Broker pipeline management: knowing which deals are actually real

Every broker has had the conversation where a principal asks 'what's in the pipeline this month' and the honest answer requires a pause, a scroll through emails and a rough guess. The pipeline exists, in the sense that deals are being worked, but it doesn't exist as something that can be reported on with any confidence. That gap between activity and visibility is where forecasting goes wrong.

The usual cause isn't a lack of effort. It's that pipeline stages were never properly defined, so 'in progress' can mean anything from 'submitted to three lenders yesterday' to 'haven't heard from the client in six weeks but I'm not ready to call it dead'. A pipeline built on stages like that will always overstate what's actually coming.

Why most pipelines lie a little

A pipeline that never gets cleaned up drifts steadily towards optimism. Deals get added when a broker has a promising first conversation, and they rarely get removed just as promptly when that conversation goes quiet. The result is a pipeline that looks healthy on a dashboard and doesn't match the number of deals that actually complete a few weeks later.

• Stages defined loosely enough that different brokers use them differently
• Deals left in an active stage long after contact has stopped
• No distinction between a deal that's likely and one that's merely possible
• Forecasts built on deal count rather than weighted probability
• No routine review of what actually converted against what was forecast

Defining stages that mean something

A stage is only useful if two different people looking at the same file would put it in the same place. That sounds obvious and is routinely not true. The fix is to define each stage by a concrete, checkable event rather than a feeling about how the deal is going.

Tie stages to evidence, not sentiment

'Submitted' should mean a pack has actually gone to a lender, not that one is being prepared. 'Terms issued' should mean a document exists with terms on it, not that a lender sounded keen on the phone. Vague stages like 'in progress' or 'ongoing' should be broken into something specific enough that anyone can audit it from the file alone.

• Enquiry: initial contact made, no qualification done yet
• Qualified: fact-find complete and product identified
• Packaged: documents collected and pack prepared
• Submitted: pack sent to at least one lender
• Terms issued: written terms received from a lender
• Offer accepted: client has accepted formal terms
• Completed: funds released

Different products will need different variants of this, and that's fine as long as each variant is written down rather than improvised per broker. Our guide on choosing a CRM built for commercial finance covers how stage design should reflect the products a firm actually writes rather than a generic sales template.

Conversion, measured honestly

Conversion rate is one of the few numbers that tells a broker or a principal something genuinely useful, but only if it's measured stage to stage rather than as one blunt enquiry-to-completion figure. A single overall percentage hides where deals are actually being lost. Knowing that seven in ten qualified deals reach submission, but only three in ten submissions reach terms, points at a completely different problem than the reverse would.

• Track conversion between each adjacent pair of stages, not just start to finish
• Segment by product, since a bridging pipeline and a development finance pipeline convert very differently
• Segment by lead source where possible, since some referral routes produce stronger deals than others
• Review conversion trends monthly rather than only when something feels wrong

This is also where a pipeline earns its keep as a management tool rather than just a to-do list. A drop in submission-to-terms conversion might mean packaging quality has slipped, worth checking against the guidance in what lenders want in an application pack. A drop in enquiry-to-qualified conversion usually points at lead quality rather than broker performance, which is a very different conversation to have with the team.

Spotting stale deals before they distort the forecast

Every pipeline accumulates deals that are technically open but functionally dead. The client's gone quiet, the lender never came back, or the deal was always more hope than substance. Left in an active stage, these deals inflate the forecast and waste attention that should go to live cases. The fix isn't to delete them hastily, since some do come back, but to treat time-in-stage as a first-class metric rather than an afterthought.

A pipeline that never loses a deal isn't healthy. It's just not being looked at.

• Set an expected maximum time in each stage, based on how the product actually moves
• Flag anything that exceeds it for a specific review, not an automatic removal
• Separate genuinely stale deals into a distinct status so they stop counting toward the active forecast
• Revisit stale deals periodically rather than writing them off permanently

Doing this manually across a full book is exactly the kind of task that gets skipped when things are busy, which is precisely when it matters most. Automated flags for stage age, visible on a dashboard rather than requiring a manual review, keep the pipeline honest without needing someone to remember to check.

Weighting the forecast properly

Counting deals is not the same as forecasting revenue. A pipeline that reports twenty live deals worth an estimated £4m tells a principal very little if half of those are early enquiries and half are sitting with signed terms. Weighting each deal by its stage-based conversion probability turns a deal count into something closer to an actual forecast.

• Apply a historical conversion percentage to each stage rather than treating every open deal as equally likely
• Update those percentages periodically as real conversion data comes in, rather than setting them once and forgetting them
• Report a weighted figure alongside the raw pipeline value, not instead of it
• Be explicit that forecasts are probabilistic, particularly when reporting to introducers or lenders about expected volumes

Getting the pipeline out of individual heads

None of this works if the pipeline lives across separate spreadsheets, personal notebooks and each broker's memory of their own patch. A shared system that everyone updates the same way is what makes stage definitions, conversion tracking and stale-deal flags possible at all. This is a large part of why firms move away from spreadsheets in the first place, as covered in our guide to moving from spreadsheets to a proper system: the value isn't a nicer interface, it's one accurate picture of the book that doesn't depend on asking around.

A well-run pipeline also supports the kind of operational discipline that separates firms as volumes rise, something explored further in why operational discipline is now a broker's competitive edge. Pipeline management isn't a reporting exercise bolted onto the real work. Done properly, it's one of the clearest signals a firm has about where its time and attention are actually going.

The bottom line

A pipeline that's allowed to drift will always look busier than it is. Tight stage definitions, honest stage-to-stage conversion tracking, and a routine for catching stale deals turn a list of hopeful conversations into something a principal can actually plan around. It's not glamorous work, but it's the difference between a forecast and a guess.