Institutional LPs do not choose between managers based on stories. They choose based on comparable numbers, and the three numbers they compare are XIRR, TVPI and DPI. Emerging managers who report these cleanly, on time, and with the underlying schedule attached, win re-ups. Managers who submit narrative-heavy PDFs with the numbers buried inside lose them.
This piece is a working definition of each metric, the mistakes that most commonly distort them, and the reporting practices LPs actually notice.
XIRR: the time-weighted answer
XIRR — extended internal rate of return — is the annualized return that discounts every fund cash flow, positive and negative, back to zero. It is the metric that answers 'per unit of time, how well did this fund do'.
It is also the most easily distorted metric in venture. Because XIRR compounds, an early markup at a high valuation can inflate the number well beyond what the fund will ultimately deliver. Emerging managers who report gross XIRR from unrealized markups without also reporting DPI look sophisticated for a year and then get discounted heavily when the markups compress.
The right practice is to report XIRR on realized plus current NAV, alongside DPI, and to explain the drivers when either moves meaningfully. LPs prefer honest thirty percent XIRR with a clear source to unexplained sixty percent XIRR that could be paper.
TVPI: the size of the pie
TVPI — total value to paid-in — is the multiple of the fund. If LPs put in a dollar and the fund holds two dollars of value (realized plus unrealized), TVPI is two.
TVPI is easier to interpret than XIRR because it is unitless and time-agnostic. It is also what LPs use to compare across vintages, because a two-and-a-half TVPI on a 2018 fund and a two-and-a-half TVPI on a 2022 fund mean different things about the manager.
The reporting mistake here is quietly changing the mark on a position without explaining why. If a position moves from 1x to 3x on a new round, the memo footnote should name the round and the price. LPs who cannot see the driver assume the worst.
DPI: the metric LPs actually trust
DPI — distributions to paid-in — is the multiple of cash returned. It is the only performance metric that cannot be inflated by mark-to-model. It is the metric institutional LPs discount the least and pay the most attention to.
For an emerging manager whose fund is early in its life, DPI is typically zero for years. That is not a problem, as long as the reporting says so honestly and shows the path to distributions — expected exits, secondaries under discussion, tender programs in flight.
The trap here is confusing distributions with realizations. A tender offer that lets the manager sell down a position is a distribution. A dividend recap is a distribution. A structured secondary is a distribution. Managers who leave DPI at zero because 'we have not had an exit yet' are underreporting the metric that LPs care about most.
The manager who reports DPI honestly and quarterly, even when it is small, builds more LP trust than the manager who reports a headline TVPI once a year.
The commitment schedule LPs quietly want
Alongside the three metrics, LPs want a clean commitment schedule. Called, uncalled, distributed, current NAV, per fund, per LP. Managers who deliver this as a schedule LPs can drop into their own model — not as a PDF they have to retype — get faster re-ups.
This is one of the highest-return operational improvements an emerging firm can make. The cost is low and the LP-side impression is disproportionately positive.
The J-curve and how to talk about it
Every venture fund starts underwater. The management fee drag and the initial write-downs mean TVPI and DPI look ugly for the first two to four years. This is the J-curve, and it is completely normal — but only if the manager explains it.
The best emerging managers report the J-curve honestly in their first two years, name the fund's expected inflection point, and then hold themselves to it. LPs who see a manager forecast and hit an inflection are much more likely to lead the next fund than LPs who see a manager mask the J-curve.
Attribution: the underrated LP question
'Which two or three positions are driving your TVPI?' is the question most LPs will ask on the third call and most managers are underprepared for. The answer requires a per-position markup schedule, current ownership, current valuation and contribution to gross return.
Managers who can produce this in a table on the spot look institutional. Managers who cannot look sub-scale.
Where Equavion fits
Equavion's returns engine computes XIRR, TVPI and DPI live from the fund's actual cash flows and current holdings, per fund and per LP. The commitment schedule is a live object that LPs can be granted read access to. Attribution is one click — every position's contribution to gross return is computable against the live graph. Emerging managers get to spend their LP calls talking about strategy, not reconciliation.
Takeaways
- LPs compare managers on XIRR, TVPI and DPI. Report all three cleanly and on time.
- DPI is the metric that cannot be inflated. Treat structured secondaries and tenders as distributions when they are.
- Ship a machine-readable commitment schedule alongside the report. It costs little and impresses LPs disproportionately.
- Explain the J-curve, forecast the inflection, hit it. Be ready to attribute your TVPI on demand.
