The state of fund data in 2026

The fund data industry is caught in a vice. Regulation keeps expanding, data volumes keep growing, and headcount stays flat. The gap between what teams need to process and what they can actually handle is widening every quarter.

The numbers

Let's ground this in specifics rather than vague trend statements.

Regulatory templates have multiplied. In 2018, a European fund distributor needed to manage EMT data for MiFID II compliance. Today, the same distributor handles EMT, EPT (PRIIPs), EET (SFDR/ESG), and increasingly TPT (pension-specific templates). Each template has dozens to hundreds of fields. The EET alone has over 600 data points. That's four times the regulatory data surface area in six years, with more coming.

Fund product counts keep climbing. EFAMA's latest figures show over 60,000 UCITS funds in Europe, plus tens of thousands of AIFs. Each fund has multiple share classes — a typical UCITS umbrella might have 20 sub-funds with 5-10 share classes each. That's 100-200 entities per umbrella, each needing its own NAVs, static data, and regulatory filings.

Data sources are fragmenting. A mid-size asset manager might receive data from their transfer agent, custodian, fund accountant, benchmark provider, ESG rating agency, and three different market data vendors. Each source has its own format, its own delivery schedule, and its own idea of what a "fund identifier" looks like. One sends ISINs, another sends SEDOLs, a third uses internal codes that require a mapping table nobody has updated since 2023.

The people problem

Here's what's actually happening on the ground. I talk to fund data teams regularly, and the pattern is consistent:

• Teams of 3-5 people are managing the same data volume that required 3-5 people five years ago — but with twice the regulatory requirements
• Senior ops analysts spend 60-70% of their time on data wrangling tasks that don't require their expertise
• New hire ramp-up takes 6-12 months because the processes are undocumented and live in people's heads
• The bus factor on most fund data teams is effectively 1 — one person's absence causes visible operational risk

Hiring more people isn't the answer, even if budgets allowed it (they generally don't). The work is growing faster than you can hire, and each new person needs months of domain training before they're productive. The maths simply doesn't work.

Where automation actually is

The industry talks a good game about automation. The reality is more mixed.

Well-automated: NAV dissemination to major platforms (Bloomberg, Morningstar, Refinitiv). If your fund is on a major platform, the daily NAV probably flows automatically. This is a solved problem for the big players.

Partially automated: Regulatory template generation. Tools exist that can populate EMT/EPT/EET templates, but they still require significant manual input for the fund-specific fields. The template structure changes with each regulatory update, breaking automations that weren't built to be flexible.

Barely automated: Source data ingestion and normalisation. This is the messy middle — accepting data from dozens of sources in dozens of formats and getting it into a consistent canonical form. Most firms still do this with Excel, VBA macros, or fragile Python scripts. It's the highest-value automation target in the industry and the least addressed.

Not automated at all: Exception handling and data quality remediation. When a NAV looks wrong, when an ISIN doesn't match, when a file arrives with a new column — these events still trigger manual investigation workflows that run on email, chat messages, and institutional memory.

What's actually changing

Two shifts are worth watching:

AI-assisted data operations. Not the "AI will replace your team" narrative. The practical version: AI that can look at a new CSV and suggest column mappings, flag anomalous values, and generate validation rules. This reduces the time cost of onboarding new sources from days to hours. We're building this at Kairo, and we're not the only ones thinking about it.

Standard convergence. FinDatEx is slowly consolidating the European regulatory template landscape. Openfunds adoption is growing for static fund data. ISO 20022 continues its march through securities messaging. None of these individually solve the problem, but the direction is toward fewer formats and more structure. That makes automation more viable.

The fund data industry doesn't need a revolution. It needs the automation that every other data-intensive industry adopted ten years ago, adapted for the specific constraints of regulated financial services.

The gap ahead

Here's my prediction for the next three years: the firms that invest in genuine data operations automation — not just dashboards on top of manual processes, but actual end-to-end pipeline automation with proper error handling — will pull away from those that don't. The regulatory burden will continue growing. The firms that can absorb that growth without proportional headcount increases will have a structural cost advantage.

The firms that are still running on Excel and email in 2028 will find themselves in an increasingly difficult position. Not impossible — Excel is remarkably resilient — but increasingly expensive in terms of operational risk, staff burnout, and regulatory exposure.

The tooling to close this gap exists now. The question is adoption speed. Based on what I've seen, the answer is: slower than it should be, but faster than most people expect.