Nora WhitfieldCo-founder & CEO
Ten years in product at logistics and fintech companies. Started Cadence after her fourth failed metrics migration.
Skip to content We got tired of
+38 engineers, designers, support and sales — in 11 countries 01One number, one owner 02Quiet by default 03Your data leaves whenever you want 04Priced on what we cost 05Support is not a department 06Boring infrastructure Ship on Monday.
About
We got tired of
analytics theatre
Three people who spent a decade building dashboards nobody opened, trying to make the opposite thing.
In 2023 we were running product at a company with four analytics tools and no agreed definition of an active user. Every review started with twenty minutes of reconciling numbers, and by the time the numbers agreed, the meeting was over.
The tools weren't broken. They were built for a different job — for analysts, working on questions that could wait a week. Product teams don't work that way. They ship on Tuesday and want to know by Thursday whether it helped.
So we built the thing we wanted: one definition of every metric, answers in under a second, and a tool that speaks up on its own rather than waiting to be opened. We shipped the first version to eleven friendly teams in 2024.
Today Cadence is used by 2,400 product teams, we're profitable, and we have no plans to sell reporting seats to people who will never log in. If a feature makes the tool louder rather than clearer, we don't ship it.
0+product teams on Cadence
0people, across 11 countries
0year the first version shipped
0moprofitable, without a growth round
Who's behind it
Small on purpose. Everyone here has shipped product, and everyone does support rotation — including the founders.
Elias VanceCo-founder & CTO
Built query engines at two database companies. Wrote the first version of the columnar store on a train.
Zami HolmesHead of Product
Employee number three. Owns the thing everyone else argues about: what stays out of the product.
What we actually believe
Four rules we've turned down real revenue to keep.
Every metric in Cadence has a person's name on it. Ambiguity isn't neutral — it always resolves in favour of whoever argues loudest.
A tool that notifies you about everything has taught you to ignore it. We ship fewer alerts and work hard on making the remaining ones right.
Full export on every plan, including Free, in a format other tools can read. Lock-in is a business model we'd rather not have.
We charge for events because events are what we store and query. Charging per seat would mean charging you for sharing, which is the opposite of the point.
Every engineer does one support day a fortnight. The fastest way to stop shipping confusing features is to answer questions about them.
We run on three regions, one database engine and no service mesh. Uptime comes from having less to break, not from a better incident process.
How we got here
2023
The fourth migration. Nora and Elias spend a quarter moving a company between analytics tools, and end up with the same disagreements in a new interface.
2024
First version. Eleven teams, one region, no billing code. The columnar engine handles 40M events and everyone is surprised.
2025
Seed round and EU region. $6.4M from Atlas Grove and a group of operators. Frankfurt goes live for customers who can't send data to the US.
Early 2026
Cadence 2.0 and profitability. New engine, plain-English queries, and the month where revenue finally covered the bill.
Today
2,400 teams, 41 people. Session replay shipped in August, and the roadmap for the rest of the year is public.
Backed by people who've run product
One institutional round, and a long list of operators who were customers first.
Atlas Grove
Northline
Fieldnote
Harborlight
Second Draft
62 operators
Get started
Ship on Monday.
Understand it by Tuesday.
Free for 30 days on the full product. Five-minute install, no card, no onboarding call unless you want one.
- SOC 2 Type II
- EU & US data residency
- Cancel in one click