WholeSum raises pre-seed to make unstructured text auditable using hybrid AI
This article covers WholeSum, a UK-based data startup that has raised £730,000 in combined funding and grant support. The pre-seed funding will be used to develop a hybrid-AI analytics layer to convert unstructured organisational text into auditable, reproducible insights for research institutions, healthcare providers and financial services teams.
WholeSum, a UK-based data startup, has built a hybrid-AI analytics layer designed to convert large volumes of unstructured organisational text into auditable, reproducible insights. The capability matters because many teams still rely on slow manual coding or opaque large language model summaries that cannot be reliably reproduced, leaving important signals buried in customer feedback, transcripts and open-ended survey responses.
Why it matters
Most organisational data is unstructured: surveys, interview transcripts, support tickets and online conversations. That material often contains the nuance teams need for product, policy and research decisions, but extracting it at scale is hard. Manual coding is slow and inconsistent. Off-the-shelf LLM summaries can be fast but suffer from hallucinations and non-reproducible outputs, which makes them risky for high-trust environments such as research, healthcare and financial services.
WholeSum positions itself as an alternative that aims to make qualitative evidence machine-readable and traceable, so organisations can act on patterns in text with statistical confidence rather than intuition.
Inside the product
WholeSum provides an API-first analytics layer that ingests free-text datasets and outputs quantified, reproducible themes and metrics ready for statistical analysis. The company says a dataset of 10,000 responses — typically weeks of work for a trained analyst — can be processed in seconds and then fed into deeper toolkits for decision-making.
WholeSum describes its approach as hybrid-AI: combining machine learning with statistical inference and controls designed to make outputs auditable and repeatable. It highlights work with partners such as Imperial College London and Female Founders Rise in collaboration with Barclays as examples of how richer signals can be found in unstructured audience data rather than in simple tickbox metrics.
On performance, WholeSum states that its models outperform leading reasoning models including GPT-5 and Gemini 3 Pro on datasets with clear thematic structure, reporting up to 100× faster processing and up to 100× lower theme attribution error in its benchmarks. Those figures are presented as company benchmarks rather than independent verification.
A priority for the product is sectors that require dependable qualitative evidence: research institutions, healthcare providers and financial services teams that need audit trails and statistical defensibility before they can act.
Who are the investors?
WholeSum has raised £730,000 in combined funding and grant support. The round includes a grant from Women TechEU and a pre-seed round led by Twin Path Ventures. Additional participation came from SFC and strategic angel investors coordinated via Ventures Together, including founders and operators from JustPark, Episode 1, ClearScore and Prolific.
JustPark is a mobility marketplace; Episode 1 is an early-stage investment firm; ClearScore is a consumer finance platform; Prolific runs an online research recruitment marketplace. The participating angels bring experience across consumer marketplaces, fintech, research tooling and early-stage investing.
In the announcement, John Spindler, Partner at Twin Path Ventures, said:
Qualitative insights have been trapped behind manual workflows and inconsistent methods for decades. WholeSum brings scientific rigour and automation to a universal problem. This is foundational infrastructure for how qualitative evidence will be generated and trusted in the future.
If you're researching potential backers in this space:
- Explore our list of data venture capital firms in the UK
- Browse our directory of data angel investors in the UK
Founder perspective
WholeSum was founded by Emily Kucharski and Dr Adam Kucharski. The pair began developing the product after encountering hallucinations and numerical inconsistencies from LLMs while analysing thousands of user experiences at a previous venture. That experience, they say, exposed a broader market problem: organisations want richer insight from qualitative data but lack tools that are both scalable and scientifically defensible.
Adam Kucharski brings a background in statistical inference and public health research; his award-winning work has informed international bodies including SAGE and WHO, and his open-source tools are used globally. Emily Kucharski has experience translating audience data into strategy for major consumer brands in advertising agencies.
In the announcement, Emily Kucharski, Co-founder & CEO at WholeSum, said:
Most organisations I’ve spoken to have tried using AI for qualitative analysis – and they’ve been frustrated and disappointed. If you can't trust the output, you can't act on it. It’s absurd that qualitative data remains such an untapped goldmine.
The company says the new funding will be used to accelerate product development, expand its science and engineering teams, scale early enterprise deployments and open API pilot partnerships.
The bigger picture
WholeSum sits at the intersection of two broader trends: increasing enterprise demand for reliable ways to operationalise qualitative evidence, and scepticism about the off-the-shelf use of large language models in regulated or high-trust settings. The company’s focus on auditability and statistical rigour is pitched at organisations that cannot accept opaque outputs.
The raise, supported by a Women TechEU grant, also reflects continued public and private interest in funding data‑driven tooling and women-led ventures across the UK and Europe. As more organisations seek reproducible analytics for text, expect ongoing demand from teams in research, healthcare and financial services for tools that prioritise traceability and defensibility.
They've invested in WholeSum
Get to know these data investors
| Investor | Sector | Stage | Activity | Team | Connect |
|---|---|---|---|---|---|
 Women TechEU | 1 investment investment | more info | |||
 Twin Path Ventures | 18 investments investments | 5 contacts contacts | |||
 SFC Capital (SFC) | 47 investments investments | 7 contacts contacts | |||
 Ventures Together | 5 investments investments | more info | |||
 Episode 1 Ventures (Episode 1) | 47 investments investments | 5 contacts contacts | |||
 ClearScore | 1 investment investment | more info |
Click here for a full list of 7,526+ startup investors in the UK