Build a pipeline backwards. Start from the revenue you need, work out how many conversations that implies, and build the list to fit. Most people start with the list, which is why pipelines end up either too thin to hit the number or full of companies that were never going to buy.
The arithmetic takes twenty minutes and it decides everything that follows.
Step one: the arithmetic
Four numbers, worked in this order.
- Revenue needed. For the period you are planning, minus whatever is already contracted.
- Average deal size. From your own closed deals, not the ones you wish you had.
- Win rate. Of opportunities that reached a real stage, what proportion closed.
- Conversation to opportunity rate. Of first conversations, how many became an opportunity.
A worked example. You need 300,000 dollars. Average deal is 15,000, so that is 20 wins. You win one in four, so you need 80 real opportunities. One in five first conversations becomes an opportunity, so that is 400 conversations.
That last number is the one that changes behaviour. Four hundred conversations is a different plan from "do some outbound", and it is the first honest answer to whether the target is reachable with the team you have.
If you have no history yet, use conservative placeholders and replace them with real numbers after the first thirty deals. A placeholder you intend to replace is fine. A placeholder nobody ever revisits becomes a forecast.
Step two: define who, before you build the list
The temptation is to go wide. Wide lists feel like progress and convert badly, because the message has to be generic enough to fit everyone on it.
Three filters are usually enough: the industry or segment where you have won before, the size band where your pricing makes sense, and a trigger that suggests now rather than someday. The trigger is the one most people skip and it is the one that moves reply rates: a new hire in the relevant role, a funding round, an office opening, a technology change, a regulatory deadline.
Step three: build the list to the size the arithmetic asked for
Now you know you need 400 conversations, the list length is a calculation rather than a judgement. If one in ten contacted people has a conversation, you need 4,000 contacts, which tells you whether this is a quarter of work or a year of it before you have spent a day on it.
This is also the moment to be honest about reach. A list of companies is not a list of people, and a list of people without a way to reach them is a document. Coverage, meaning the proportion of your target market where you have a named contact you can actually reach, is the constraint that quietly caps most pipelines.
Step four: one record, from the first touch
Every conversation goes on the record from the first one, including the ones that go nowhere. Two reasons, and the second is the important one.
The obvious reason is follow-up. The real reason is that the rates in step one only get accurate if the failures are recorded too. A pipeline that only contains deals that became real cannot tell you your conversation to opportunity rate, which means next quarter's plan is built on the same guess as this one.
Step five: review what is ageing, not just what is new
A new pipeline grows for the first month and then starts to rot quietly. The measure that catches it early is the proportion of deals that have changed stage in the last thirty days. If most of the pipeline has not moved, the total at the bottom is not a forecast.
Build the review in from week one, while the pipeline is small enough that reviewing it takes ten minutes. Teams that start reviewing when the pipeline is large never start.
The part that is genuinely hard
Not the arithmetic and not the list. It is that pipeline building pays out one sales cycle later, so the work you do today shows up after the quarter you are worried about. That is why it gets started in response to a bad month and why it is always late.
The only real fix is to make it continuous rather than a project, which means a small amount every week rather than a push every quarter.
Empiraa Signal is a CRM and more. It is also the B2B prospect database, the enrichment, the email sequences, the booking link and the proposal tool, in one system with unlimited users on every plan, so finding the list, reaching it and recording what happened are the same system rather than three.
Related reading: sales pipeline stages, and what probability to put on each, and how to write a sales email that gets a reply.
Sales pipeline has a definition on this site.
