Data Products & Innovations
We transform hard-to-access information into trusted, decision-ready evidence for the livestock and aquaculture sectors.
From fragmented data to decision-ready evidence
Working with partners, SEBI-Livestock transforms hard-to-access information into trusted, decision-ready evidence for the livestock and aquaculture sectors. We operate at the intersection of rigorous data science and practical application, ensuring that every product we build is fit for purpose.
The challenge: disconnected sources
Decision-makers we work with often face a barrier: critical data is scattered across private project reports, public censuses, and unstructured grey literature. These datasets rarely "talk" to each other naturally. Without intervention, valuable insights remain trapped in PDF files or incompatible formats, leaving decision-makers with an incomplete picture.
Our approach: the data pipeline
We bridge these gaps through a deliberate pipeline of Ingestion, Processing, and Analytics.
- Ingest: We bring together structured data (such as Gates Foundation Grantee Investment Data and global open databases) and unstructured data (including peer-reviewed literature and investment reports).
- Process (Find, Extract, Consolidate): This is where we make data talk to each other and the next phase of our work is about making data ‘Artificial-Intelligence (AI) ready'. We harmonise data from disparate sources, integrate them into a cohesive whole, and describe them rigorously.
- Analyse: The output is high-quality data ready for analytics, powering our interactive dashboards and models.
Innovation in action
We actively collaborate with partners at the University of Edinburgh to deploy AI tools that expand the breadth of data we can use. Our innovations target specific bottlenecks in the pipeline:
- Discovery: Using AI to discover, extract, and consolidate hard-to-reach data including investment outcomes, national livestock metrics and price data.
- Consolidation: We will be evaluating tools to synthesise trends and enable conversational analytics.
- Gap Filling: When our quality checks identify missing critical context, we are exploring AI methods to help fill these data gaps, ensuring decision-makers have the complete evidence they need.
Grounded in user needs
Our work is grounded in the real questions posed by funders and stakeholders, such as identifying sector bottlenecks. By automating the search and synthesis of information, we free up experts to spend less time hunting for data and more time analysing results, capturing learnings and improving decisions.
Testing, learning, improving
Innovation rarely moves in a straight line. We are deliberate about how we share progress, clarifying the output at each stage—whether it is an internal prototype, a refined method, or a mature tool ready for external use. By sharing work in progress and lessons learned, we maintain transparency about our iteration and timelines. This approach allows us to evaluate AI tools for reliability and fit, designing pipelines that are responsible by default and always prioritising security, fairness, and transparency.
Transparency and quality
Our goal is to provide more of the right data, sooner. Our dashboards clearly indicate data quantity (from "Good" to "Sparse"), ensuring users understand the confidence levels of the evidence. Where possible, we will share the outputs on our website through data downloads or dashboards. Whether you are a funder tracking impact or a researcher exploring new technologies, SEBI-L ensures that data serves as a reliable foundation for better decisions in the livestock sector.
Powering the wider programme
Our data work is the engine for SEBI-L's broader mission. The structured evidence we generate directly enables our other Areas of Work:
- Monitoring & Learning: Providing the collated investment and national data needed to track sector progress.
- Evidence Synthesis & Modelling: Supplying the rigorous, harmonised datasets required for modelling and policy insights.
- Gender Integration & Insights: Ensuring our data infrastructure supports the use of gender-disaggregated data, to identify gaps in equitable systems.
- LD4D Network: Where possible, making our data tools and insights available to the global livestock community to help inform decisions and address evidence challenges.
People
Isabell Orlishausen
Johanna Wong
Phimchanok (Pim) Seelawongseree
Ana Miranda
Vanessa Meadu
Clare Manning
Sophia Jenkinson
Louise Donnison
Vlad Cherman
Alice Aakerberg
Karen Smyth
Gareth Salmon
Products
Stories
Making evidence synthesis easy, the hard way
What are the prospects of using Artificial Intelligence (AI) to automate steps of the evidence synthesis pipeline? Prof Malcolm Macleod explains why there is no easy way to make evidence synthesis easy.
From Static Reports to Interactive Insights: Unlocking Livestock Strategy Data
Explore livestock sector priorities with our interactive Livestock Strategy Documents Dashboard.
How can AI improve national livestock metrics? Insights from our latest workshop
Over the last half-year, SEBI-Livestock has been working with partners across the University of Edinburgh to explore how data science and AI can help address persistent livestock data challenges.
Harnessing Artificial Intelligence for Livestock: Insights from the AI for Good Global Summit
SEBI-Livestock team members report back from the AI for Good Global Summit 2026.
How machine learning can help advance and accelerate AMR research
An innovative machine learning approach spearheaded by SEBI-Livestock and partners could help advance research on antimicrobial resistance (AMR).
Activities
Best practice guidelines for evidence synthesis
The methodological foundation for the SEBI-Livestock evidence synthesis programme.
Explore moreUnlocking livestock investment insights from global open data
We're leveraging machine learning to understand the landscape of investments into livestock development.
Explore moreIntelligent evidence synthesis for animal health
Through our strategic collaboration with the CAMARADES group at the University of Edinburgh's Medical School, we are applying innovative technological approaches to streamline the animal health evidence pipeline.
Explore moreLitXpress - supporting systematic reviews through automation
By automating the laborious task of searching literature, LitXpress frees up researchers to spend more time analysing results. This acceleration ensures that fit-for-purpose evidence can reach decision makers faster, supporting better policies and investments in animal health and production.
Explore more