Run IMM from due diligence to LP reports on one platform
Impact measurement and management (IMM) is the practice of identifying the social and environmental effects a portfolio intends to create, measuring those effects against stated commitments, and using the results to steer decisions — capital allocation, investee support, and reporting — through the life of the fund. Measurement produces the evidence. Management is what the fund does with that evidence between reports, which is where most IMM programs quietly stop.
The practitioner version is blunter. Every investee reports in its own format, on its own calendar, against its own definitions. Someone on the team spends three weeks a quarter assembling the same LP letter from six sources. The fund can produce an annual report; what it cannot do is answer a portfolio question on an ordinary Tuesday.
IMM stands for impact measurement and management. In impact investing, IMM is the discipline — standardized by bodies such as the GIIN and Impact Frontiers — of measuring a portfolio’s social and environmental results and managing investments with that evidence, from due diligence through exit. Sopact treats IMM as one continuous record per investee rather than a yearly reporting exercise.
The abbreviation collides with unrelated terms — IMM dates in banking, immunization records in medicine. Here, the full form of IMM is impact measurement and management.
Key takeaways
Sopact calls it the Tuesday question: the mid-cycle ask — which investees are behind on their year-2 outcome commitments, and what changed — that arrives on an ordinary Tuesday, months before any report is due. A stack tuned for the annual report answers it in weeks, because answering means archaeology: dig out last quarter’s exports, re-key the numbers, reconcile definitions that never matched. A stack tuned for the Tuesday question answers it the same afternoon.
The difference is the data model, not the dashboard. Legacy IMM tooling is report-centric: each quarterly submission is a document that gets filed, and the connection between an investee’s diligence promises and its year-3 results lives in an analyst’s memory. Sopact Sense is record-centric. Every investee carries one longitudinal record, under a persistent ID, that inherits every stage — the diligence memo, the signed impact agreement, every quarterly submission, the exit report. That is Sopact’s One Bet: when every document, metric, and narrative lands on one record, portfolio analysis becomes a query instead of a project.
Fund teams describe the before state in almost identical words: consistency is impossible across investees, and every LP-letter number must be defended by whoever remembers where it came from. The after state is the one Sopact designs for — the cleanup is the work; the LP letter is the residue.
Era one was the data call: Excel templates emailed to investees, returned late, reconciled by hand. Era two added software around the same model — survey platforms such as SurveyMonkey and Qualtrics for collection, portfolio trackers and CRMs such as Salesforce or Airtable for storage, BI layers such as Power BI or Tableau on top. These systems store and sum. The documents that carry the actual evidence — quarterly narratives, board letters, site-visit notes — sit attached and unread, so the dashboards look current while the understanding is a year old.
The current era reads on arrival: each submission is parsed the day it lands, mapped to the fund’s dictionary, and joined to the investee’s record. The one evaluation test that separates the eras: ask a vendor to show one investee’s diligence commitment and its year-3 quarterly submission on one screen, with the variance computed and the investee’s own narrative beside it. Era-one and era-two tools detour to a dashboard. If the demo cannot survive that question, the LP letter will not survive it either.
The data dictionary is the operable artifact of IMM: the field-by-field definition of everything investees report — the exact wording, the unit, the wave, the owner, and the framework each field maps to. A framework you cannot query is a PDF; a dictionary is what makes Five Dimensions, IRIS+, or a theory of change computable against real submissions. Sopact’s guidance is to start from the dictionary, not the framework — and the right dictionary depends on the fund. The build steps are in how to build a data dictionary; the SROI variant starts from an SROI value map.
| Dictionary flavor | Built from | Best fit | What the roll-up yields |
|---|---|---|---|
| Custom fund-based | The fund’s own theory of change and each fund’s theme; fields named in the fund’s language | Multi-fund portfolios where each fund runs a different theme — climate, workforce, health — and no LP mandates a public standard | Outcome variance per investee against fund-specific commitments; themes comparable within each fund |
| IRIS+ / IMP-aligned | IRIS+ metric IDs and the Five Dimensions — What, Who, How Much, Contribution, Risk — bound to each field | Funds whose LPs expect comparability against a public standard across managers | A Five Dimensions profile per investee and a portfolio roll-up benchmarkable beyond the fund |
| SROI value-map | Stakeholder outcomes, financial proxies, and deadweight and attribution estimates from the value map | Funds and foundations that owe a monetized social-return figure | A blended portfolio SROI with every proxy and assumption stated and traceable |
The flavors mix: a common pattern is a custom core dictionary with IRIS+ IDs tagged onto the fields where LPs want comparability, and an SROI value map layered on the sleeve that owes a monetized figure. Proxy selection is covered in how to pick a financial proxy, and the portfolio-level ratio in how to calculate a blended portfolio SROI.
Run IMM as six stages on one investee record: score the deal in diligence, lock the dictionary at onboarding, collect clean every quarter, grade against commitments, learn mid-cycle, and report as views over the record. Each stage card below shows today’s practice, where it breaks, and the same stage running on Sopact’s Loop — the same lifecycle a fund runs on due diligence software and portfolio monitoring software, connected instead of adjacent.
Sopact’s principle for the reporting stage: every report is a view. The LP letter, the board pack, the compliance annex, and the fund’s public impact report are not documents someone assembles — they are queries over the same audited investee records, each figure citing the submission it came from. Change the underlying record and every view updates; question any figure and the source is one click down.
The same records emit mid-cycle outputs no assembled report can: a Missing Data note when an investee’s submission skips a committed field, and an Unusual Insight when a narrative contradicts its own KPIs. Those two outputs turn measurement into management: they arrive while the fund can still act.
The design point is the mid-market impact fund: roughly 10 to 80 portfolio companies and $50M to $2B AUM, a small team, and LP reporting owed on a fixed calendar. Honest boundaries below.
| Scenario | Fit | Why |
|---|---|---|
| Mid-market impact fund, 10–80 portfolio companies | Strong | The design point: a multi-stage lifecycle per investee, quarterly submissions, LP and board reporting owed |
| Foundation or DFI with a grant or investment portfolio | Strong | Many funded relationships to aggregate into one defensible board figure |
| Fund of funds and blended-finance vehicles | Strong | Roll-up across vehicles with thin-data estimates flagged as estimates |
| A single program measuring its own participants | Not this page | That is participant-level work — start from impact measurement instead |
| A purely financial book with no impact thesis | No | There is nothing to measure; a standard portfolio tracker suffices |
| A one-off impact report for a deadline | Weak | The Loop pays back as a weekly habit; a single deadline does not amortize a dictionary |
IMM fails as an annual ceremony and works as a cycle. That is the premise of the Loop, Sopact’s method for continuous impact intelligence: collect clean at the source, analyze the moment a submission arrives, improve while capital and attention can still move.
The Loop is also what makes the LP number defensible. Every figure traces to the exact submission it came from — the standard described in traceability and transparency — and thin data is estimated honestly, with assumptions stated, per how to estimate impact with thin data.
One method, three moves that never stop
Then the cycle runs again, a little sharper each quarter. Read the method: the Loop methodology →
The fastest way to test the model is to run one slice of your own portfolio through it. Each prompt below is written for Sopact Sense’s Assistant; the arrow above each links the Academy walkthrough with the expected output and tips.
Academy walkthrough → How to build a data dictionary
Build a fund-level data dictionary from these sources: [PASTE FUND THESIS + 3 INVESTEE KPI LISTS]. For each field give the definition, unit, wave, owner, and framework mapping (custom, IRIS+ ID, or SROI value-map line). Flag every field where two investees define the same term differently.
Academy walkthrough → Onboard a portfolio and lock the impact agreement
Draft the impact agreement for [INVESTEE]. From our dictionary [PASTE], select the fields this investee will report quarterly, set year-1 baselines from their diligence data [PASTE], and list the evidence each commitment will require at review.
Academy walkthrough → How to calculate a blended portfolio SROI
Using these investee outcome records: [PASTE OR ATTACH], compute a blended portfolio SROI. State the proxy for each outcome, the deadweight and attribution assumptions, flag thin-data investees as estimates, and show how sensitive the ratio is to each assumption.
Academy walkthrough → Write a cited impact narrative
Draft the impact section of our year-[N] LP letter from these records: [ATTACH]. Cite every figure to its investee submission, include one Missing Data note and one Unusual Insight, and keep measured results separate from benchmarked estimates.
Each walkthrough is practical and short: what to do, the prompt to run, the output to expect, and the tips that make it reliable. Start with the Portfolio Intelligence course for the full arc.
Watch: impact measurement and management on one connected investee record, diligence to reporting.
IMM stands for impact measurement and management. In impact investing it is the discipline of measuring a portfolio’s social and environmental results and managing investments with that evidence, from due diligence through exit. Sopact implements IMM as one longitudinal record per investee rather than an annual reporting exercise.
Impact measurement and management is the practice of defining the effects a fund intends to create, measuring them against commitments locked at onboarding, and steering capital and investee support with the results. Sopact’s framing: management means answering the Tuesday question mid-cycle.
Impact measurement establishes what changed: outcomes against baselines and commitments. Impact management is acting on that evidence between reports — reallocating support, correcting drift, renegotiating targets. Sopact’s test is timing: if the insight arrives only in the annual report, the fund measured but never managed.
An IMM framework is the lens a fund uses to classify impact — the Five Dimensions of Impact, IRIS+ metrics, or the fund’s own theory of change — and the frameworks are complementary, not competing. Sopact’s guidance: a framework becomes operable only when a data dictionary binds each of its concepts to a defined, queryable field.
No. IRIS+ buys cross-fund comparability when LPs expect a public standard, but many multi-fund portfolios run custom fund-based dictionaries matched to each fund’s theme, and SROI-driven funds start from a value map. Sopact supports all three dictionary flavors and lets them mix on the same portfolio.
As six stages on one investee record: score the impact thesis in diligence, lock the dictionary and commitments at onboarding, collect clean quarterly submissions, grade variance against commitments, learn from cross-portfolio themes mid-cycle, and generate reports as views. Sopact keeps all six stages on one persistent investee ID, so no stage starts from zero.
No. Sopact Sense runs beside the portfolio tracker or CRM as the reading and evidence layer — an AND, not a rip-and-replace. Your system of record keeps ownership and pipeline; Sopact keeps every investee’s evidence on one record and pipes structured output onward.
Trackers and dashboards store and sum what was already structured; the evidence in quarterly narratives and reports stays unread. Sopact’s Intelligent Cell reads each document on arrival and its Intelligent Row grades each investee against commitments, so the dashboard-level number is backed by cited evidence instead of attachments.
Reliable enough only when every figure is traceable, which is the constraint Sopact builds in: each number in a generated report cites the investee submission it came from, thin data is flagged as an estimate with assumptions stated, and the reviewer approves the draft. The Loop’s traceability standard exists precisely because LP reporting is adversarial.
The dictionary comes first — typically a working session per fund to lock fields, definitions, and framework mappings — then investees onboard against it and the first quarterly cycle runs clean. The design point is a mid-market team of one to three people covering 10 to 80 portfolio companies without adding headcount.
Next: see the live roll-up and blended SROI in portfolio analytics for impact, or start one stage earlier with due diligence software.
