Koi DocsKoi Docs

Facilitating Solutions

Last Updated: 2025-10-06

Facilitating solutions do not directly achieve GHG impact themselves but enable systems to reduce emissions. Koi considers facilitating solutions to be composed of both solutions that enable direct solutions and solutions that improve the emissions performance of existing systems (that may themselves not be solutions).

Modeling Approach

As a general approach, we only model facilitating solutions that have a measurable physical change in the system. We do this by forcing the Koi Engine to find an appropriate baseline to match. If no baseline can be reasonably matched, the model won't be made. This requirement exists because there must be sufficient specificity to a facilitating solution to model its effects with real-world systems.

The key dividing line for modeling is: Does the software achieve sector-specific, direct and measurable emissions reductions through targeted technical workflows? Or is it generalized, financial, or enabling broad sustainability action? Only solutions that can demonstrate the former are suitable for reliable modeling.

For example, a solution that uses software to change consumer behavior may be modeled if there is specificity to the expected behavioral change, such as reducing food waste with product discounting. However, if the software is broadly claiming more sustainable consumer behavior without a clear mechanism of impact and/or industry application, such as with a general claim of enabling consumers to make sustainable decisions, a model will not be generated.

A solution may be classified as facilitating but not proceed through the Koi Engine because it doesn't meet the specificity requirements for reliable modeling.

Note that even with well-defined mechanisms of impact, facilitating solutions typically require more assumptions around their downstream effects than direct solutions. All assumptions should be documented and amenable to modifications (e.g., what share of customers will change their purchasing behavior and by how much). Therefore, expert review and customization is especially critical for trustworthy faciliting solution models.

Classification Criteria

Facilitating Solutions Well-Suited for Koi Modeling

These types of solutions are classified as facilitating and generally have sufficient specificity to be reliably modeled. This is not an exhaustive list but is intended to represent the types of facilitating solutions commonly modeled:

  • Industry-specific solutions: Solutions designed for clearly defined physical industries (e.g., steelmaking, cement, manufacturing, logistics) that enables measurable emissions reductions (e.g., remote energy monitoring and optimization for factory machinery, platforms for building waste prevention)
  • Targeted technical workflows: Solutions with narrowly targeted, well-defined intervention mechanisms focused on specific areas for optimization across sectors (e.g., predictive maintenance software for industrial equipment)
  • Enabling components: Physical infrastructure that enables direct solutions, such as fast EV chargers

Difficult-to-Measure Facilitating Solutions

These solutions are classified as facilitating but can only be modeled if they meet the specificity requirements outlined in the Modeling Approach section above. In many cases, these solutions are too vague, varied or unestablished in their mechanism of impact to meet these requirements.

  • Financial and carbon/sustainability platforms: Software focused on finance, carbon credits, carbon trading, ESG investing, sustainability-linked lending, environmental finance, insurance, accounting, analytics, management, sustainability incentives, or general sustainability enablement
  • Generic technological solutions: Platforms providing general-purpose technological solutions for carbon or sustainability improvements, often broadly applied across value chains
  • Information and market solutions: Solutions whose main benefits are "better information," "unlocking markets," or broad sustainability tracking not linked to specific physical process changes
  • Monitoring and measurement tools: Software whose primary function is observation, monitoring, or measurement of carbon, GHGs, or sustainability indicators

Multi-System Services or Infrastructure: Facilitating solutions providing infrastructure or services that reduce emissions across multiple systems (e.g., green startup accelerators) are not currently amenable to automated Koi Engine modeling. Instead, each solution for modeling would need to be input individually. For example, the Engine could generate models for each portfolio company at an accelerator and those models can be assembled by the user into a Portfolio Collection. The Engine cannot currently accept an aggregate of solutions as a single input for modeling.

Baseline Matching Requirement

For any facilitating solution to be modeled, there must be an appropriate baseline activity in the Koi Data Lake to match upon. This baseline represents the conventional system or process that the facilitating solution is designed to improve or replace, providing the foundation for quantifying the measurable physical change in emissions.

Allocation Considerations

How to approriately allocate avoided emissions for facilitating solutions is an unsettled science. Koi does not pre-segment any analysis data for allocation and instead provides the entire system savings as a model. Users can apply an adjustment factor using their allocation method of choice within a Collection.