Graphext: The Spanish Startup That Spent 7 Years Building the 'Formula 1' of Data Analysis
Source: Dev.to
The “Formula 1” Startup Story
In the fast‑paced world of startups, the motto is often “grow fast or die.”
Launching a minimum viable product, gaining traction, and raising multi‑million‑dollar funding rounds seems obligatory.
But what if there was another way? A patient, almost‑artisanal strategy focused on building technology so advanced it creates an insurmountable competitive moat before hitting the commercial accelerator?
Graphext at a Glance
| Milestone | Date | Detail |
|---|---|---|
| Founding | 2015 | Graphext is born. |
| EU Funding | 2018‑2021 | ~€2 M in non‑dilutive grants (Horizon 2020, EIC Fund). |
| Key Seed Round | Jun 2023 | $4.6 M led by Hoxton Ventures. |
| Team | 2024/25 | ~50 employees (sources vary). |
- Founders: Victoriano Izquierdo & Miguel Cantón – Spanish computer engineers with an entrepreneurial spirit since childhood.
- Core Vision: Explainable Artificial Intelligence (XAI) that lets companies understand their data, not just process it.
How Graphext Began
The spark didn’t come from a business plan but from curiosity.
The duo first built contexto.io, a tool for analyzing connections on Twitter. They soon realized the real opportunity lay in:
- Creating broader information contexts.
- Visualising hidden networks that connect people and organisations in any dataset.
Thus, Graphext (a blend of graph + context) was born in 2015 with the mission to democratise data science—bridging the gap between coding experts and business analysts who have important questions but lack the tools to answer them directly.
“A tool as interactive as Figma, but for data science.” – Founders’ vision.
The Technology Edge – The “Formula 1”
After years of intensive R&D, Graphext’s architecture resembles a Formula 1 race car: ultra‑fast, finely tuned, and difficult to replicate.
| Component | What It Does |
|---|---|
| WebAssembly (Wasm) | Executes 80‑90 % of data processing in the browser, eliminating server latency. |
| WebGL | Renders massive visualisations at native‑GPU speed. |
| Apache Arrow | Enables zero‑copy data interchange for lightning‑fast analytics. |
| Proprietary compression libraries | Shrink datasets without sacrificing performance. |
| Internal low‑code language | Lets power users script complex workflows without writing traditional code. |
The result? Exploring and filtering millions of rows feels instantaneous.
The Platform – A Full‑Stack No‑Code/Low‑Code Solution
-
Universal Connection
- Import from a simple CSV.
- Connect directly to modern data warehouses (Snowflake, BigQuery, Databricks, Redshift).
-
Interactive Visual Exploration (EDA)
- Filter, group, cross‑reference variables, and enrich data on the fly.
-
Advanced No‑Code Modelling
- Apply ML algorithms (clustering, NLP, image analysis) with clicks, not code.
-
Prediction with Explainability (XAI)
- Build predictive models (e.g., churn, sales‑lead scoring).
- Why the model makes a prediction is shown transparently—core to Graphext’s strategy.
The “Formula 1” Dilemma: Power vs. Accessibility
- Power: The platform is so capable that a skilled “driver” (business analyst, data scientist, power user) is needed to unlock its full potential.
- Accessibility: It is not a toy for absolute beginners.
Business Model
| Segment | Offering | Goal |
|---|---|---|
| Self‑service | Free & Pro plans | Attract users → organic, product‑led growth. |
| Enterprise | Custom pricing + data‑engineering & training services | Serve large corporations that need support. |
The company is shifting from selling €1 k tickets to closing six‑ or seven‑figure contracts with brands like McDonald’s and Roche.
Funding Strategy – Patience Over Dilution
| Funding Source | Details |
|---|---|
| Modest Seed Capital | Small rounds, including one led by K Fund. |
| Public Grants | ~€2 M from EU programmes (Horizon 2020, EIC Fund) – non‑dilutive R&D financing. |
| Seed Round (Jun 2023) | $4.6 M led by Hoxton Ventures; >80 angel investors (Freepik, CARTO, Snowflake, GitHub, Meta). |
This patient approach let Graphext build its “Formula 1” without excessive dilution. Once the product matured and technological risk dropped, top‑tier VCs became eager to invest.
The Road Ahead
- Explainable AI Leadership: In a world where AI is powerful but opaque, Graphext’s XAI focus builds trust and drives business adoption.
- Generative AI Integration: Not just another feature—AI will act as a co‑pilot, turning natural‑language questions into complete, interactive analyses, solving the “Formula 1” accessibility dilemma.
- Commercial Scaling: The next challenge is expanding sales, especially in the international enterprise market, and turning the platform’s raw power into widespread, user‑friendly value.
TL;DR
- Graphext spent 7 years and ≈€7 M perfecting a browser‑based, Wasm‑powered analytics engine.
- Its no‑code/low‑code platform covers the entire data‑analysis lifecycle while delivering Explainable AI.
- After a patient, grant‑heavy funding strategy, it raised $4.6 M in 2023 and is now scaling commercially.
- The key future work: make the “Formula 1” power accessible to a broader audience through generative‑AI‑driven guidance.
Graphext: Advanced Data Science & Analytics Platform
The tension between power and usability is a constant challenge. Their story is a reminder that there’s no single path to success. Sometimes, patience, technical depth, and a clear vision can be more powerful than speed at any cost. Graphext has built its “Formula 1”; now the real race begins to prove it can win in the global market.
Resources
- Graphext YouTube Channel – YouTube channel with many practical examples of using the Graphext platform.
- Blog Naranja – Graphext, connecting a large volume of data to understand the world (Spanish).
- YouTube – Interview with Victoriano Izquierdo (NTYPodcast – Spanish) – a deep dive into their strategy and technology.
- Business Insider – Graphext, the Spanish startup in which the EU invested €1.7 M (Spanish).
- Graphext Blog – Announcement of the $4.6 M funding round.