The Sourcing Capacity Ceiling

Deal flow is the one number almost every sourcing team can quote without thinking.

How many opportunities arrived this quarter, how many sit in the pipeline now, how this year's funnel compares to last. It is the metric the industry has agreed to care about, and there is a reason for that.

The funnel really is wide, and deliberately so. A team might review scores of opportunities, take perhaps ten or fifteen to a preliminary look, three to five into genuine diligence, and close one or two of those. When the odds look like that, the instinct is easy to follow. Widen the top. See more, and you will close more.

It holds only while something else can grow alongside it, though. And the thing that has to grow, a team's capacity to actually look at a deal properly, does not scale the way the funnel does. There is a point past which more opportunities stop producing more good decisions and quietly begin producing worse ones. That point is the sourcing capacity ceiling, and a lot of firms run clean through it without noticing, mostly because nothing on the dashboard is built to tell them they have.

What actually runs out first

It is worth being precise about what capacity is, because the easy assumption is that it means hours, and it does not, or not only. The real constraint on a deal team is the supply of considered attention: the senior judgement available to read a situation properly, to ask the second and third question, to notice the line in the accounts that does not quite sit right. That supply is finite in a way that headcount and working hours tend to disguise. You can always add another name to the pipeline review. What you cannot do is hand another unit of real attention to a partner who is already carrying six live processes in their head.

This is where a quiet piece of behavioural research earns its place. Sophie Leroy's work on what she called attention residue found that when people move from one task to another, a part of their attention stays behind on the first, and that residue measurably weakens performance on whatever comes next, the effect being strongest when the earlier task was left unresolved.

That describes the daily condition of a sourcing team almost exactly: many open threads, few of them ever quite put to bed. A partner holding eight half-formed views on eight targets is not giving each one an eighth of their attention. Each is getting a fraction, and a fraction already degraded by the residue of the other seven. The pipeline looks full. The thinking underneath it is thinner than anyone means it to be.

The funnel is effectively infinite. The capacity to read it well is not. A great many of the sourcing problems worth solving live in the gap between those two facts.

Why the ceiling stays invisible

The awkward part is that reaching the ceiling does not feel like reaching anything. There is no warning light. A team at capacity looks, from the outside, simply like a busy team, which is exactly what everyone hopes a sourcing team looks like. The cost arrives indirectly and on a delay. It usually surfaces as one of a few quiet things:

• Diligence that goes a little shallower than it should.
• A reply to a promising intermediary that takes four days rather than four hours.
• The genuinely strong opportunity that gets a competent but slightly distracted look, because three mediocre ones were taking up the room that week.

None of these read as failures at the time. The deal that did not get enough attention rarely puts its hand up to say so; it just converts at a lower rate, and the loss gets quietly filed under a competitive process or bad timing.

So a firm draws the natural conclusion from a flat quarter, which is that it needs more coming in. More flow, more coverage, more sources. And because the funnel will accept whatever you pour into it, the strategy rarely fails in any visible way. It simply keeps adding weight to a beam that is already bending. (There is an honest version of the opposite error, worth naming. A firm can also sit well below its ceiling, too cautious, passing on things it had the room to pursue. That is a real cost too. It is just the rarer one, and it is not the direction the industry's incentives tend to push.)

Capacity is mostly a question of what you refuse

If the ceiling is real and more or less fixed in the short run, then the lever that matters is not how much arrives. It is how little is allowed to occupy active attention at any one moment. This is the same place the rest of this series keeps landing, approached from a different side, and it is worth saying why.

‍Saying no earlier, qualifying before the model gets built, treating the move from watching a target to pursuing it as a decision rather than a drift: these usually get framed as questions of discipline, as if the missing ingredient were willpower. Underneath, the thing they are really managing is more mechanical. They are how a team rations a scarce resource, so that the deals which do get attention get enough of it to count.

Read that way, a pass stops being a missed opportunity and becomes attention handed back to the pool, freed up for something with a better claim on it. A firm that guards its capacity is not being timid. It is declining to let plausible but unexamined targets quietly tax the evaluation of the ones that genuinely deserve the work. The aim was never a smaller pipeline for its own sake. It is a pipeline matched honestly to the attention there is to work it with.

Raising the ceiling, not only defending it

There is a second move, and it happens to be the one that data and tooling can actually help with. If the binding constraint is the attention each deal consumes, then anything that lowers the cost of a serious early read effectively lifts the ceiling. Most of what a team needs for a sensible first judgement already exists and is reachable before anyone commits real time:

• The shape of a company's numbers across several years, rather than a single one.
• The ownership and group structure, once it is properly mapped.
• Who actually holds the decision on the other side.
• Whether there are distress or regulatory signals already visible.

When that picture takes an afternoon instead of a fortnight, each opportunity draws down less of the scarce resource, and the same team can hold more in genuine view without spreading itself thin.

That is the less obvious reason front-end intelligence matters, and it is a different argument from the usual one. It is not only that better information makes for better decisions, true as that is. It is that cheaper early information changes the arithmetic of capacity itself. A target you can assess properly in hours simply costs less of the thing the whole team is short of.

The number nobody puts on the dashboard

None of this is an argument for sourcing less ambitiously. It is an argument for being honest about a constraint the standard metrics are designed, almost by accident, to hide. Deal flow is easy to count, which is part of why it gets managed so attentively, and capacity is hard to see, which is part of why it gets spent so freely. A firm can widen its funnel more or less indefinitely and feel productive the whole way, right until the quality of its attention has been stretched so thin that its best opportunities are getting the same half-present glance as its weakest.

The firms that compound well across a cycle tend not to be the ones with the widest funnel. More often they are the ones that worked out, sooner than their peers, that the scarce resource was never the supply of deals. It was the room to think clearly about them.

‍

Further reading

This blog is part of a series on how deal teams make better pursuit decisions.

• The Deal That Wasn't Worth It: what late-stage collapse really costs, and why the most expensive deals are often the ones that never close.
• What Makes a Deal Worth Pursuing?: the front-end framework that separates genuine opportunities from merely plausible ones.
• The Two Sourcing Mindsets: why the firms winning on deal economics tend to be more deliberate at the front, rather than better at the back.