Your pipeline is a leaky bucket, not a funnel

A funnel is a device for making liquid go somewhere it was already going. You pour, gravity does the work, and the only real variable is how much you pour in at the top. This is a lovely metaphor if you sell funnels and a terrible one if you sell anything else, because it quietly tells every sales team in the world that the answer to a bad quarter is more volume.

Your pipeline is a bucket. Buckets leak. And the interesting thing about a leaking bucket is that pouring faster is almost never the cheapest fix.

Where it actually leaks

Run the numbers on any pipeline and the losses cluster in three places, in roughly this order of size.

The biggest hole is between "responded" and "first real conversation." Somebody replies with interest, a meeting gets loosely proposed, and then it evaporates in the scheduling gap. Nobody logs this as a loss because it never became an opportunity. It just quietly is not there anymore. For most teams this leak is larger than every later-stage loss combined, and it is invisible in every dashboard because dashboards start counting at stage one.

The second hole is single-threading. A deal with exactly one contact has roughly a coin-flip chance of surviving any organizational disturbance, and organizational disturbance is the normal state of a company. The contact changes jobs, gets reassigned, goes on leave, or simply stops caring, and the deal does not get lost so much as it becomes unreachable.

The third hole is the late-stage silence that follows a proposal. Everything was fine, the proposal went out, and then nothing. This one gets attributed to price about eighty percent of the time and is usually not price. It is usually that the proposal went to somebody who then had to sell it internally without you, using materials designed to persuade them rather than materials designed to persuade their boss.

Plugging them costs almost nothing

The scheduling gap closes with a calendar link and a same-hour reply. That is the whole fix, and it is worth more than any amount of additional prospecting. If a prospect says "sure, let's talk" and you answer four hours later with three proposed times, you have introduced a decision into a moment that did not require one. Answer immediately, propose one specific time, attach the invite. The response rate difference between "does Thursday at 2 work" and "here are some times that might work" is not subtle.

Single-threading closes with one habit: never leave a good call without asking who else touches this. Not "who else is involved in the decision," which sounds like you are trying to route around them, and people hear that. Something softer that gets the same map: "when this goes well, whose world does it change most?" That question gets you a name and a reason in one answer, and it makes the second contact a favor to the first rather than a betrayal.

The proposal silence closes by giving your champion the internal version, not the external one. A one-page summary written for the person they have to convince, in the language that person uses, with the number in it. Most reps send a beautiful deck to a champion who then has to translate it, and translation is where deals die quietly.

The part that stings

Here is the uncomfortable implication of the bucket model. If your pipeline leaks at these three points, then increasing top-of-funnel volume increases your workload proportionally and your revenue somewhat less than proportionally, because you are pouring into the same holes.

This is why "just do more activity" is such durable advice despite working so poorly. It is measurable, it feels virtuous, and it produces a small real improvement, which is enough to keep everyone believing in it. Fixing a leak produces a much larger improvement and is nearly invisible, because the deals that did not fall out never appear anywhere as a saved deal. Nobody gets a certificate for the meeting that did not get lost in a scheduling gap.

You can only find leaks by looking at where things stopped, which means keeping records of stalls and not just outcomes. A CRM that only knows won and lost cannot tell you anything about your bucket. A CRM that knows "responded, never scheduled" as a distinct state can tell you almost everything, and it costs one extra field.

The sales industry generates metaphors at a rate that outpaces its ability to test them. The funnel, the flywheel, the journey, the engine, the machine. Most of them share a feature, which is that they imply the process runs on its own once configured, and they are popular for exactly that reason: a system that runs itself is a much nicer thing to sell software for than a bucket somebody has to keep patching.

The bucket is less inspiring and considerably more accurate. It also has the advantage of suggesting an action that works, which is to go find the hole rather than to go find more water.

Look at where deals stop, not just where they end.

— DealArena Team