IRIS+ is the Global Impact Investing Network's metric catalog and measurement system, built to give investors a shared vocabulary for describing impact goals, strategies, and outcomes across otherwise unrelated investments. Its core function isn't to tell investors what impact to pursue — that remains a strategic decision specific to each fund's thesis and mandate. Its function is to standardize how impact, once a strategy is chosen, gets described and measured, so that data collected across different portfolios, funds, and organizations can eventually be compared on something closer to common terms rather than each organization inventing its own metrics from scratch.
The structure
IRIS+ organizes around Core Metrics Sets: curated groups of metrics tied to specific impact themes and explicitly aligned with the UN Sustainable Development Goals. Rather than asking an investor to select from many thousands of possible indicators individually — an approach that earlier, pre-IRIS+ efforts at impact metrics struggled with precisely because of the resulting fragmentation — an investor instead anchors on a defined impact category, such as financial inclusion, clean energy access, or quality jobs, and IRIS+ points toward a relevant, pre-vetted metric set built specifically for that category by practitioners with domain expertise.
Metrics within these sets are further organized around IRIS+'s five dimensions of impact: What, Who, How Much, Contribution, and Risk. This structure governs not just what gets measured in isolation, but how a single metric is meant to fit into a fuller, more honest picture of an investment's actual impact — a large number on its own, without context on who was reached or whether the outcome would have happened regardless, tells an incomplete story that IRIS+'s broader structure is explicitly designed to discourage.
What selecting metrics actually involves
The practical work of using IRIS+ well isn't reading through the catalog, which is straightforward enough on its own. It's deciding, ideally at the point of investment — before capital deploys, not retrofitted afterward once results start coming in — which specific metrics will actually govern reporting for that investment going forward. This requires balancing three considerations that frequently pull in different directions at once.
Comparability favors choosing metrics that are already widely used elsewhere in the portfolio or across the broader sector, so the resulting data can eventually be aggregated with other investments' data. Relevance favors choosing metrics that actually capture the specific impact thesis of this particular investment, which may not be among the most commonly used metrics available in the catalog, especially for investments pursuing a genuinely novel or specific theory of change. Collection feasibility favors choosing metrics the investee can actually report on reliably and consistently, given their existing data infrastructure and internal capacity, rather than metrics that look ideal on paper but that the investee has no realistic way of tracking accurately over the life of the investment.
Investors who default entirely to the most standardized, widely-used metrics available get strong comparability at the direct cost of relevance — reporting something conveniently measurable and common across the portfolio, rather than something that actually reflects the specific impact thesis this investment was made to pursue in the first place. Investors who instead prioritize bespoke relevance over any standardization get the mirror-image problem: rich, genuinely meaningful data for this one investment that can't be meaningfully aggregated or compared with anything else in the portfolio, undermining exactly the kind of portfolio-level view that motivated adopting a standardized system to begin with.
Aligning portfolio reporting over time
At the portfolio level, IRIS+ usage tends to fail less often at the point of initial metric selection and more often on consistency of application over time, which is a less visible but ultimately more damaging failure mode. A metric selected at deal signing needs to remain the actual reporting basis for the life of that investment if any resulting trend data is going to mean anything at all. Switching metrics mid-relationship, because a seemingly "better" or more favorable-looking metric becomes available or convenient, breaks the time series the original selection was meant to build, and often does so quietly, without the change being clearly flagged to whoever is reading the resulting report.
This is a governance discipline as much as it is a measurement one. It requires the original metric decision to be documented clearly and treated as binding, not casually revisited or informally renegotiated at each subsequent reporting cycle simply because circumstances, or convenience, have changed.
What IRIS+ doesn't solve on its own
IRIS+ standardizes what gets measured across a portfolio; it does not, on its own, verify that the underlying data reported against those metrics is actually accurate. It does not resolve genuine disagreements about which Core Metrics Set best fits an ambiguous or genuinely novel investment thesis that doesn't map cleanly onto any existing category. And it does not prevent an investee from reporting a technically correct but selectively framed number that satisfies the letter of the chosen metric while omitting important context.
Adopting IRIS+ gets a portfolio to a common measurement language, which is a real and valuable achievement in a sector that has historically struggled with exactly this kind of fragmentation. It is a necessary condition for producing comparable impact data across a portfolio — it is not, on its own, a sufficient condition for producing genuinely trustworthy impact data, and treating the two as equivalent overstates what metric standardization alone can actually deliver.