How to build a referral partner program that partners use
Revenue share alone rarely motivates. What referral partners actually want, the disclosure rules you are responsible for, and why three beats thirty.
Read
If a deal's stage does not change your confidence in it closing, the stage is decoration and the forecast is a guess.
Most B2B companies do not have a pipeline. They have a list of people they have spoken to, sorted roughly by how encouraging the last conversation felt. That is not the same thing, and the difference shows up the first time someone asks what will close this quarter.
A pipeline is a model. It takes what you know about deals in progress and turns it into a number you can plan against. If it cannot do that, it is a contact list with optimistic formatting.
The first move is not choosing a CRM. It is writing down what has to be true for a deal to enter the pipeline at all. Without that, every conversation qualifies, the number inflates, and the forecast becomes a work of fiction that everyone privately discounts.
Ours is deliberately unglamorous: the right role is involved, a specific problem has been named by them rather than by us, and there is an agreed next step with a date on it. Miss any one and it stays out. That third condition removes more deals than the other two combined, which is exactly why it earns its place.
Stages exist so that a deal's position implies a probability. If two stages carry the same real chance of closing, one of them is decoration.
You do not have to invent this from nothing. HubSpot ships a default sales pipeline and publishes the probabilities attached to it, which is a useful reference point precisely because it is not ours: Appointment scheduled at 20%, Qualified to buy at 40%, Presentation scheduled at 60%, Decision maker bought-in at 80%, Contract sent at 90%, then closed won at 100% and closed lost at 0%.
Two things are worth taking from that list. The first is that the gaps are large. Twenty points between stages means moving a deal forward is a real event, not a status update. The second is that the documentation is explicit about what the probability is for: the weighted amount shown in the pipeline is the total in each stage multiplied by that stage's probability. The percentages are not commentary, they are arithmetic that produces the forecast.
Copying a default pipeline wholesale is a mistake, because your stages should describe what actually happens in your sale. The order to build it in:
This is the part of an engagement we do first, before any outreach, and it is why our deliverables start with targeting and definition rather than volume. Five stages is plenty for most companies under fifty people. Seven is the point at which people start guessing which one a deal is in, and a stage that gets guessed produces a forecast that is worse than having no stages at all.
A stage nobody can name the exit criterion for is not a stage. It is a place deals go to wait.
Once deals sit in stages, three numbers do all the work, and none of them require a tool you do not already have.
None of the three needs new software, which is worth saying because the instinct is to solve a definition problem with a purchase. The examples on our homepage are companies that fixed this with a spreadsheet and a weekly review before anything else. That last one is where most of the false comfort lives. Clearing out stalled deals makes the pipeline number drop, which feels like a bad week and is actually the first accurate week you have had.
Structure is not a substitute for having something worth buying, and a well-kept pipeline that keeps stalling at the same stage is not a discipline problem. It is telling you something about the offer or the audience, and reorganising the columns will not change the answer.
It also will not fix the underlying issue we wrote about in most B2B companies do not have a business development problem: if four people in the company describe the process differently, a shared pipeline just gives them a shared place to disagree. Agree the process first, then model it.
One honest limitation of everything above: it measures deals that already exist. If not enough conversations are starting, no amount of stage hygiene produces a forecast worth reading, and the constraint is demand rather than pipeline management.
That is a different discipline with different measurement, and the long sales cycle makes it genuinely hard to attribute. KF Agency has a practical write-up on measuring marketing ROI across a long B2B sales cycle (in Arabic) that covers the attribution side properly, which is the half a pipeline cannot see.
Do not buy anything. Export whatever list you have, apply the three entry conditions, and delete every deal that fails them. Then put a date against each survivor. The number you are left with is your real pipeline, and for most companies it is between a third and a half of what they were reporting.
If the number that survives is far short of the target, the next decision is a resourcing one rather than a process one, and we have written the honest version of that comparison in outsourced BD versus hiring a salesperson. Either route needs the pipeline definition above first, or you are handing an undefined process to a new person and calling it delegation.
That is not a bad discovery. It is the first quarter you can plan.
Send us the last twenty deals, won and lost. We will tell you where they actually die, and whether the problem is the pipeline or the demand feeding it.
Book a Free Consultation