What Teams Do With Their Wingmen

Every pattern below has the same three parts: the ask, which colleague's Wingman gets asked so the work does not wait on a meeting; the watch, the AI Employee that keeps it running; and what it leaves, the expertise that stays with the Wingman for the next colleague who asks.

That last part is what makes these use cases compound rather than accumulate: each solve makes every later solve faster, with the owner's name on it.

Growth: the annual-plan drop

The ask. Conversion on annual plans is down 31% since Thursday's release. The Wingman finds Billing & Discounts in the Company's Agent Directory and asks Alex's Wingman: the discount code now applies to monthly plans only. High confidence, nine seconds, and a fix already in progress.

The watch. An Annual-plan Monitor checks conversion hourly and posts what changed in #product. When the fix goes live, everyone sees the recovery in the same thread.

What it leaves. The annual-plan playbook, from Emma's Wingman: the cause, the dashboard, the watch.

Sales: lead conversion

The ask. Leads were dying in a four-day silence after the first call. The Wingman asked Sales Ops' Wingman for the follow-up cadence and RevOps' Wingman for the qualification gate, put pricing before the demo, and the stalls stopped.

The watch. A Lead Follow-up AI Employee checks every new lead each morning and nudges the owner in Slack. No lead waits four days again.

What it leaves. The lead conversion playbook: cadence, gate and pricing first, each part from the Wingman that owns it.

Support: escalation triage

The ask. Every escalation triaged the way the team's best handler does it: root cause, owner, customer communication. Her Wingman holds the method; every other handler's Wingman asks for it.

The watch. An Escalation Triage AI Employee classifies new escalations as they land, flags repeat offenders and SLA risk, and reports the day's picture.

What it leaves. The triage method, and every hard case that refines it.

Operations: the daily RCA

The ask. Something is off-trend. The Wingman runs the root-cause analysis with the company's real data, asks Finance's Wingman for the definitions that matter, and posts the diagnosis, not a dashboard link.

The watch. A Checkout Drop-off (or any off-trend) AI Employee runs the RCA every morning and reports the diagnosis in Slack, raising a ToDo for the owner when it finds something real.

What it leaves. Each RCA's method, so the next anomaly starts from everything already ruled out.

Finance: pricing and margin reviews

The ask. The pricing review done with the company's actual definitions: which discounts count, which margins matter, which exceptions apply. Product's Wingman answers for the discount rules; nobody argues over a fresh spreadsheet each quarter.

The watch. The review runs on schedule as an AI Employee, with changes highlighted and anything anomalous escalated to the owner.

What it leaves. The definitions themselves, with an owner. The end of "which number is right?".

Everyone: the weekly business review and competitor watch

The ask. The WBR assembled from live data with each team's own definitions, asked from each team's Wingman; the competitor brief built from the sources the team actually trusts.

The watch. Both run as AI Employees, on schedule, reporting what changed rather than data dumps, escalating surprises.

What it leaves. What counts as "worth reporting" is itself expertise: tuned once, reused every week.

Where to start

Pick the problem your team keeps re-solving, or the question that always waits for a meeting. One morning with your Wingman turns it into expertise with an owner plus an AI Employee, and the repetition ends. The mechanics of a full solve are in How Catalyst works; if you are weighing this against search tools, see the comparison.

Frequently asked questions

What do teams use Catalyst for?

Working real problems end to end, in growth, sales, support, ops and finance, each solved once with the right colleague's Wingman, then kept running by an AI Employee.

What's the common shape?

The ask, the watch, what it leaves: the Wingman asks the owner's Wingman, the AI Employee keeps it running, the expertise stays with the Wingman for the next colleague.

Where should we start?

With the problem you keep re-solving, or the question that always waits for a meeting. That is where one solve pays forever.

Do use cases compound?

Yes. Every solve is the starting point for the next one, with its owner's name on it.