things:

1. **Contact count (30 days).** How many distinct people inside the account have had a substantive two-way exchange with your team. Phone calls, meetings, meaningful email threads. Not forwards, not CC lines.
2. **Seniority ceiling.** The most senior person your team has spoken to directly. Not "we know the CFO is involved." Directly spoken to.
3. **Champion status.** Named, confirmed, has explicitly advocated internally. Not "probably our champion."
4. **Coach status.** Someone who has given you real intelligence about the internal process, budget owner, competing priorities, or decision timeline. This is different from champion.

If a deal at Stage 4 or 5 fails two or more of those checks, it goes into a separate review bucket. Not written off. Not pushed from the forecast automatically. But treated as a relationship-risk deal, not just a stage-progression deal, and given specific actions in that week's review rather than a nod.

If you want a structured way to score these alongside other deal health signals, the [Deal Health / Risk Scorecard](https://www.peerlab.ai/library/tools/deal-health-risk-scorecard/) gives you a ready-built framework you can adapt to this contact-map logic without starting from scratch.

## Why this keeps not happening

The honest answer is that contact-depth data is harder to pull than stage data. Stage sits in the CRM. Contact engagement, especially at the quality level that distinguishes a CC from a real conversation, usually lives in call notes, email threads, and the rep's head. Getting it into a structured review requires either better CRM hygiene or a willingness to ask direct questions in the room and write down what you hear.

Most managers have been trained to review what the system shows them. The system shows stage, amount, and close date. So that is what gets reviewed.

If your team is running any kind of account-based motion on these deals, the [Account-Based Prospecting Playbook](https://www.peerlab.ai/library/account-based-prospecting-playbook/) has a contact-mapping section that maps directly onto the four-check framework above and can help reps build multi-threaded coverage before deals reach late stage.

## The forecast accuracy question

Stage without relationship topology is a single data point pretending to be a forecast. You can track your forecast accuracy in aggregate with something like the [Win Rate & Forecast Accuracy Tracker](https://www.peerlab.ai/library/tools/win-rate-forecast-accuracy-tracker/), but aggregate accuracy will not tell you which structural variable is driving the miss. When you start tracking contact depth alongside stage and go back to your historical slipped deals, you will find the same pattern almost every time: single-threaded late, no confirmed coach, most senior contact was not the economic buyer.

That is not bad luck. That is a measurement problem.

The teams that get consistently good at forecasting are not the ones with the best reps. They are the ones whose review cadence asks the questions that actually correlate with outcomes, and treats a gap in the answer as a signal worth acting on rather than a box that wasn't filled in correctly.

Stage tells you where you think you are. Relationships tell you whether you're right.
