> ## Documentation Index > Fetch the complete documentation index at: https://docs.getvector.me/llms.txt > Use this file to discover all available pages before exploring further. # Feature dossier > Find out everything known about one feature, fill the gaps with research, check how each competitor does it, and decide whether it belongs in your product. Every feature in the [catalog](/features/features-in-domain) has a dossier: one page with everything known about that capability across your market. It's where you go to answer three questions about a feature β€” what do we actually know about it, how do competitors do it, and should we build it. To open a dossier, click a feature's name anywhere in Vector: in the catalog, on the product map, on a delivery card. A feature dossier with the feature name, Merge and Archive, the row of section pills, and the Dossier tab showing the Where it lives card, the Do we have this status buttons with Plan delivery, and Notes ## Find out what's known A dossier is split into eight sections, each answering one question about the feature. Switch between them with the row of pills under the feature's name. A pill with a check mark has content; one without is still empty. | Section | What it tells you | | ----------------------- | ---------------------------------------------------------------------------------- | | πŸ”¬ Deep how-it-works | How each competitor actually implements it: scope, limits, their own terminology | | πŸ‘₯ Public Reviews | What users praise and complain about, from public reviews, forums and social media | | πŸ—£ Meeting analysis | What your own customers said about it in calls and tickets | | πŸ“ˆ Trajectory | How it has changed in competitors' changelogs β€” where the market is taking it | | πŸ“Š Competitor coverage | Who has it, how good their version is, and how well that's backed | | 🎨 UI references | The screens and flows competitors put it in | | πŸ›‘ Re-verify evidence | Whether existing evidence still holds against the live sources | | 🎀 User interview guide | Questions to ask users about it, built from everything above | The same eight sections are the **Known** dots in the catalog, so you can see from the list how much you know about each feature before you open it. The summary under the feature's name β€” the TL;DR β€” is the short version; click **edit** to rewrite it in your own words. ## Fill in what's missing Open an empty section and it tells you what it's for, with a button to fill it β€” **Research**, **Score** or **Re-verify**. The research runs in the background, so you can leave the page while it works, and **Stop** cancels it. Once it's done, the button reads **Refresh** so you can bring the section up to date later β€” or **Retry** if the run failed. Each run uses your [AI key](/getting-started/connect-a-provider). To research the same section for many features at once, select them in the catalog instead. A few sections fill differently: * **πŸ—£ Meeting analysis** fills on its own when meetings you add to [Meeting analysis](/features/meeting-analysis) mention this feature. * **🎀 User interview guide** is built when you need one: pick a **Template** and a **Language**, click **Build the guide**, and **Copy** the result. * **πŸ‘₯ Public Reviews** keeps praise and complaints on separate tabs. Once there's both, **Summarise** on the **✦ Summary** tab condenses them. ## See how each competitor does it **πŸ“Š Competitor coverage** is where you check a competitor's version of the feature. For each competitor it shows whether they have it, their own name for it when it differs from yours, a 0–10 score of how good it is, and how strong the evidence is: | Evidence | What it means | | -------------------- | ---------------------------------------------------------------------- | | Strong, Medium, Weak | How well the claim is backed, judged by the kind and number of sources | | no evidence | Nothing backs it yet β€” treat the status as an assumption | Hover the badge to see what kinds of sources back it. A claim with weak or no evidence is the one to verify before you rely on it. The same data makes up this feature's row in the [comparison](/competitors/compare-competitors) of all your competitors. ## Correct what research got wrong Research can be wrong, and your knowledge wins: * **A competitor's status or score.** In **πŸ“Š Competitor coverage**, click **Edit** on their row. Set the status, describe **What exactly they have**, and set a score if you know better. A score you set by hand stays yours β€” later runs won't overwrite it. * **Deep how-it-works and Trajectory.** Edit the text directly. Your version is kept alongside what the run found: you can switch between the two, and **Drop my edits** goes back to the run's version. ## Decide whether it belongs in your product **Overview** β†’ **Dossier** is where your own decisions about the feature live. Under **Do we have this?**, set your status with one click: **not assessed**, **we do not have it**, **on our roadmap**, **partially** or **we have it**. It's the same status the catalog and the comparison show. If you don't have the feature yet and want it, **Plan delivery** marks it **on our roadmap**. That records the decision; to start the actual work, plan a [delivery cycle](/delivery/plan-a-delivery-cycle) from the **Delivery** tab. Both are reversible. ## Place it on the product map Under **Where it lives**, pick a **Cluster** and, optionally, an **Area** inside it, and click **File it here**. **Directly in the cluster** is a valid choice. If the area you need doesn't exist yet, type a **New area name** and click **Create** β€” the area is created and the feature is filed into it. See [Product map](/features/product-map) for how clusters and areas work. ## Keep your own notes and files Some things research can't tell you: why a feature is parked, what the team decided, a doubt someone raised. Write them in **πŸ“ Notes**, on the **Dossier** tab β€” **Add a note**, then **Edit** to change it. It's the one part of the page that's yours rather than the model's. The other tabs of **Overview** collect material about the feature from elsewhere: | Tab | What you'll find | | -------- | --------------------------------------------------------------------------------------------------------------------------- | | Files | Your own material: specs, screenshots, recordings. See [Files in a dossier](/competitors/dossier-files) | | Delivery | The delivery cycles planned for this feature | | Voice | Evidence from customer meetings that matched this feature | | Archive | Passages from competitors' help centers and changelogs that were [archived](/competitors/content-archive) and matched to it | | Activity | What happened to the feature, and when | ## Rename, merge or archive a feature * **Rename** β€” hover the feature's name and click **rename**. Links to the old name keep working. * **Merge** β€” if another feature is the same thing under a different name, fold it into this one with **Merge…** next to the name. The **🧩 Possible duplicates** tab lists likely matches. See [Duplicates and merging](/features/duplicates-and-merging). * **Archive** β€” if the feature doesn't matter to you, **Archive** takes it out of the catalog and the comparison but keeps its history. An archived feature opens only from a direct link, where **Unarchive** brings it back. ## Where to go next The catalog these dossiers belong to. Where new features come from.