Overview
A run prepares researched listing suggestions for your selected markets. You can follow its progress from the Agent page or leave and come back.
Proposals appear as each market finishes rather than all at the end.
Starting a run
Start a run from the agent page once setup is complete. It continues in the background if you close the tab. Return to the Agent page to check progress and review the results.
If a run cannot finish, check its displayed status before starting another. Contact support if you need help with an unfinished run.
Progress
A running agent shows a checklist, and ticks each line as it finishes.
- Reading your store metadata
- Loading your keyword research
- Comparing against competitors
- Finding keyword gaps
- Ranking opportunities
- Drafting metadata suggestions
- Estimating search-visibility gain
If you granted the screenshots scope, a final Rendering localized screenshots line appears once that work starts. The checklist keeps you informed while the suggestions are being prepared.
Run duration varies with the selected markets and work. You can leave the page or reload it: the run continues in the background and the checklist picks up where it was.
Generate suggestions
Generate suggestions opens the same market picker as setup. Choose up to five storefronts — ones you already run, new ones, or a mix — to request fresh suggestions within the scope you allowed. The full-screen progress checklist stays up until this pass finishes; a reload while it is still running brings that screen back.
Markets are a per-run choice: this week's five can be entirely different from last week's, and a market you do not pick keeps its suggestions in the inbox untouched.
Run cadence
Nothing runs on a schedule. Every pass starts when you press Generate suggestions (or ask in chat).
Once seven days have passed since your last pass, the inbox shows a reminder that fresh suggestions are worth pulling — a week is enough time for store rankings to move.
After approved metadata reaches the store, Spek keeps that version stable for a seven-day measurement window. The Agent page shows the remaining days and hours. When the window closes, the next pass you start checks for stronger keyword opportunities; it may still decide that your current metadata is best.
Every pass produces proposals the same way, and every pass never publishes anything to a store. Its output waits in the inbox until you look at it. See what the agent will never do.
Common questions
How long does a run take?
Duration varies with the selected markets and work. The Agent page shows progress, and you can return later without keeping the tab open.
Check whether the selected markets are still progressing or whether a specific failure needs attention. If you contact support, include the app, markets and approximate start time. Keeping the first run identifiable is more useful than starting repeated requests without understanding its current state.
Can I close the tab?
Yes. The run continues in the background and the checklist picks up where it was when you return.
Do suggestions refresh automatically?
No. You start every pass yourself. After seven days without one the inbox shows a reminder, and after approved metadata goes live the seven-day measurement window tells you when a fresh pass is worth starting.
When will new keyword suggestions appear?
The Agent can check again after the published metadata has had a seven-day measurement window. The countdown appears in the inbox. Finishing the window does not guarantee a new card: if no stronger option wins, the Agent keeps the current metadata.
What should I do when a run is taking longer than expected?
Check the current progress and any message on the Agent page. Duration depends on the selected markets and work, so the most useful signal is the recorded state rather than how long the browser tab has been open. You can leave and return while a run continues. If it reports a failure or stops making progress, record the app, markets, approximate start time and visible message for support. Starting several overlapping requests without checking the first one can make it harder to understand which result you are reviewing.
Is daily rank tracking the same as starting a new Agent run?
No. Ranking observations build the history used to understand search presence, while an Agent run prepares a set of proposals for you to review. Seeing a new data point in Analytics therefore does not mean a new metadata draft was created or published. Use Generate suggestions when you intend to request another proposal pass and review its market scope. This distinction also helps when planning credits and team workload: collecting observations and deciding to produce a new set of recommendations are different activities.
How should I review results when markets finish at different times?
Read each market's visible status and proposals in its own context. A completed market can have useful material to review while another market is still preparing its results. Keep the app, locale and run context clear as you work, and do not assume an unfinished market has no opportunities simply because its cards are not available yet. Before publishing a group of locale drafts, inspect the final scope of the publishing action rather than assuming it matches whichever results happened to be visible first.
When is another proposal pass useful?
A new pass is useful when you have a specific reason to reassess the listing, such as a new market, a changed product offering or enough observation history to review an earlier update. Check the current Agent status and any measurement guidance before starting. A new run should lead to a reviewable decision, rather than becoming a reflex after every small ranking movement. The next set of proposals may support keeping the existing wording; request and evaluate the work without assuming that a fresh run must always produce something to publish.