Fin360AIForm 5500 signals
Sign inOpen the lists

The data

One federal filing, and everything that comes out of it

Every workplace benefit plan covered by ERISA has to file a Form 5500 with the Department of Labor each year. The Department publishes those filings as a public government record. That is the whole source.

This page is what a careful buyer asks for before the demo: exactly what we hold, exactly what we do not, and where it runs out.

Where it comes from

And what that means for you, legally and commercially.

A public US government record

The Form 5500 files, downloaded from the Department of Labor’s own file service. Public record, published by the agency that collects it.

Not licensed, not scraped

Nothing here is licensed from another vendor, scraped from a competitor’s product, or bought from a data broker. No third party can tell you what you may do with it, and no supplier can take it away.

You could download it yourself

Anyone can. The hard part is not the download. It is keeping track of which plan and which provider is which across sixteen years, and deciding what counts as a plan actually having done something. Those are the parts documented in the most detail here.

What we hold right now

3.47 GB of compressed files from the Department of Labor, across 13 data files, filing years 2009–2025. Read from the same data the product serves.
DatasetForm yearsWhat is in it
55002009–2024The main annual return. Employer, plan, how many people are in it, how it is set up.
5500_SF2009–2024Form 5500-SF, the short form smaller plans file. A separate file with its own column names, not a slice of the main return. Treat it as one and you lose every plan that crossed the size line.
OSHA_300A2024–2025Form 5500 structured dataset.
SCH_A2009–2024Form 5500 structured dataset.
SCH_A_PART12009–2024Form 5500 structured dataset.
SCH_C2009–2024Schedule C header. A header record only: the provider names and compensation are not in this file.
SCH_C_PART1_ITEM12009–2024Schedule C Part 1 Item 1 — service provider identification.
SCH_C_PART1_ITEM22009–2024Schedule C Part 1 Item 2 — where provider names and compensation actually live. This is the service-provider relationship table.
SCH_C_PART1_ITEM32009–2024Schedule C Part 1 Item 3 — indirect compensation detail.
SCH_C_PART32009–2024Schedule C Part 3 — terminated accountants and actuaries.
SCH_H2009–2024Schedule H, the financial statement large plans file. Assets, running costs, and the fee history the fee list is built on.
SCH_I2009–2025Form 5500 structured dataset.
SCH_SB2009–2025Form 5500 structured dataset.

What we deliberately leave out

Worth saying out loud, because a claim this product does not make is one you should not assume it makes.
  • Attachments, including the Schedule of Assets. They sit behind a separate request and nothing here depends on them. It matters because when someone in this market tells you what a plan is holding, that attachment is usually where it came from. Fin360AI does not make that claim.
  • Anything about individual people. Form 5500 describes the plan and the employer sponsoring it. There are no records about individual participants in the source and none in the product.
  • Defined benefit plans, for now. The current lists cover defined contribution plans only. A check on every refresh confirms that filter actually removed something, rather than that the result looks about right. A filter that quietly does nothing looks exactly like one that works.
  • Asset managers selling funds into plans. They are paid by the fund rather than by the plan, so they never appear on a Schedule C. We cannot count them from this data, let alone build a list for them.

Why the newest year is never complete

And why we label it rather than hide it.

Filings arrive nine to eleven months after a plan’s year end, so a year is still filling in long after it has ended and the newest year we hold is never complete. Every list says how much of that year has come in, rather than showing you a half filed year as if the market were shrinking.

The most recent year we can check an outcome against is 2022. To know whether a plan changed recordkeeper in a given year, next year’s filing has to exist. That is why the testing stops where it does. It is how the filing calendar works, not a choice we made.

The Department quietly folds amendments into the same download link, so a file can go out of date without its year changing. On every refresh we ask the Department’s own server what each file’s size and date are now, and re-download only what has actually changed.

The thing most likely to go wrong is identity, not math

Five separate bugs in this project came back to the same cause. It is worth understanding before you trust any number here.

When an ID number changes, it looks like the plan itself changed

  • The plan. A new employer ID or plan number looks like the plan shut down. About four in ten plans that seem to vanish have simply fallen below 100 participants and moved to the short form. They are alive, and we account for the large majority of those disappearances rather than counting them as closures.
  • The provider. A provider changing its own tax ID looks like it grew several hundred times over in one year.
  • The event itself. One provider buying another looks like thousands of plans deciding to switch on the same day. In one year it was a single firm renaming itself. Those years are taken out of the count rather than recorded as plans that stayed put, because an acquired plan might have been switching anyway.

For a product whose whole job is spotting change, keeping track of who is who is a bigger part of the build than the modeling. The testing page lists the checks that keep it honest, and how they stand on the current build.

Where it runs out

The things we would rather you read now than find out after you have paid.

A plan closing and a plan being absorbed look the same

In the filings today we cannot tell them apart. Both are worth a call. They are not the same call, and the one that is an opportunity rather than a loss is not something we can point you at yet.

We know who serves a plan, not who earns the most from it

A provider counts as serving a plan when it appears on that plan’s Schedule C. Ranking providers by what they are paid moved around too much from year to year to rely on.

How often these things happen keeps moving

On one list, how often the event happens shifted noticeably between the years we learned from and the years we tested on. Any model gets worse as the world moves away from the years it learned on, and this is a market with consolidation waves and rate cycles that do exactly that.

Data last refreshed 2026-10-18. Files last checked against the Department’s copy 2026-10-18.