Dashboard
The Content Engine dashboard answers two questions: What did the last audit find on my listings? and What is waiting for my decision? Three numbers for the findings, one for the proposals, and below them the timestamp of the last run. Every number is clickable and opens exactly the records it counted.
There are deliberately no settings here — those live on the setup card.
How to get there: In the merchantCENTRAL Setup under Extensions, click Content Engine — as with the Price Engine, the click opens the dashboard (Set Up next to it still opens the settings). Also via Tell me Content Engine Dashboard and via the Dashboard action on the setup card.
Why nothing is audited
This group appears only when it applies. It names the first gate the engine currently fails, in exactly the order the engine hits them:
| Order | Message | What to do |
|---|---|---|
| 1 | The Content Engine is switched off | Switch Enabled on, on the setup card |
| 2 | No active license | Activate the module CONTENTENGINE or start a demo — the License FactBox |
| 3 | There are no marketplace listings yet | Connect a marketplace or shop and list items on it; what is audited are listings, not items |
| 4 | The scheduled job queue entry is missing or on hold | Create Job Queue Entry on the setup card |
Only the first open gate is named: it is the only one that can be fixed first. The group disappears once nothing is in the way any more.
Findings of the last audit
Three numbers, each a drill-down into the findings list, filtered to its severity:
| Number | What it counts | Colour |
|---|---|---|
| Errors | Findings that block ranking: a missing title, meta title, meta description, description or image — and two listings behind one URL slug | Red while it is above zero |
| Warnings | Findings that hurt ranking: texts too long or too short, duplicates, an invalid slug, a missing category | Amber while it is above zero |
| Hints | Findings that are merely nice to have: keywords, slug, brand or EAN not maintained | neutral |
The numbers are not a history: the findings table always shows the state of the last audit per listing. A corrected field loses its finding as soon as the listing is audited again — at the latest with the daily run, immediately via Audit This Listing Again or Audit SEO on the marketplace item card.
Content proposals
| Number | Meaning |
|---|---|
| Waiting for review | Drafted content proposals nobody has read yet — status Proposed. The click opens them |
A waiting proposal is not a problem, it is the engine having done its part. While it is open, every batch skips it: no second draft for a listing whose first one is still unread.
Last audit
| Field | Meaning |
|---|---|
| Last Audit Run | When the audit last ran. If the timestamp is more than a day old while no notice is shown above, the Job Queue Entries are worth a look |
| Last Audit Items | How many listings that run checked — the test of whether the scope is right (Audit Offline Items) |
| Last Audit Findings | How many findings it produced in total |
Beside them, the License FactBox shows the state of the module. It is read from the cache when the page opens — it is refreshed on the setup card.
Actions
| Action | Effect |
|---|---|
| Run SEO Audit | After a confirmation, audits every listing in scope and replaces its findings with the current result — the same run the job queue entry starts on its schedule, started by hand. At the end a message reports listings checked, errors, warnings and hints |
| Draft Content for Errors | Asks Copilot for new texts for every listing with an error finding, up to the run limit in the setup. The drafts land as proposals for review; nothing reaches a channel before you apply it |
Under Navigate you find Content Proposals, All Findings and Marketplace Items. The last one is the path for targeted runs: filter the marketplace item list — one shop, one category, the items put online since Monday — and start the audit on that selection from there.
The daily routine
- Is there a notice at the top? If so, fix that first — everything else is waiting on it.
- Open Errors and work through them: those are the listings a search engine cannot rank as they are.
- Decide the ones Waiting for review — apply or reject the proposals. Applied texts go out with the connector's next item update.
- Scan the Warnings: duplicates first, they always concern at least two listings.
- Hints are stock for quiet days — keywords, EAN, brand.
On the hub dashboard
The Content Engine also reports into the waiting for you card of the hub dashboard, with two rows:
| Row | Tone | Click opens |
|---|---|---|
| Content Engine · SEO errors on listings | Attention | the error findings |
| Content Engine · content proposals waiting for review | neutral | the open proposals |
Warnings and hints stay on the engine's own dashboard: they are work to plan, not work waiting. Both rows appear only when the module is set up and Enabled.