Why VoyanceHQ’s Data Warehouse Ambitions Faded in Africa

In 2019, Africa’s tech ecosystem was seeing an unprecedented surge in transaction volumes. Fintechs, digital lenders, and e-commerce platforms were generating vast amounts of consumer data. Yet most African enterprises still couldn't use predictive analytics. Traditional businesses relied on fragmented legacy databases and isolated spreadsheets, and couldn’t afford the scarce, expensive data science teams needed to clean, structure, and model their information.

Founded in Lagos by seasoned machine learning engineer Ebot Tabi (who previously worked as a senior data engineer at the $60 billion quantitative hedge fund Two Sigma) alongside early co-founder Abdulhamid Hassan, Voyance set out to democratize enterprise data science. The thesis was ambitious: abstract away the brutal engineering overhead of big data. Through flagship tools like OMNI (an automated machine learning engine) and Manhattan DB (a scalable cloud data warehouse), Voyance promised to turn everyday business analysts into predictive data scientists with a few clicks. Later, recognising the acute fraud crisis plaguing Nigerian digital finance, the company launched Sigma—a graph-based industry alliance that lets fintechs share bad-actor identifiers (BVNs, phone numbers, and IPs) in real time.

Voyance's Early Traction

Voyance’s technical depth immediately set it apart from typical consumer-facing clones, attracting high-signal institutional backing. In February 2022, the startup announced a $500,000 pre-seed round led by Beta.Ventures, with participation from Zedcrest Capital, HoaQ Club, Assembly Investors, and prominent angel operators including Fola Olatunji-David, Seni Sulyman, and Nelly Chatue-Diop.

Shortly afterwards, in the spring/summer of 2022, Voyance was selected for the prestigious Techstars New York City Accelerator, bringing its total disclosed funding to roughly $520,000 (with private tracking estimates reaching up to $620,000). At its height, Voyance expanded its corporate footprint between Lagos and California, growing its engineering and operations team to over 40 employees while pitching itself as the foundational data layer for modern African enterprises.

Issues with Scaling Enterprise Data Infrastructure

Despite world-class engineering leadership, Voyance encountered severe structural friction when attempting to scale B2B enterprise data infrastructure across emerging markets:

1. The "Garbage In, Garbage Out" Reality: Automated machine learning and predictive analytics require clean, structured, and accessible historical data. Voyance found that many target African enterprises and mid-market firms lacked basic digital data hygiene. Before selling predictive AI, Voyance had to provide extensive, custom data-engineering hand-holding—turning a scalable SaaS product into slow, consultative integration work.
2. The Collaborative Trust Deficit (Sigma): While Sigma’s shared fraud-tracking database was technically brilliant, Nigerian fintechs and banks fiercely protected their proprietary customer data and were wary of regulatory exposure under strict data privacy frameworks (such as the NDPR), slowing adoption of network effects.
3. Competition from Global Hyperscalers: By 2023–2024, global cloud giants such as Snowflake, Databricks, AWS, and Google Cloud (BigQuery) aggressively expanded their turnkey AI and data warehouse tooling. Convincing CTOs to entrust an early-stage startup with core enterprise data warehouses—rather than an established hyperscaler—proved an uphill battle.
4. The Series A Vacuum: Deep-tech and enterprise infrastructure startups require long, patient capital cycles to navigate lengthy B2B sales pipelines. When the global venture capital winter froze follow-on funding for African B2B startups between 2023 and 2025, Voyance’s ~$520k pre-seed war chest came under immense strain.

The Shutdown

Voyance did not suffer a dramatic public implosion; instead, it experienced the quiet operational fade characteristic of enterprise SaaS casualties. Following its 2022 accelerator run, public product releases, developer documentation updates, and social media communications gradually stalled.

By February 2026, private venture intelligence platforms including PitchBook officially reclassified Voyance’s corporate status to "Out of Business" (with regional trackers like Lucidity Insights similarly flagging the company as Inactive). While the legacy developer documentation portal remained cached online, active commercial operations and enterprise onboarding ceased, quietly closing the chapter on one of Nigeria’s earliest pure-play big data and machine learning infrastructure bets.

Post-Mortem Analysis: Lessons for Entrepreneurs

Build for the Market’s Actual Data Maturity: Selling advanced machine learning and data warehousing in frontier markets often fails because prospective clients are still grappling with foundational digitisation. In emerging ecosystems, startups that solve immediate, transactional pain points (such as payments or reconciliation) gain traction faster than pure data-infrastructure layers.

Beware the "Hyperscaler Crush" in Developer Tooling: Building proprietary databases or cloud data warehouses (such as Manhattan DB) puts a lean startup in direct competition with multi-billion-dollar global infrastructure giants. Without a hyper-localised regulatory moat or a proprietary dataset that global tools cannot replicate, local infrastructure startups struggle to defend enterprise contracts.

Enterprise Sales Cycles vs. Pre-Seed Runways: Selling core infrastructure to African banks and large enterprises typically takes 9 to 18 months per deal. Raising ~$520,000 in pre-seed capital leaves little margin for error when navigating slow corporate procurement and compliance reviews.