Perspectives · Standards & Evidence

Regeneration Requires Evidence. What the Profession Has to Be Able to Show Before It Uses the Word.

A project team submitted a sustainability assessment describing the ecological outcome of a mixed-use site development as regenerative. The submission listed native planting, a constructed wetland, permeable paving across the parking areas, and a commitment to biodiversity gain. It was a competent package. The team had done real work.

The assessment asked for the documented baseline ecological condition of the site prior to disturbance. The team produced two site photographs and a sentence lifted from the planning application describing the land as degraded scrub.

That is where the regeneration claim ended and a more useful conversation started. Degraded scrub is not a baseline. It is an impression. Without a documented starting condition, there is no way to establish whether the wetland improved hydrological function or simply moved water around, whether the planting increased habitat value or replaced an existing assemblage with a less resilient one, or whether the site's contribution to the surrounding catchment went up or down. The team had also not fixed a boundary. Their claim covered the development parcel, while the ecological system they were describing extended well past it, and the material impacts of the project (aggregate sourcing, haulage, and the upstream draw on water during construction) fell outside the line they had drawn.

The monitoring horizon was the third gap. Planting survival at six months tells you the contractor watered the plants. Survival, recruitment, and function at five to ten years tell you whether an ecosystem is establishing.

The team reclassified most of the outcome as restoration, identified two elements that were mitigation, and retained one genuinely regenerative commitment tied to the wetland with a defined baseline, three indicators, and a seven-year monitoring period with named ownership after handover. The project did not get worse. The description got accurate, and the monitoring budget survived value engineering because someone in the room could explain what it was buying.

That sequence (claim, missing baseline, undefined boundary, absent time horizon, and then a narrower but defensible statement) is now the most common conversation I have about regeneration. The word has moved into the profession faster than the practice behind it.

Why the pressure is real

The case for moving past harm reduction has strengthened considerably in the last few years, and it is worth being precise about the evidence rather than gesturing at it.

The 2025 Planetary Health Check, produced by the Planetary Boundaries Science Lab at the Potsdam Institute for Climate Impact Research, assessed seven of nine planetary boundaries as breached, with ocean acidification crossing for the first time and all seven breached boundaries showing worsening trends.¹ Only ozone depletion and aerosol loading remain within the safe zone. On climate, UNEP's Emissions Gap Report 2025 found that warming projections based on current policies sit at 2.8°C over this century, and that full implementation of national pledges gives 2.3–2.5°C.² Reductions of 35 per cent and 55 per cent against 2019 levels by 2035 would be needed to align with 2°C and 1.5°C pathways respectively.

Those numbers describe systems that are degrading while being managed. Reduction targets applied to an already-degraded system slow the rate of decline. They do not restore capacity. That is the operational argument for regeneration, and it holds independently of who is making it.

The credibility problem is equally documented. The European Commission's 2020 inventory of environmental claims examined 150 claims across product categories and found that 53.3 per cent gave vague, misleading, or unfounded information, and 40 per cent were unsubstantiated.³ Those figures were the basis for a decade of European regulatory work on claims.

That work now has a date attached to it. Directive (EU) 2024/825, the Empowering Consumers for the Green Transition Directive, was required to be transposed by Member States by 27 March 2026 and applies from 27 September 2026.⁴ It bans generic environmental claims such as "environmentally friendly" and "eco-friendly" where recognised environmental performance cannot be substantiated, prohibits sustainability labels not founded on a certification scheme or authorised by public authorities, and targets impact claims that rely on offsetting. Third-party verification independent of both the scheme owner and the trader becomes necessary for sustainability labels.

Seven weeks from now, in the largest single consumer market in the world, an unsubstantiated environmental claim becomes an enforcement matter rather than a communications risk. Regeneration is a stronger claim than sustainability. It asserts improvement beyond a prior state. Any organisation putting that word in front of a client, a regulator, a lender, or a jurisdiction operating under this kind of regime should expect to be asked what the prior state was and how the improvement was measured.

How regeneration entered this body of work

GPM began in 2009 with an environmental focus and a fairly narrow question about how project delivery decisions produce environmental outcomes. The work moved through the P2 method, then PRiSM, then the P5 impact ontology covering People, Planet, Prosperity, Process, and Product. Each step widened the boundary of what a project team was expected to account for.

Regeneration first appears in the second edition of Sustainable Project Management: The GPM Reference Guide, published in 2018. In that edition it sits inside the discussion of the Blue Economy model, identified as a perspective alongside circular economy, resource efficiency, and decoupling, with the surrounding text describing systems in which by-products become inputs and the environment regenerates as a consequence. It is a concept named and placed in a project management guide, not a developed method. I want to be exact about that, because the distinction between naming an idea and operationalising it is the whole argument of this paper.

The developed treatment arrived in the third edition in 2024, which sets out a progression across five positions (Conventional, Green, Sustainable, Restorative, Regenerative) and locates project resource consumption against the planetary boundaries on either side of that scale. Conventional practice sits at unchecked consumption. Green practice makes incremental environmental improvements and, as the guide puts it, generally falls short of improving environmental and social conditions. Restorative practice repairs damaged environments and communities within the scope of the project. Regenerative practice designs for enhanced resilience and health in the natural and social systems the project touches.

The current standard makes it measurable. The PMI^® GPM^® P5™ Standard for Sustainability in Project Management, version 4.0, published in 2026 by the PMI GPM Sustainability JV, defines regeneration as the measurable enhancement of system capacity beyond baseline conditions, resulting in improved ecological, social, or economic resilience over time. It structures impact response as a continuum of mitigation, restoration, and regeneration, and it states plainly that not all projects will achieve regenerative outcomes.

That last point matters more than it looks. A standard that permits a project to conclude it cannot regenerate is a standard that can be used honestly.

P5 carries three distinct regeneration elements (Social Regeneration, Ecological Regeneration, and Economic Regeneration), each with its own baseline requirements, evaluation questions, and reporting linkages to GRI, SDG targets, and the ISSB and TCFD structures. The ecological element instructs teams to establish documented baselines prior to disturbance, avoid reliance on unverified offset mechanisms in place of local restoration, and allocate resources for long-term monitoring. The economic element distinguishes durable system enhancement from temporary economic activity. The social element asks whether activities strengthen or weaken the long-term capacity of affected communities and institutions, and whether current arrangements create dependency, exclusion, or erosion of trust over time.

The same standard contains an element on Green Claims and Greenwashing, which requires that environmental claims be supported by documented evidence, that material adverse impacts be disclosed alongside claimed benefits, and that supplier claims be independently verified. Claim integrity is treated as a project impact with the same status as water consumption or emissions.

Competence was handled separately. The GPM^® Sustainability Competence Standard, version 2.2, published in 2024 under a Creative Commons licence, defines Living Systems Stewardship as an assessed unit covering biodiversity protection, natural resource preservation, and the restoration of ecological balance, broken into performance criteria at the level of individual practitioner behaviour: establishing and monitoring targets within an area of responsibility, tracking and sharing resource and emissions data to support decisions, collaborating on land and water use. Regenerative Design appears as Unit 8, defined as development in the service of life, with performance criteria covering net-positive outcomes, restoration of degraded ecosystems, alignment with global biodiversity goals, and frameworks connecting design choices to measurable long-term impacts.

Unit 8 is not assessed. The standard says why: regenerative design is an emergent practice, and there are not enough people doing it to assess it credibly. The stated intent is to change that.

Seventeen years of work does not make anyone right. It does establish that the operational questions have been asked before, and that answers exist which can be examined, criticised, and improved.

What regeneration requires at project level

The distance between the word and the practice is mostly technical, and it is worth setting out concretely.

A documented baseline. The condition of the affected system before the project acts on it, recorded with methodology, date, and uncertainty. Ecological baselines need species and habitat context, not land-use descriptors. Social baselines need conditions among affected populations, not census averages for the region. Economic baselines need capability and capacity measures, not headline spend.

A defined boundary. The system the claim covers, and the systems it does not. Most weak regeneration claims are boundary failures. A site improves while the catchment it draws from degrades. A community gains employment while displacement occurs two kilometres away. A material is described as restorative while its extraction footprint sits outside the assessed scope.

Indicators tied to function, not activity. Trees planted is an activity measure. Canopy survival, recruitment, soil carbon, and hydrological function are condition measures. Jobs created is an activity measure. Wage levels, retention, local ownership, and the distribution of benefit are condition measures.

A time horizon that matches the system. Ecological establishment operates on multi-year cycles. Institutional capacity does the same. A monitoring period that ends at project closure will report on construction, not on outcome.

Named responsibility after handover. Regenerative outcomes mature after the project team has demobilised. If nobody owns the monitoring and nobody funds it, the claim expires the day the contract closes.

Honest treatment of unavoidable impact. A project claiming ecological regeneration while generating significant emissions, consuming scarce water, or driving land conversion in its supply chain is making a partial claim. P5 asks directly whether claimed improvements exceed unavoidable project impacts. Most of the time the answer is no, and the correct description is mitigation or restoration.

Applied to a construction project, the difference shows in the decisions. Reduction-oriented practice limits clearing, controls runoff, manages noise, and specifies lower-carbon materials. Regeneration-oriented practice starts earlier, at siting and design, and asks what condition the land, water, and community were in, what the project could measurably improve, what materials would return value to the systems they came from, and what arrangement leaves local capability stronger than it found it. Those choices affect procurement criteria, contract terms, supplier selection, monitoring budgets, and the definition of project success. They change cost and schedule. A team that says regeneration and changes nothing in those documents has not made a regenerative decision.

Regeneration washing

A working definition, offered so the profession has something to argue with:

Regeneration washing is the use of regenerative language, claims, programmes, frameworks, events, or commitments without sufficient evidence that regenerative principles materially influence an organisation's governance, operations, decisions, impacts, and performance.

It shows up in recognisable forms. Regenerative terminology adopted without defined criteria for what would qualify. Public commitments with no assigned governance or decision rights. Claims without baselines, without measurement, or without stated boundaries. Net-positive assertions with no disclosed method. Selective disclosure in which benefits are reported and material adverse impacts elsewhere in the system are not. Event and programme branding around regeneration with no disclosed organisational performance. Certification or teaching of regenerative practice without comparable internal application.

None of these require bad faith. Most greenwashing in my experience was not deceptive by intent. It was language that ran ahead of capability, adopted by people who believed it, inside organisations that had not built the measurement infrastructure to support it. The result is the same. Claims that cannot be substantiated devalue the claims that can, and practitioners doing rigorous work end up competing with organisations doing none, at the same price point, with the same vocabulary.

The profession has already been through this cycle once with sustainability. Regeneration is a stronger claim and will discredit faster.

The standard professional bodies should meet

Organisations that publish standards, certify practitioners, define competence, convene the profession, and shape what counts as good practice occupy a particular position. Their claims propagate. A weak evidence standard at that level becomes a weak evidence standard in thousands of project plans.

A reasonable expectation, applicable to any such organisation, has six parts.

Governance. Named ownership of sustainability and regenerative commitments at board or executive level, with defined decision rights, and evidence that those commitments constrain actual decisions.

Materiality. A stated assessment of which ecological, social, and economic systems the organisation materially affects, with the boundary made explicit.

Application. Evidence that the principles reach procurement, supplier selection, investment, partnerships, travel, and events, which for professional bodies is usually where the material impact sits.

Measurement. Baselines, indicators, targets, methodology, time horizons, and a statement of uncertainty.

Disclosure. Public performance data, the method behind it, progress over multiple periods, and disclosed gaps and limitations. A policy statement is a commitment. Performance data over time is evidence.

Outcomes. Results that can be examined by someone who did not produce them.

There is a seventh expectation that applies specifically to certifying bodies, and it is the one most likely to be skipped. An organisation that certifies people in a competence should be able to say what evidence a candidate must produce, and should decline to certify where that evidence base does not yet exist. Regenerative Design sits in the GPM competence standard as a fully specified unit that carries no credential, because the population of practitioners able to demonstrate it against performance criteria was too small to assess honestly. Publishing an unassessed unit is commercially unhelpful. It removes a marketable credential from the catalogue at exactly the moment the vocabulary is in demand. It also keeps the standard truthful about the state of the practice, which is the only reason to publish a standard at all.

This standard should be applied to GPM as readily as to anyone else. GPM publishes its sustainability policies, disclosures, and reports, and has for years released standards and guidance openly, in several cases under Creative Commons licensing, rather than behind subscription or membership access. That decision was practical rather than principled at the outset (practitioners could not implement what they could not read) and it has had a useful side effect, which is that the work is available to be checked. Applying our own requirements internally has been consistently uncomfortable and consistently productive. It exposes unclear language in the standard, data that turns out to be difficult to obtain, indicators that measure the wrong thing, unassigned ownership, and boundary problems we had not anticipated. Several of the revisions in the current P5 exist because our own reporting failed against an earlier version of the text. Anyone publishing regenerative guidance without running it against their own operations is shipping untested requirements.

None of this positions GPM as exemplary. It positions the work as inspectable, which is the only claim that matters in this context.

What the profession should expect

The practical implications are narrow and unglamorous.

Practitioners should expect to produce baselines. Where a client, sponsor, or standard asks for a regenerative outcome, the first deliverable is documentation of the starting condition of the systems in scope. Where that is not funded, the correct response is to say that the claim cannot be supported and to describe the work as mitigation or restoration instead.

Organisations should expect claim substantiation to become a compliance function rather than a marketing one. The European timeline makes this concrete for anyone selling into that market from September, and comparable regimes are developing elsewhere. Project documentation is where substantiation either exists or does not.

Certifying bodies should expect to be asked what a practitioner holding a regenerative credential can actually do: which assessments they can run, which baselines they can specify, which indicators they can select and defend. Where that answer is thin, the appropriate response is to publish the competence and withhold the credential until the practice catches up.

Bodies that convene the profession around regeneration should expect to be asked for their own governance, measurement, and disclosure. Not as a rhetorical move, and not selectively. As the ordinary consequence of teaching a subject.

The work of restoring degraded systems is achievable, and projects are the mechanism through which it happens. That is why the vocabulary is worth protecting. A claim that cannot be checked does not restore anything, and every unsupported use of the word makes the supported ones harder to distinguish. The profession has the standards, the assessment methods, and the competence frameworks to hold a higher line. What remains is the discipline of not saying more than the evidence allows.

¹ Planetary Boundaries Science Lab (2025). Planetary Health Check 2025: A Scientific Assessment of the State of the Planet. Potsdam Institute for Climate Impact Research. https://www.planetaryhealthcheck.org/

² United Nations Environment Programme (2025). Emissions Gap Report 2025: Off Target. UNEP, Nairobi. https://www.unep.org/resources/emissions-gap-report-2025

³ European Commission (2020). Inventory of environmental claims, 150-claim sample. Figures reported in the Explanatory Memorandum to the proposal for a Green Claims Directive, COM(2023) 166 final. https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:52023PC0166

⁴ Directive (EU) 2024/825 of the European Parliament and of the Council of 28 February 2024 amending Directives 2005/29/EC and 2011/83/EU as regards empowering consumers for the green transition. https://eur-lex.europa.eu/eli/dir/2024/825/oj/eng

Terms used in this article, including Regenerative Design, Regenerative Development, Regenerative Sustainability, and Greenwashing, are defined in the GPM Glossary of Sustainability and Sustainable Project Management Terms.