For founders

They shipped. Did it actually land?

Every competitor tool tells you something happened. That's the easy half. Track the handful of companies you really race against and see which of their moves the market cared about — and which ones went out to silence.

Deterministic scoring — no model in the numberEvery line cited to a source URLRead-only — nobody is notifiedFree to start, no card
01The problem

A feed tells you what shipped. It can't tell you what mattered.

A funding round announced into a skeptical room and a quiet feature ship developers adopted immediately look identical in a changelog feed. They are not remotely the same event. Sorting by recency is how you over-react to noise and miss the move that counts.

A recency-ordered feed of competitor events beside the same events re-ranked by the public reaction each one earned — the top item in the feed drops to the bottom once reception is measured.
the same week, ordered by recency and then by what landed
02Your set

Eight companies, watched properly.

Add the companies you actually compete with. The engine watches their releases, pricing, hiring and public repos on a schedule, diffs every observation against the last one, and scores what changed. Nothing is inferred — a move exists because two snapshots disagree.

ido.app / dashboard · your competitive set
IDØ dashboard showing a founder's tracked competitive set, each move scored as signals arrive
your set, scored as the signals arrive
03The opening

The most useful thing we find is a rival's silence.

When a competitor ships something genuinely significant and the market says nothing, that gap is the single most actionable signal in the product. It means the need wasn't felt, the framing missed, or the lane is still wide open. You only ever see it if reception is measured — a feed reports that launch as a win.

A high-scoring competitor launch plotted against the public reaction it received, where the reaction stays flat — the gap between significance and silence marked as an opening.
high significance, no reception — the gap is the opportunity
04What you get

Four things, and honest labels on the rest.

your_set
8 competitors
live

Your real set

The companies you actually race against — not a 400-row market map.

did_it_land
HN · Bluesky · GitHub
live

Reception

The reaction each launch earned, coupled to that specific launch.

friday_brief
ranked by impact
live

Weekly brief

What shipped, what landed, what it means — one forwardable page.

alerts
high-signal only
soon

Alerts

Pushed only when a move is genuinely worth interrupting you.

05Boundaries

What we won't do — and why that's the point.

Most of these are things a competitor-intelligence tool could ship tomorrow and charge for. Each one would cost the thing that makes the output worth reading.

01

No verdicts

We don't tell you to build it or kill it. You get evidence and the call stays yours.

02

No LLM in the score

The number comes from the diff and the measured reaction. Same inputs, same number, every time.

03

No uncited claims

Every line carries its source URL and timestamp. If we can't cite it, we don't show it.

04

No scraping LinkedIn

Hard constraint. We never hold your credentials for anything.

05

No notifying anyone

Nobody is told you're watching them. Read-only by architecture, not by policy.

06

No selling your data

Your tracked set is yours. It is not a dataset we resell.

06Questions

Straight answers.

Is a language model making any of this up?
No. The impact score is deterministic — it comes from the diff between two observations plus the measured public reaction, and every line links to the page it was read from. A model writes the weekly brief's prose at the very end; it never produces the number.
How many competitors should I track?
Around eight. The engine is built for the set you genuinely race against, and a tight set is what makes the weekly brief readable. Tracking forty companies gets you a feed you'll stop opening.
What if my competitors are quiet?
That's the useful case. A rival shipping something significant to no reaction at all is the clearest opening you'll get — it means the market didn't care, and the lane is still open. The engine surfaces those explicitly rather than burying them under whatever was loudest.
How fast do I see something?
The engine is a change detector, so the first run records baselines and the moves start appearing on the runs after it — typically within a day. Reception builds over the following days as reactions accumulate.
Does anyone find out I'm tracking them?
No. Everything is read from public sources, nothing is written anywhere, and no company is ever notified. There is no mechanism in the product to do it.
Your lane

Point it at the eight companies that matter.