Use case
Monitoring a client’s category is work every agency agrees is valuable and almost none does consistently, because it only happens when someone has a spare hour. Multiply runs it as a scheduled workflow per client — daily, weekly or monthly — gathering signals against that client’s context and reporting back, so the scan happens whether or not anyone had time for it.
Today
The trigger for wanting category intelligence is almost never a fixed point in the calendar — it is mid-project, when someone realises a relevant report exists and they could send it to the client. That is exactly the kind of work that loses to the deadline in front of it. So it happens in bursts before a quarterly review, which is the opposite of monitoring.
How it runs
Describe what you want watched and how often. A workflow is a scheduled recurring agent run — daily, weekly or monthly — using the same agents and tools as any other task.
A workflow connects to a specific workspace or project, so it monitors against that client’s brand, category and history rather than returning a generic industry feed. This is the difference between a signal and a newsletter.
Output lands as a task on the project, with sources, so it can be read, forwarded or exported. Runs are autonomous — they do not sit waiting for someone to click approve — and if a run genuinely cannot proceed it asks a clarifying question instead of stalling silently.
Each run adds to what the workspace knows about that client’s category, so month six is better informed than month one rather than the same scan repeated.
A monitoring workflow that reports everything gets muted in a fortnight, and a muted workflow is worse than none because it still costs something to run. Write the instruction around what would actually change a decision — a competitor’s positioning moving, a regulatory shift, a category narrative turning — rather than around coverage volume.
For the metrics teams keep asking these reports to compute, the free calculators cover share of voice, engagement rate and the rest, and the glossary defines them.
Related: the product for how the pieces fit together, multi-client operations for running several of these at once, agency operations for the underlying idea, and the other use cases if monitoring is not the job you want to start with.
We will write the first workflow with you on the walkthrough, against a real client.
A listening tool is built to measure mentions and hand you a dashboard. This is built to answer a question you wrote, in the context of one client, and hand back something readable with sources. They are not mutually exclusive — several agencies run both, and feed the listening tool’s output in as material.
That is the normal setup: one workflow per client, connected to that client’s workspace. It was an agency request — monitoring world events per client — that shaped how workspace-level workflows work.
Weekly, for one client, for a month. Daily is usually noise unless the category genuinely moves daily, and a single client is enough to find out whether the instruction was written well before it is copied across the roster.
A single well-written weekly workflow tells you more about whether this fits than a demo of ten.