Most circular-economy take-back programmes that look healthy on paper run badly at scale. Either the corporate sustainability team has the model right but no operating system underneath it, or the operations team has the system but cannot make it work consistently across countries. The harder version of the problem is the one where the model exists, the technology works, and the company has even committed to it publicly. What is missing is the operating coherence to run it across markets that all look different from each other.
Between 2018 and 2020 I worked with the SYSTEMIQ team on the relaunch of Interface ReEntry, the company’s global carpet-tile take-back programme, across six EMEA priority markets. Interface had the technology (a working recycling plant in the Netherlands), the brand (a long-running sustainability commitment via Mission Zero and Climate Take Back), and the customers. What it needed was a way to run the programme consistently across the United Kingdom, Ireland, Spain, Benelux, France and Germany, four of which I worked on directly.
This article is the third in a small series on designing and scaling circular systems. The first looked at a single-stream industrial case in the Seychelles. The second walked through a multi-stream municipal case in coastal Indonesia. This one is about multi-market commercial B2B. Same family of problem, different operating constraints.
Why Multi-Market Take-Back Is Harder Than It Looks
The natural assumption walking into a programme like Interface ReEntry is that the hard work has already been done. The recycling plant exists. The corporate commitment is public. The product is the same across markets. The marketing collateral can be translated. Surely launching in another country is just a matter of finding a local partner and turning on the tap.
The assumption breaks on contact with each market. Three forces drive that.
Waste licensing regimes are not interoperable. Moving used carpet across borders within the EU triggers transfrontier shipment rules and country-specific waste-carrier licences. The UK requires a Waste Carrier, Broker and Dealer registration. The Netherlands operates a more decentralised model. Spain layers regional autonomous-community variations on top of national rules. Ireland tracks the UK in some respects but not all. A programme that works smoothly in one market will, by default, not work in another, and the work of making it legal is rarely budgeted properly at the start.
Partner ecosystems differ by an order of magnitude. The reuse stream of the ReEntry model depends on local resellers and social enterprises that take usable carpet tiles, refurbish them where needed, and resell them at a discount to second-life buyers. In the United Kingdom, that ecosystem exists with several established players. In Ireland, the ecosystem is much thinner and a programme launch has to either build partner capacity or accept a sort-only model. In Spain the ecosystem is fragmented regionally. In Benelux it is concentrated geographically. The same programme deliverable looks very different in each market.
Customer profiles vary by sector concentration. Carpet tile is a B2B product. Its buyers are commercial property occupiers, office fit-out designers, public-sector facility managers, and the architects and contractors who specify them. The mix of these buyers differs significantly by market. London is heavily corporate-occupier. Madrid leans more design-led. Dutch buyers have a higher baseline sustainability expectation. Each profile needs a different sales conversation. The brand message can be shared. The pitch has to be re-pitched.
Most programmes underweight all three. The default failure mode is a relaunch that gets the marketing right and the operations wrong, runs for eighteen months on goodwill, and then quietly degrades when the central team rotates out.
The Shared Programme Architecture
The first piece of work in the relaunch was to define a shared structure that every market would follow. Templating was the only way to avoid each market reinventing the deliverables from scratch and producing eight different versions of “ReEntry” with no common spine.
The template covered eight workstreams. Each had a defined owner, expected outputs, and a review checkpoint with the central programme team.
- Value proposition. Market-specific articulation of why this customer in this geography should choose Interface ReEntry over the default of skip-to-landfill. Translated from the global proposition, not copy-pasted
- Target customer profile. Segmentation of which buyer types in this market the launch would prioritise first, with the rationale for sequencing
- Competitive analysis. What local alternatives existed (other take-back schemes, generic recyclers, internal carpet refurbishers) and how ReEntry compared on price, quality and credibility
- Operating P&L. Bottom-up cost and revenue model for the market over three years. Different per market because input prices, partner economics and transport costs all varied
- Partner ecosystem. Identification, qualification and onboarding of local reuse and recycling partners, with formal commercial agreements
- Operations assets. Process flows, templates and checklists for the actual carpet-tile take-back workflow, from customer request to material delivered to its next destination
- Marketing assets. Sales collateral, customer case studies, internal training packs and external communications materials
- Training. Sessions for the country sales and operations teams who would run the programme day-to-day after launch
The discipline that mattered was treating these as a single coherent deliverable, not eight parallel ones. The partner ecosystem fed the operations design. The operations design fed the P&L. The P&L fed the value proposition (you cannot promise what you cannot afford to deliver). The training closed the loop by making sure the people running the system understood why each piece was the shape it was.
What Had to Be Local
The template provided the spine. The flesh varied by market, and being honest about what could be standardised vs what had to be local-first saved time. Four things consistently had to be designed market-by-market.
The legal layer. Waste-carrier licences, transfrontier shipment notifications, end-of-waste documentation. This is the slowest workstream by far. A programme that does not start the licensing process in the first month is the one that misses its launch date in month nine.
The partner roster. The two-model split (full reuse-and-recycle vs sort-only with shipment to the Dutch plant) depended entirely on whether mature local reseller capacity existed. The first move in any market was a partner-ecosystem scan: who already does this, how big are they, what is their capacity, what are their margins, do they have the licences. The answer determined the operating model, which determined the P&L, which determined the launch sequence.
The customer-segment entry point. Not all customer segments could be sold simultaneously. In each market, the team picked one segment to lead with based on existing Interface sales footprint and customer readiness. London corporate occupiers in the UK. Design-led specifiers in Madrid. Public-sector clients in Ireland. The launch sequencing then expanded into adjacent segments once the first wave was running.
The marketing register. The global Interface message about ReEntry resonates differently in different markets. UK customers respond to compliance-and-reputation framing. Dutch customers respond to operational efficiency framing. Spanish customers respond to design-and-quality framing more than either. The deck and the pitch had to be re-recorded for each market even when the underlying programme was identical.
Recognising this distinction up front (template = shared, four things = local) shortened the launch path materially. Programmes that tried to standardise the four local elements lost time renegotiating them market by market. Programmes that templated everything ended up with a thin global story that did not land in any single country.
The Two Operating Models
Across the markets I worked on, the operating model split into two patterns, and the choice between them came down to local partner availability.
Model A is reuse-first with recycling for residuals. Used in markets with mature reseller partner ecosystems. Tiles are uplifted from the customer site, transported to the reseller, sorted and graded for resale where condition allows, with the residual material going to the Dutch recycling plant. The economic logic is strong because reuse captures more value than recycling. The operational logic depends on having a partner you can trust to grade tiles consistently and represent the Interface brand appropriately.
Model B is sort-only with central recycling. Used in markets without mature reseller capacity. Tiles are uplifted, sorted at a local consolidation point to remove contamination, and shipped to the Dutch recycling plant. The economics are tighter (shipping cost per tonne is high and the value capture per tonne is lower), but the model is operationally simpler and lower-risk for a market launch. Model B was often a starting point that could evolve into Model A once a local reuse partner could be developed.
The mistake to avoid is forcing a market into Model A when the partner ecosystem cannot support it. A launch under Model A with weak partners produces inconsistent service, customer complaints and reputational damage that takes years to undo. The honest call in those markets is Model B at launch, with a deliberate two-to-three-year plan to develop local reuse capacity that will eventually justify the model upgrade.
What Worked Across Markets
Looking back at the relaunch, three things worked consistently well.
Cross-market lesson sharing. Each market launched in a slightly different sequence, which meant the markets launching later could draw on the lessons of the earlier ones. We set up a shared repository of operational lessons learned that each country team contributed to and consumed. This is unsexy work but it is what stopped each market from repeating the same partner-onboarding mistakes or producing the eighth slightly-different version of a sales deck.
Joint marketing assets with local adaptation slots. The marketing materials were built as bilingual templates with explicit placeholders for market-specific case studies, regulatory references and customer testimonials. Country teams could swap the local content in without rewriting the whole asset. This kept the brand consistent and the local relevance high.
Quarterly cross-market reviews. The country leads met every quarter to compare progress on the same metrics: customers signed up, tonnes recovered, partner capacity built, P&L health. The cross-market visibility kept country teams accountable and made it impossible to hide a stalling market behind talk of “local conditions”. The meetings also surfaced the patterns that mattered. When two countries reported the same partner-quality problem, the central team could investigate. When one country was running ahead, the others could learn what they were doing differently.
What Was Hard
Two pieces of the work remained genuinely unsolved by the time I rotated out.
The economics of small markets. Some priority markets did not have enough Interface installed-base to support an economically viable independent take-back operation. The launch worked operationally but the unit economics were marginal. The honest options for these markets are either (a) accept a sustained internal subsidy from the central programme as the cost of brand consistency, or (b) run a lighter-touch reactive model where the programme exists for major customers but is not actively promoted. Neither is fully satisfying. The trade-off between commercial discipline and brand consistency is a real one.
The handoff from launch to steady-state. The relaunch model relied on a central programme team supporting each market through launch and the first year of operations. The handoff to fully country-owned operation is the moment of truth. Some markets executed this handoff well and the programme continued under local stewardship. Others lost momentum once the central support reduced and reverted to a more reactive posture. The pattern correlated less with country and more with whether the local country lead treated ReEntry as a strategic differentiator or a sustainability tax. The hand-off work needed more deliberate succession planning than the original programme design budgeted for.
A Working Method for Multi-Market Take-Back Design
If you are designing or relaunching a take-back programme across multiple markets, the method that emerged from this work generalises as follows.
- Template the spine, localise the limbs. Decide explicitly which workstreams must be shared across markets (value proposition framework, programme architecture, technology, brand) and which must be designed local-first (legal, partner roster, customer segment entry point, marketing register). Document the distinction before launch planning starts.
- Run the partner-ecosystem scan first. In each market, the first deliverable is not the marketing plan or the P&L. It is a clear picture of who already operates in adjacent space, what their capacity is, and which gaps the programme has to fill itself. The scan answers whether the market can run Model A (reuse-first) or has to start with Model B (sort-only).
- Start the legal track in month one. Waste-carrier licences, transfrontier shipment notifications, and end-of-waste documentation are the longest-lead-time items. Launches slip on legal more often than on anything else.
- Sequence customer segments instead of launching across all of them. Pick one segment per market for the first wave. Build the operations and the partner relationships around that segment. Expand once the first wave is stable. Trying to be everything to everyone produces patchy service quality and lukewarm sales conversations.
- Build the cross-market repository before the first market launches. The system for capturing and sharing operational lessons should exist before there are any lessons. Otherwise it becomes a post-hoc effort and the second market loses most of what the first market learned.
- Design the handoff to country-owned operation deliberately. Identify the local owner before launch. Define the transition phases. Specify what central support will be reduced and when. Without this, the programme reaches steady-state in some markets and degrades in others, and the difference will not be where you expect.
- Accept that some markets will not be self-sustaining and decide what to do about it. The brand-consistency case and the unit-economics case point in different directions in small markets. Make the call explicitly. Subsidising for brand reasons is a legitimate choice. Pretending the economics will work when they will not is not.
How This Analysis Was Put Together
This article was written in 2026 from project materials produced between 2018 and 2020, including the EMEA programme pitch deck, the per-market launch templates across the UK, Ireland, Spain and Benelux, and the operations process documentation produced for each market. The work itself was done in person across multiple country visits, with design phases in London and remote operational support afterward.
The synthesis and re-framing for this piece was supported by AI workflows of the kind I now write about at azvai.com. Project materials were re-read and summarised using Claude Code skills for structured-document extraction. The methodology arguments and the four-to-six-year retrospective are mine. The AI work was on the desk side, not the field side.
For public context on Interface ReEntry, the Interface sustainability page describes the programme, the European Circular Economy Stakeholder Platform case study covers Interface’s broader circular model, and the EMEA ReEntry brochure sets out the current programme for European customers.
If you are designing or relaunching a take-back programme across markets and the temptation is to copy-paste the model from your strongest market into the others, that is almost always a sign that the relaunch will produce inconsistent service in eighteen months. The discipline of explicit template-vs-local design is the cheapest insurance you will buy on a programme of this kind. Skip it at your cost.