A public resource for understanding and eventually evaluating, the tools, data, methods and standards used to make sustainability decisions in agriculture.
This is a working preview
What is SADDLE?
There are a lot of tools available to measure, model, manage and report sustainability in agriculture. The harder question is:
Which tools are useful for which decisions, and what data, methods and standards sit behind them?
SADDLE is designed to help answer that question.
Rather than organizing tools into a static market map, SADDLE is built as a relational database designed to be explored by people and queried through an AI agent. That means the different parts of the landscape are connected: a tool can be linked to the decisions it supports, the data it requires, the outputs it produces, how those data are collected, the standards or methodologies it uses, and the environmental outcome pathways it supports.
Those relationships can then be reviewed, filtered, explored and evaluated from different starting points, based on the criteria that matter for a particular decision.
Why an “Evaluator”?
SADDLE does not assign a universal score or declare one tool “best.”
Instead, the underlying data are structured so users can eventually evaluate tools against their own criteria and decision context.
The goal is to let someone ask:
“Given what I’m trying to do, what tools appear to fit, and why?”
Over time, we intend for users to be able to provide the criteria that matter to them and receive a transparent shortlist of relevant tools, including why they may fit, where there are tradeoffs, what information is missing, and what alternatives may be worth considering.
Why not just make another landscape map?
Because tools don’t fit neatly into boxes.
A single platform may support several different jobs. Two tools that appear to be in the same category may require completely different data or generate completely different outputs. And many real decisions require a combination of tools, datasets, models and services, not one winner.
SADDLE is intended to help people start with the job they need to do, rather than with a software category.
How SADDLE is organized
- Use cases — the job to be done. These represent decisions or jobs performed by a primary actor, such as corporate GHG reporting, procurement and sourcing, on-farm performance tracking, project accounting, or risk and resilience assessment. [Explore use cases →] [Suggest a missing use case →]
- Data — the common language underneath the tools. Input data and generated outputs are normalized across the landscape so the same data element can be connected to tools, use cases and methods. [Explore data →] [Suggest missing data elements →]
- Data collection methods — how the data are obtained. These include producer records, equipment data, soil sampling, remote sensing, APIs, third-party datasets and other methods. [Explore collection methods →] [Suggest missing data collection methods →]
- Standards and methodologies — what rules or methods sit behind the tool. Where applicable, SADDLE identifies the accounting standards, protocols, regulatory frameworks and quantification methodologies a tool directly supports or uses. [Explore standards →]
- Environmental outcome pathways — what change the tool can help measure, manage or enable. These include pathways related to carbon, soil health, nitrogen, water, biodiversity, livestock and other environmental outcomes. [Explore outcome pathways →]
- Tools, platforms, models and service providers. The current landscape includes nearly 150 entries, with relationships across all of the elements above. [Explore the tool interface →]
Here’s a snapshot of the full landscape of tools that will be featured. If a tool is missing, or you think we have characterized one incorrectly, please tell us.
This is still a working resource. With a landscape this large, there will be errors, omissions and judgment calls.
The people who build and use these tools are best positioned to help us improve it.
Want deeper access?
SADDLE is designed to be publicly useful through the landscape you can explore above. We’re also developing additional ways to interact with the underlying data.
For people and organizations that want to go deeper, we expect to offer:
SADDLE Agent access: query the underlying relational database conversationally based on the decision, criteria, data, standards or outcomes that matter to you. Agent access is currently in development and will be available free of change to selected beta users.
Contributor access: direct access to explore the underlying relational database for selected researchers, practitioners and others actively helping improve SADDLE.
Licensed data access: a structured copy or snapshot of the underlying SADDLE database for internal analysis.
White-label versions: customized versions of the SADDLE architecture, data and interface for specific organizations, sectors, geographies or decision contexts.
Custom analysis and advisory support: deeper tool evaluation, landscape development, requirements definition and decision support.
Interested in using the SADDLE Agent, accessing the data, contributing, or exploring a version for your organization?
[Tell us what you’re interested in →]
