The LIFE STRESS project successfully delivered all of its planned activities and objectives. Highlights include:

• The development of a new risk Climate Risk Intelligence Solutions Platform (CRISPY) that, first of its kind, offers financial institutions, banks, asset managers and owners, supervisor and central banks fully end-to-end climate stress testing analysis from portfolio data matching using dedicated granular climate asset-level data, through the design of scenarios or choice of preexisting ones to integrated nature, social, just transition, physical and transition risk portfolio analysis of financial instruments linked to SMEs and listed companies.

• A scenario repository of +854 climate scenario pathways, coupled with a new Scenario Designer tool that has been applied by +300 financial market participants, is powered by the expert insight of +240 climate transition experts around the world, and is supported by an established large climate finance initiative the Inevitable Policy Response.

• The total user base (offline and online) of the CRISPY tools reached an estimated +500 financial market participants.

• The publication of a new kind of model JTRISK designed to assess the implication of climate transition on the labour market to achieve a just transition. The model links our newly developed 1in1000 asset level database with employment figures of large steel manufacturers in the EU and beyond.

• The development of the first integrated climate-nature-social risk exercise NRISK and SRISK, with the model road-tested with the Dutch Central Bank as part of an internal exercise. The model has since been applied by a non-European financial supervisor as part of a planned stress-tests and a second application with an additional supervisor is under way;

• The development of a the tilTRISK climate risk model, building on the LIFE CB-PASTAX project and the tilt database, and its application on +200,000 European SMEs. As part of this project, we partnered with the Bundesbank on a research application of SME risk data for the German financial sector, with a planned publication by Q4 2025 / Q1 2026. This SME risk analysis is to our knowledge the first of its kind;

• The project directly or indirectly generated a range of thought leadership of the project driving public awareness of the core objectives of the project. This thought leadership involved discrete publications building on our tools and analysis designed to make the technical work more accessible beyond the methodology papers and software solutions generated within the project. They also served as an important internal tool to document implications of our work in the context of applications. Examples include a public paper integrating nature-social risks, building on the research work with the Dutch Central Bank, work on ‘adaptation’ factors (e.g. climate insurance policy measures) reflecting the need for our models to consider adaptation responses, and a paper on strengthening corporate resilience to climate risks. In addition, the work of this project also fed into a book published in German, English and Spanish (forthcoming) (Note, writing of the book was not funded by the project itself), and various blogs and opeds, including on the implications of the financial disruption seen in 2025 for long-term risks;

• A strong sustainability plan, including but not limited to: i) the integration of the project outputs by commercial partners (e.g. tilt, Morningstar, Fitch Ratings); ii) the replication of the project outputs in other markets outside of the EU that provide follow-up funding (e.g. UK government, German government) and ensure global synergies; iii) The development of an institutional partnership with the University of Oxford that ensures the anchoring of the work in academia, iv) governmental partnerships to scale the research internationally, v) the publication of tools that can be maintained with minimal costs, vi) the development of alternative datasets that are open-source that reduce the base costs of the infrastructure and model, vii) the partnering with strong existing brands (1in1000, Inevitable Policy Response) to enhance visibility and ensure integration into larger programs, viii) synergies with the outputs of another LIFE project (LIFE CB-PASTAX), and finally ix) the partnership with the Principles for Responsible Investment to ensure broad market uptake.

The project also surfaced a number of lessons learned and challenges including notably the following:

• We believe that  project did meet its objectives around social risks (development of SRISK, JTRISK) but there is potential to go significantly beyond, especially in their application with financial institutions. While the focus on these risks was recognized by the NGFS as not being considered, highlighting the need for this kind of work, this area of analysis still is perceived to be my market participants too much in its infancy. In addition, the focus on nature / ecosystem risks meant there was little bandwidth for both topics across stakeholders.

• Similarly, while the project successfully built on LIFE CB-PASTAX, the other LIFE project laid new ground in developing data for SMEs, our timing to deliver SME climate stress testing to the market in already updating and upgrading this work was ‘too quick’ and we expect this work to achieve its impact targets beyond the time horizon of this project. We should have been more mindful of the user journey on SMEs, overestimating the ‘adoption rate’ on our work in this area, as outlined above. While the synergy will ensure that our work can seamlessly plug in moving forward, the actual ability to road-test two steps for SME data by banks was limited and so the feedback was more focused on qualitative aspects, rather than a full-fledged deployment. Similarly, we expect the final report of the Bundesbank to be more limited in scope in terms of the kind of metrics they will feature.

• At the inception of the project and during the first half of it, we envisioned the software to ‘traditionally’ rely on data from commercial data providers and we could have been faster to recognize the opportunities for open-source data application. We have now developed in-house our own 1in1000 asset-level database that, although not as comprehensive in coverage, gives us the freedom, flexibility and cost efficiency we could not have delivered with commercial data providers.

• Finally, all of our deliverables are ‘living documents’ that we continue to fine-tune, adjust and improve throughout the project. The distinction between ‘road-testing’ and Launch is somewhat artificial as we still improve models after launch in the spirit fo ensuring sustainability. We need to do a better job in deliverable scoping as we frequently write proposals with intermediate deliverables scheduled during the project. For our project, we always end up updating our work until the end of the project given its dynamic nature (even just as simple as debugging code if errors are noticed). We should put a bigger emphasis on intermediate and final deliverables in future project design.

• We realized that the actual comparability of financial scenarios was limited ultimately given the idiosyncratic design by different central banks. We thus put greater emphasis on the interplay between financial and economic scenarios (as shown in our this is the way report), both in terms of our work and the Scenario Designer.

• We overestimated the extent to which users would actually want to use our platform without handholding. We thus did not have significant ‘independent users’, but a meaningful number of users with our support. This implied that there was a) more demand in the offline use of our platform rather than independently online, b) higher degree of resource demand for stakeholder support than anticipated, and c) bigger barriers to scaling. We have implemented this lesson in moving away from open independent access on our CRISPY platform to a direct assisted access model.