Research your team can question, edit, and rerun.
Start with a brief. Signals returns signals and reports with sources, scores, and verdicts. Keep the work in a Workspace and return to it when the decision moves.
Compare model output before you rely on it.
- One brief runs across several models.
- Repeated signals merge, so agreement stays visible.
- Sources, scores, and verdicts stay with each signal.
From question to work your team can use.
Signals finds candidate signals, checks the evidence, and gives planning, strategy, and innovation teams a shared record to build on.
Set up the work
A Workspace, a clear question, and a first pass across several models.
Give the work a place to live.
A Workspace is the home for one research program. It keeps the brief, signals, sources, reports, and structure together as the work develops.
- The Workspace holds your Sessions, signals, structure, and brand.
- Each Session is a focused line of research your team can return to.
- The work builds in one place, run after run.
Get clear on the question first.
A short chat works through the scope, audience, time horizon, and exclusions. That brief guides every later step and your team can edit it when the question changes.
- The chat draws out your objectives, your scope, and what to skip.
- Every later step traces back to the brief, so the work stays on the question.
- Anyone with access can edit it, and earlier versions stay available.
Run several models against the same question.
Each run sends your brief to a selected group of models. When their answers overlap, Signals merges them into one signal and records which models agreed.
- Several models give your team more than one perspective.
- Overlapping answers merge into a single signal.
- You can see where the models reached the same answer independently.
Check the signals
Review the evidence, explore the work, and collect what matters.
See the evidence behind every signal.
A signal includes sources, a verdict on each one, and the reasoning behind its confidence. A source that contradicts the claim stays on the record.
- Each source is marked as supporting, contradicting, unrelated, or unreachable.
- Confidence comes from those verdicts, not from the model's own say-so.
- When sources disagree, the signal says so.
A radar your team can use.
Signals appear on a radar. Quadrants follow your categories and rings use the metric you choose, such as readiness, impact, or urgency. Your team can explore the work instead of exporting a static chart.
- Your categories, metrics, and tags set the layout, defined once for the Workspace.
- Click to read, filter to narrow it down, or copy the URL to share the exact view.
- The same signals also read as a table, card grid, or detailed view.
Collect signals across Sessions.
Signals do not have to stay in the Session they came from. A collection brings together the ones you want from anywhere in the Workspace, for a particular story or audience.
- Gather signals from several Sessions into one collection.
- Collections point to signals rather than copying them, so the work stays in sync.
- Brief an executive, prepare a workshop, or seed the next Session.
Keep it current
The Workspace builds over time, and your team can reshape it when the question changes.
Return to the work when the world moves.
Sessions build up inside the Workspace. Run the brief again next quarter and the radar can add, verify, revise, or retire signals. The activity feed records every change.
- Categories, metrics, tags, and Workflows stay in the Workspace across every Session.
- The activity feed shows what changed and when.
- Set a weekly, monthly, quarterly, or custom cadence.
Change the structure when you need to.
After a run, tell the analyst chat what to change: merge categories, raise the bar for verification, or narrow the time horizon. The radar updates as you work, and the sources remain traceable.
- The assistant can edit signals, sources, reports, and collections for you.
- While a run is going, edits are held so nothing collides.
- Every theme and report can be rebuilt from the signals underneath.
It keeps watch after you close the tab.
Pick a cadence and Signals re-runs the brief on its own, nightly, weekly, or monthly. It keeps an eye on your sources, catches what changed, and updates the radar without you asking.
- Re-runs on a schedule: nightly, weekly, or monthly.
- Between runs it flags new, moved, and dead links.
- Only what changed gets touched. The rest of the Workspace stays put.
What gives your team something to work with.
Several models work from the same brief, and sources show what each candidate signal rests on.
Models work from the same brief without seeing one another's answers. Your team reviews the record, not one model's first answer.
Repeated signals show agreement. Standalone signals stay visible as possible outliers or early warnings.
Make the work your own.
The first run gives your team a starting point. Edit signals, add context, merge duplicates, and keep the reasoning visible.
- Edit a signal's title, summary, category, and scores.
- Add signals the models missed, by hand or from chat.
- Merge duplicates, or unmerge when convergence over-collapsed.
- Organize with categories, tags, and your own collections.
- Delete what doesn't belong, and restore it if you change your mind.
- Steer it all from the assistant; edits persist to the Workspace.
“ The most valuable asset is trend momentum, the speed to generate an accurate visual that scans the latest news. ”
Ideation Manager
Make it fit your method.
Use your own categories, metrics, brand, and radar shape. The research structure matches how your team works.
- Pick your accent, dark surface, and radar geometry.
- Set the Workspace up once and every run inherits it.
- Publish a run publicly, keep it private, or share read-only.
Ready when reviewers have questions.
Bring your own SSO provider, set access by role, and review AI usage by run. Security and procurement teams get the material they need.
Signals records the model and usage details for every run in the Workspace, so finance and procurement can review the line items.
Sign in through your own provider, scope access to your organization, and hand out owner, editor, and viewer roles. Runs and findings all follow the same rules.
Publish a Session, keep it inside your organization, or share a read-only link with a partner. Set visibility at the signal level.
We test the models that power Signals.
34 models ran the same 12 industry briefs. We score verifiability, specificity, currency, and coverage. The methodology and data are open, and we refresh them quarterly.
Questions
Can I customize the categories and framework?
Yes. Scans default to a five-category STEEP framework, but the categories, the research brief, and the assessment metrics are all yours to change. Tune them to your industry, or swap in the lens your team already works with.
Can I track a theme over time?
Yes. A Workspace keeps the same brief, categories, and method, and you take snapshots over time, so you can see which signals are heating up and which are cooling, instead of starting from scratch each scan.
How does sharing work? Can I share a radar with people who don’t have an account?
Yes. Each Workspace has viewer, editor, and admin roles, and you can share a radar through a public explore link or a private share link. Read-only viewers and invited guests do not count against your seats.
Can I edit or correct what a scan returns?
Yes. Generated signals are a starting point, not a locked result. You can edit a signal’s title, summary, category, and scores, add ones the models missed, merge duplicates or unmerge them, and delete what doesn’t belong. You can make the same edits by asking the assistant in chat.
Does Signals run our whole innovation or strategy process?
No, and that’s deliberate. Signals works the sensing phase: discovering, grounding, and verifying external signals of change. It gives planning, strategy, and innovation teams a sourced record to build on. It does not replace the people or the process downstream.
See it on a question you already have.
Your first scan is on us.