Almost every firm we work with can tell us how many investor meetings its business development team held last quarter. Very few can tell us what those meetings changed.
That gap is not a failure of discipline. Meeting counts survived because they were the only thing anyone could reliably count. Allocator intent lived inside an allocator's head, and heads are not instrumented. So the sector settled on a proxy, built Monday morning reviews and a good deal of variable compensation around it, and quietly accepted that the proxy measured effort rather than progress.
The argument for changing this is not that activity metrics were always wrong. It is that the reason for tolerating them has gone.
A meeting count measures the seller, not the buyer
Activity metrics describe your own behaviour. An allocator who takes a courtesy call because a placement agent asked for one, and an allocator who has started modelling your fund into next year's pacing plan, both register as a single meeting. The metric is not slightly imprecise. It is measuring a different subject entirely: how hard your team worked, which is genuinely useful for managing people, and nothing at all about whether capital is moving towards you.
The second-order effect is worse than the measurement error. Anything attached to a review cycle gets optimised, and meetings are easy to optimise. It is far simpler to arrange a catch-up with a supportive existing LP who re-upped last year than to open a difficult conversation with a new insurance allocator who has never backed your strategy. Both count as one. Late in a quarter, the incentive points clearly at the easier one. We have sat in enough pipeline reviews to know that the list of meetings held in the final three weeks of a quarter tends to look different from the list held in the first three.
None of this makes business development teams cynical. It makes them responsive to the instrument they are being read by, which is what any competent professional does.
Intent was invisible, so the sector measured motion instead
The usual remedy has been the CRM stage field, and it has a structural problem: the person who updates the stage is usually the person compensated on movement through it. Self-reported probabilities inherit the same weakness. A relationship sits at sixty per cent for four quarters, and everyone in the room knows that sixty per cent means the partner still likes the LP.
Meanwhile the actual evidence of intent has always existed. It sits in the email threads, the meeting notes, the data room access logs, the consultant correspondence, the questionnaire returns, the side letter discussions. The record was not missing. It was unstructured, scattered across five systems and two inboxes, and reading a few hundred relationships' worth of it every week was not a job any firm was going to staff. Reading was the cost, so nobody read, so motion became the measure.
It is worth being clear that this was a reasonable settlement for as long as it lasted. A proxy that everybody understands and nobody trusts completely is still better than no instrument at all, and plenty of funds have been raised on the back of one. The settlement only becomes indefensible once the cost that justified it falls away.
The signals of progression were always in the record, just not readable at scale
What distinguishes a real allocator conversation from a polite one is not subtle. It is simply buried. The things experienced fundraisers watch for are concrete and, in principle, observable in the correspondence:
Who joined the second call, and whether their seniority went up or down.
Whether the questions moved from strategy and track record towards terms, governance and capacity.
Whether an operational due diligence name appeared on a thread without being invited.
Whether the questionnaire sent to you was a template or genuinely bespoke to your strategy.
Whether they asked about a named portfolio company rather than the portfolio in general.
Every one of those lives in material the firm already holds. What has changed is that reading the whole record consistently, every week, across the entire relationship base, is now a tractable task rather than an aspirational one. Once it is tractable, you can score progression on allocator-side evidence instead of on your own team's stamina, and the two numbers turn out to tell very different stories.
What you would measure instead, in illustrative terms
Consider a mid-market buyout manager raising a fund of around 750 million dollars over eighteen months. These figures are illustrative, constructed to show the shape of the argument rather than drawn from any particular fund. Suppose the team holds roughly 600 meetings across 240 allocator relationships. That is the activity view, and it is the number most firms would report to their management committee.
Now take the progression view of the same raise, again illustrative rather than measured. Of the 240 relationships, 95 never advance past a first conversation. Sixty stall somewhere in strategy discussion. Fifty reach operational due diligence. Thirty-five go to an allocator investment committee. Twenty-two commit. The interesting number in that sequence is not 600. It is the fifteen relationships that reach an allocator IC and do not convert, because that is the only step in the funnel where the firm was close enough to learn something precise about why it lost.
If, on inspection, nine of those fifteen fell out on the same objection, say a concentration limit or a question about the size of the GP commitment, then the firm has found something a meeting count could never have surfaced. It has also found something it can act on before the next fund rather than after it. And it changes what a good quarter means. A team that held forty fewer meetings but moved eight relationships from diligence into allocator IC had an unambiguously better quarter, and under the old instrument would have looked worse.
Any measure attached to compensation will be gamed, and signal-based measures are no exception
The strongest objection to all of this is not that the signals are unreadable. It is that once you score them, people will manufacture them. If an operational due diligence contact on a thread raises a relationship's score, threads will acquire operational due diligence contacts. If a bespoke questionnaire counts for more than a template, someone will ask an allocator to send one they were not planning to send. The failure mode is precise: you end up scoring the artefact rather than the intent it was standing in for, which is exactly the trap meeting counts fell into.
There is a second objection with more mathematical force. Fundraising is a small-sample activity. A firm might have a few hundred live relationships and a couple of dozen commitments per fund. Patterns fitted to that much data will be substantially made of accident. A probability estimate carried to two decimal places on a base of twenty-two commitments is false precision dressed as rigour, and any instrument that produces one should be treated with suspicion.
The third objection is the honest one about what stays invisible. A chief investment officer's private reservation about your succession plan, an allocator's internal politics, the fact that they are already over-allocated to your strategy and cannot say so without giving something away. None of that appears in the correspondence, and some of it is decisive.
What follows from these objections is not that the old metric was fine. It is that the new one needs handling with discipline. Report progression in bands rather than decimals. Keep the senior partner's qualitative override, but require it to be written down, so that in two years the firm can look back and see who was right, the instrument or the judgement. And resist wiring the score directly into variable compensation, at least at first. Its job is to direct attention, not to settle it.
What changes when the pipeline stops flattering everyone
The first quarter of honest pipeline reporting is uncomfortable. Relationships that have been carried for years as live turn out to be dormant, and someone has to say so out loud. After that, it becomes something closer to a relief. Senior time gets pointed at the twenty conversations that are genuinely in motion rather than spread thinly across two hundred. Finance gets a close forecast it can plan against instead of a number that drifts. And business development people, who have spent a decade being assessed on how much they travelled, get defended by evidence for the first time.
The best fundraisers we know have always run a version of this model in their heads. They knew which conversations were real, which were courtesy, and which were a year away from being either. What changes is that the model stops being private. It becomes something the firm can inspect, argue with, improve and, importantly, keep when the person who built it moves on. The work was never to hold meetings. It is encouraging that the measure is finally catching up with the job.
