Impact investing due diligence that reads every impact report and claim on arrival - and surfaces the claim the evidence cannot support, before an LP asks.
Sopact is the risk-intelligence layer for impact investing due diligence. It reads every report, metric, and impact claim a fund or portfolio company submits — the moment it lands — and surfaces the claim the evidence cannot support, before an LP asks how you know. It is built for the impact fund managers and LPs who have to defend every number.
Most impact investing due diligence is a screen: a scorecard run once, before the capital is committed. The impact claim it screens for has to hold for the whole holding period. Here is the same investment, run both ways.
A diligence scorecard is a point in time. It clears the deal on the thesis the fund was given — and goes quiet until an LP asks for the evidence.
Continuous diligence is a layer, not a screen. It reads every impact claim on arrival — so a weak one surfaces while the fund can still ask the portfolio company to fix it.
It is the same impact claim on both tracks. Point-in-time diligence meets it again at the LP’s question; continuous diligence reads it on arrival. The years between those two dates are the years an unverified claim sat in your impact report.
Impact investing due diligence is the process of examining whether an investment will create the social or environmental impact it claims — before the capital is committed, and for as long as it is at work. The weak version is a one-time scorecard run before the deal. The strong version reads every impact report and claim on arrival — the metrics, the theory of change, the beneficiary evidence — scores it against the evidence behind it, and keeps the diligence live across the whole holding period.
It is distinct from ESG due diligence. ESG diligence asks whether an investment carries environmental, social, or governance risk. Impact investing due diligence asks the harder question: is the impact it promises real, additional, and defensible.
An impact claim is rarely unsupported on its face. The evidence that would confirm it — or undermine it — has been submitted. Impact diligence reads two of these places reliably. The other four are collected and filed.
The numbers that lead the impact report — people reached, tonnes avoided, jobs created. Read closely, because they are the claim.
The structured impact KPIs the fund tracks each quarter. Counted accurately — and only ever as good as the method behind them.
The logic that connects the activity to the outcome. Where additionality and the unproven assumption actually live — reviewed once, at diligence.
The pages behind the headline number — the method, the caveats, the shortfall. Read for the highlights, then filed.
What the people the investment is meant to serve actually say. The most direct evidence an impact claim is real — and the one diligence rarely reaches.
The independent evaluation, the third-party data, the unflattering coverage. Public, consequential, and outside the fund’s own report.
The four unread sources are where an impact claim is confirmed or quietly undone. A diligence that reads only the headline metric is checking the claim against itself — not against the evidence.
An impact investing due diligence checklist covers three things: the impact thesis, the evidence behind it, and the fund or company that has to sustain it. The scope below is the standard. What decides whether the diligence works is whether each line was checked against evidence — or taken on the thesis.
These eighteen lines are the standard impact investing due diligence scope. But a checklist is a list of questions, not a verdict. The diligence is only as strong as the evidence read behind each line — and across a portfolio, most of that evidence is never read twice.
For years, impact investing due diligence was a scorecard. A fund screened a deal against an impact thesis, scored it, committed the capital, and collected an annual impact report. The report was read for the highlights and filed. Impact was a story the fund told, and it was largely taken on trust.
Trust got expensive. As impact capital grew, so did the scrutiny: LPs now ask funds to evidence every impact claim, regulators are moving against impact-washing, and a headline number with nothing behind it is a reputational liability, not an asset. The Operating Principles for Impact Management ask for independent verification. A scorecard run once, before the deal, cannot answer a question asked every year after it.
Meanwhile the evidence kept arriving — the quarterly reports, the beneficiary surveys, the evaluations, the news — and the scorecard had no way to read it. So the value moved. It is no longer in the diligence scorecard or the annual report. It is in the layer that reads every impact claim on arrival and checks it against the evidence. The scorecard era told an impact story. The audited era has to prove one.
This is not an argument that the diligence scorecard or the impact report is useless — they remain how a thesis is framed and a year is summarised. It is an argument that framing an impact claim and verifying it are two different jobs — and an LP now asks for the second.
Sopact is a risk-intelligence layer that reads what impact diligence already collects. It does not replace your fund administration or your data room. It reads the material the fund stores and never re-examines — the impact reports, the metrics, the theory of change, the beneficiary surveys, the evaluations — against the impact framework the fund defined, the moment each one arrives.
Three things happen on every impact report, in order. None of them waits for the annual review.
Every impact report, metric, theory of change, and beneficiary survey is read the day it lands — in any language it was written in, tied to one record per portfolio company. Nothing is filed unread.
Each claim is scored against the evidence behind it — the baseline, the method, the source, the beneficiary voice — on the impact framework the fund defined, with the source sentence kept behind every flag.
A standing view shows which claims are weakest, across the whole portfolio. The fund aims its independent audit where the evidence is thinnest — and the LP report is built from claims that have already been checked.
An impact report read at the LP’s question is a claim you now have to defend. The same report read on arrival is a chance to strengthen the claim, or correct it, first. The only variable is when it gets read.
A fund with forty portfolio companies cannot independently verify every impact claim every year. The question is not whether to sample — it is how to choose the sample.
You pick a handful of portfolio companies to audit each year — by rotation, by deal size, or at random. The deep audit goes wherever the schedule points, which is rarely where the weak claim is. The portfolio company with the thinnest evidence is audited in year four, if at all.
Every impact claim across the portfolio is read on arrival and scored against its evidence. The deep, independent audit is then aimed at the claims that scored weakest — the thin baseline, the metric with no method, the number with no source. The limited audit budget goes where the risk is.
Ask of any impact diligence process: how do you decide which claims to audit? If the honest answer is “rotation” or “the biggest deals,” the audit budget is being spent off the risk — not on it.
AI is now on the label of almost every diligence tool. Two paragraphs on what it genuinely changes, then the test.
What AI genuinely changes is the cost of reading impact documents — reports, theories of change, beneficiary surveys, evaluations — against a defined set of impact criteria. Work that took an analyst weeks of manual review now runs in minutes, and re-runs every time a new report arrives. That is the single change that makes continuous impact diligence possible across a real portfolio.
What AI does not change is where the reading has to sit. There is a real difference between asking a general AI to summarize an impact report and a layer reading each claim against your framework on arrival. Run the same portfolio company through a chat window twice and the impact rating drifts — a strong claim one day, a weak one the next — because nothing holds the definitions still.
You paste an impact report into a chat window and ask whether the claim holds up. It answers — once. There is no fixed definition of what makes a claim credible, no link from this report to the last, and no source sentence behind the rating. Ask again next quarter and the answer has moved.
The impact framework is defined once and held. Every report is read against that same definition, tied to one record per portfolio company, with the source sentence kept behind every flag. Run the same company in March and in June and the method is identical — what changed is the company, not the ruler.
Ask any AI diligence tool: run the same portfolio company twice, a quarter apart — does the impact rating hold, and can you see the sentence behind it? A locked answer is a claim you can put in an LP report. A drifting one is a guess with a logo.
An impact fund manager screening a deal, an LP screening a manager, a foundation answering to a board — different mandates, the same job: know the impact claim is real before someone asks you to prove it.
A diligence thesis before the deal, and an impact claim to defend every year after it — across a portfolio too large to audit by hand.
Diligence on the manager, not only the deal — does the fund’s impact management system hold up, and is its reporting evidence or narrative.
Diligence that has to answer to a board and a mission — is this investment additional, and is the impact worth the trade-off.
A GP, an LP, and a foundation run the same loop: an impact claim arrives, the evidence behind it has to be read, someone has to answer for it later. They differ on the mandate — not on where the weak claim hides, and not on what it costs to repeat it in a report.
Impact investing due diligence is not an improvised exercise. The field has built shared systems for what to measure, how to manage impact, and how to verify it — and diligence is the act of holding a claim up to them.
IRIS+, managed by the GIIN, is the generally accepted system of impact metrics. It gives diligence a common language for what a portfolio company should be measuring — and what its claims should be built on.
The Operating Principles for Impact Management set nine principles for managing impact across the investment lifecycle — including independent verification of the impact management process.
The Impact Management framework defines five dimensions — What, Who, How Much, Contribution, and Risk. Impact risk, the chance the impact does not occur as expected, is exactly what diligence is built to assess.
Sopact cites these frameworks to share their vocabulary — IRIS+ metrics, the five dimensions, the principle of independent verification — not to certify against them. The frameworks say what good impact looks like; diligence is the work of checking whether a claim meets it.
An impact investing due diligence platform is not a scorecard with a dashboard. It is the set of jobs that turn the reports a portfolio submits into a claim you can defend. Sopact runs six, in one place.
Collect impact data through Sopact, or read the reports and metrics a portfolio already submits. One record per portfolio company, from diligence onward.
Every impact report, theory of change, metric, and beneficiary survey read on arrival, in any language. Nothing is filed unread.
Each claim scored against the evidence behind it — the baseline, the method, the source — with the source kept behind every flag.
The headline metric and the evidence under it on one record — per portfolio company, across the whole holding period.
The same impact framework applied to every company and every cycle — so a portfolio is comparable and a trend is real, not a wording change.
An LP-ready impact report and a portfolio risk view, generated from the live record — a custom answer to an LP query without a month of analyst work.
Bring a real portfolio batch — a set of impact reports, metrics, and theories of change. We will run it through Sopact and show you the impact claims read against their evidence.
Most impact investing due diligence searches start with the wrong question. “Which diligence platform should we buy” returns a shortlist of databases and scorecards that all demo well. The useful question is narrower: walk one portfolio company from the diligence screen to its third annual impact report, and find the seam where the claim goes unverified.
If the scorecard clears the deal but the annual reports are never re-examined, the gap is reading. If every company’s impact is rated by a different analyst with a different definition, the gap is a locked framework. If diligence goes quiet between annual reviews, the gap is continuity. If the audit budget is spent by rotation rather than by risk, the gap is sampling. And if a headline number cannot be traced to the evidence behind it, the gap is exactly what an LP will ask for.
That diagnosis decides whether you need a better scorecard or a different layer over the whole portfolio. A fund that skips it buys a faster way to score a deal — and the impact report that held the weak claim is still sitting in the data room, read for the highlights and filed.
Take one impact claim your fund made in a past LP report. Ask of any tool you are evaluating: would this have shown you the evidence behind that claim — or the absence of it — before the report went out? If the answer is “only if an analyst had gone looking,” it scores impact — it does not verify it.
Impact investing due diligence is the process of examining whether an investment will create the social or environmental impact it claims — before the capital is committed, and for as long as it is at work. The weak version is a one-time scorecard run before the deal. The strong version reads every impact report and claim on arrival — the metrics, the theory of change, the beneficiary evidence — scores it against the evidence behind it, and keeps the diligence live across the holding period.
An impact investing due diligence checklist covers three areas. The impact thesis: a clear theory of change, intended outcomes and beneficiaries, additionality, alignment to IRIS+ or the SDGs, impact risk, and unintended effects. The evidence: a credible baseline, outcome metrics, a documented method, beneficiary voice, third-party verification, and a source behind each headline number. The fund and portfolio: an impact management system, impact-linked incentives, reporting quality, impact-washing safeguards, impact at exit, and transparent LP reporting.
Auditing impact claims means checking each headline impact number against the evidence behind it — the baseline, the data collection method, the beneficiary feedback, any independent evaluation — rather than accepting the number as reported. The practical problem is volume: a portfolio has more claims than an analyst can deeply audit. The strong approach reads every claim against its evidence on arrival, scores how well each is supported, and aims the deep, independent audit at the claims that score weakest.
Risk-based auditing of an impact portfolio means choosing which impact claims to audit by where the risk is, not by rotation or deal size. Every claim across the portfolio is read and scored against its evidence; the deep, independent audit then goes to the claims with the thinnest support — a missing baseline, a metric with no method, a number with no source. It spends a limited audit budget where a weak claim is most likely, and produces a sample a fund can defend.
Best practice for audit sampling across an impact portfolio is to make the sample risk-based and evidence-led rather than random or rotational. Read every portfolio company’s impact claims first, score each against its supporting evidence, and select the deep-audit sample from the lowest-scoring claims. Document why each was chosen. The result uses the limited audit resource efficiently, surfaces weak claims earlier than a rotation would, and gives an LP a defensible answer to how the sample was set.
ESG due diligence asks whether an investment carries environmental, social, or governance risk — the kind of issue that could damage value. Impact investing due diligence asks a different question: is the positive impact the investment promises real, additional, and measurable. An investment can pass ESG diligence and still fail impact diligence if its impact claim has no evidence behind it. A fund doing both uses the same reading layer, applied to two different questions.
Impact-washing is the practice of presenting an investment as more impactful than the evidence supports — a strong headline number with a thin or absent basis. Due diligence catches it by reading behind the claim: checking the baseline, the data collection method, the beneficiary evidence, and whether an outcome was measured or only an output counted. A claim that cannot be traced to evidence, or that an independent source contradicts, is the signal. Reading every claim on arrival makes that check routine rather than occasional.
For an LP, impact due diligence covers the manager, not only the deal. It examines whether the fund’s impact management system is real and operating, whether impact is built into incentives and decisions, whether reporting is evidence or narrative, and whether the fund verifies its own portfolio’s claims. The aim is to allocate capital to managers whose impact is verifiable — and to screen out impact-washing before the commitment, when it is still a choice.
IRIS+, managed by the GIIN, provides a shared catalog of impact metrics; diligence uses it as the common language for what a portfolio company should be measuring. The Five Dimensions of Impact — What, Who, How Much, Contribution, and Risk — give diligence a structure for assessing a claim, with impact risk as an explicit dimension. Diligence does not certify against either; it reads a claim against them to judge whether the impact is well-defined and well-evidenced.
AI changes the cost of reading impact documents — reports, theories of change, beneficiary surveys, evaluations — against a defined set of impact criteria, replacing weeks of manual analyst review. The distinction that matters is whether the AI runs against a locked framework. A general AI summarising an impact report drifts between runs; a layer reading each claim against a fixed framework, on arrival, produces a rating that holds and a finding traceable to its source — which is what an LP report needs.
Impact investing due diligence software is the system a fund uses to run diligence — screen deals, store impact data, and track portfolio reporting. A due diligence platform, done well, goes further: it reads every impact report and claim on arrival, scores each against its evidence, keeps the source behind every flag, and maintains a continuous record per portfolio company. Sopact is built for the reading and the record — the parts a scorecard or a database leaves to an analyst.
Custom impact diligence reports for specific LP queries are slow when the data is scattered across portfolio reports in different formats. They are fast when every impact claim has already been read, scored, and tied to its evidence on one record per company. A query — by SDG, by region, by outcome — then becomes a view of data that is already structured, not a month of analyst work. The report is custom; the reading behind it was done on arrival.
An impact due diligence report sets out the impact thesis, the evidence assessed, the claims that are well-supported and the claims that are not, and a view on impact risk. A strong report shares two qualities: every finding is traceable to the source document behind it, and the same framework was applied as on every other company, so the portfolio is comparable. A report built from headline metrics alone is a summary of the claims — not a verification of them.
Yes — and this is the main change from the old model. Pre-investment diligence screens the thesis, but the impact claim has to hold for the whole holding period, and an LP can ask for evidence at any point in it. Continuous impact diligence reads each new report and metric on arrival, re-scores the claim against its evidence, and keeps the picture current — so a weak claim surfaces while the fund can still address it, not when it appears in a report.
Start from where the current process breaks, not from a feature list. Walk one portfolio company from the diligence screen through several annual impact reports and find the seam where the claim goes unverified. If the scorecard clears the deal but the reports are never re-examined, the gap is reading. If every company is rated differently, the gap is a locked framework. If the audit is spent by rotation, the gap is sampling. If a number cannot be traced to evidence, the gap is what an LP will ask for. The diagnosis decides what you need.
Framework and standard names referenced on this page are the property of their respective organizations. Information is based on publicly available documentation as of May 2026 and may have changed since. To suggest a correction, email unmesh@sopact.com.
Bring a real set of portfolio material — a batch of impact reports, metrics, and theories of change, in whatever languages they arrived. We will run it through Sopact and show you the impact claims read against their evidence: the thin baseline, the metric with no method, the headline number with no source — every flag traceable to the document it came from. A parallel pilot you can run alongside the diligence process you have today.
30 minutes · your real portfolio reports · no migration commitment
