Skip to content

Our mission

We believe that the future of the small business economy will be driven by data.

Enigma builds comprehensive intelligence about the identity and financial health of small and medium businesses across the United States — engineering better data from hundreds of public and third-party sources, so that companies of every size can access the financial services they need to grow and thrive.

See the data ▸Open roles
100M+

registrations resolved

30M+

brands in the graph

2B+

relationships mapped

99.2%

pairwise precision

At our core

01

Evidence over assertion

Every record carries the source it came from. Where we can't point at the evidence behind a claim, we don't publish the claim.

02

Observed, not self-reported

Revenue is read from card transactions — a panel covering 40% of U.S. card volume across the top 60 issuing banks — not from what a business writes on an application.

03

Resolution before the query

Hundreds of sources are reconciled into one profile per business before anyone asks a question. You query the answer, not the raw feeds.

04

Auditable by default

Every decision links back to the records that produced it, so a reviewer, a regulator, or a model can retrace the path.

Our data

One profile per business, built from the record.

We apply next wave data science and proprietary machine learning to hundreds of online and offline sources to build the identity, activity, and risk profile of every small business.

Identity, activity, and risk are read off the same resolved profile rather than three disconnected systems. That is what makes answers comparable across a portfolio — and what keeps disagreement between sources visible instead of averaged away.

1B+

screening requests a month

40%

of U.S. card volume observed

Top 60

issuing banks in the panel

Source: hundreds of public and third-party sources, resolved into one profile per business.

Our people

The work is the data.

Enigma is a data company. The people here build the sources, the resolution, and the evidence behind every record on this site. Roles are posted to our job board as they open — that board is the full list.

See open roles →

The idea, as filed

We wrote it down in 2013.

A provisional application in February 2013, a filing in 2017, a grant in 2021. The drawing on sheet one is the pipeline this site still describes — many sources in, one shared index built before the question is asked, then the query.

It is here because it is the one claim on this page that does not rest on our telling you so.

US 10,963,513 B2 — sheet 1 of 15

Filed 2017 · provisional 2013 · granted 2021

Sheet 1 of US patent 10,963,513 B2, reproduced. Figure 1A shows six numbered stages in a vertical flow: source data set acquisition (105), data warehouse storage (110), data post processing and high level index (115), low level data indexing (120), data querying (125), and output results and user interface (130), with a return path from the output back to querying.

FIG. 1A, transcribed

  1. 105Structured and/or unstructured source data set acquisition into at least one system
  2. 110Data warehouse storage
  3. 115Data post processing, high level index
  4. 120Low level data indexing
  5. 125Data querying
  6. 130Output results & user interface

And back: 130 returns to 125. Querying is not the last step in the drawing either.

Stage 115 is the one that matters: a single high-level index shared across every object, built before anyone queries. That is the same order of operations the rest of this site describes.

Inventors Marc B. DaCosta and Hicham Oudghiri. Assignee Enigma Technologies, Inc. 12 claims, 15 drawing sheets.primary document

21 of the top 30 U.S. banks use Enigma.

See what the record says.

Bring a sample of the businesses you decide on. We'll show you the evidence behind each one.

Request a demo →