Quick answer (TL;DR): Marketing-ops and RevOps requests arrive from everywhere: a Slack DM from a sales manager, an email from finance, a comment on a ticket, a new row in the SDR team’s shared sheet, an aside in pipeline review that nobody wrote down. Things get dropped because no single surface holds the whole picture. The reconcile is a periodic sweep across every channel where requests land, cross-checked against the team’s tracker, to surface what’s been missed. It’s tedious, mechanical cross-referencing, which people do badly and AI does well. That makes it one of the highest-leverage things an ops team can hand to an assistant.
This is the most immediately useful habit in The Ops Operating System, because nearly every ops team has the problem and few have a system for it.
The scattered-intake problem
Think about where requests reach an ops team in a typical week:
- A Slack DM: “can you check why this lead went to the wrong rep?”
- An email thread where finance asks, three paragraphs down, for a pipeline number that matches the bookings report.
- A comment on a ticket asking whether the webinar workflow can be switched off.
- A new row in the sales team’s shared sheet asking for a territory change.
- A request made out loud in the weekly pipeline meeting and never written down.
- “We should fix attribution at some point”, said by the CMO a month ago.
None of these surfaces holds the full list. So things slip, not through carelessness but through fragmentation. The request was real; it just never landed where the team would see it again. And the people sending requests often fire them across several channels and then forget, which leaves the ops team as the only party tracking the full set. No one can hold that many threads in their head.
What the reconcile is
The reconcile is a periodic, systematic sweep across every channel where requests can arrive, cross-checked against the team’s source of truth, to answer one question: what has fallen through the cracks?
It’s miserable work for a person: switching between five tools, scanning for open asks, holding them in memory, comparing against the tracker. For an AI assistant that can read across those surfaces, it’s routine. The assistant doesn’t get bored, doesn’t lose the thread and doesn’t forget the Slack message from three weeks ago.
How to run a reconcile
1. List your intake surfaces
Every place requests can reach the team:
- Slack or Teams channels and DMs
- Email, including shared ops inboxes
- The ticketing or project tool
- Comments and tasks inside HubSpot, Marketo or Salesforce
- Shared trackers and spreadsheets owned by sales, marketing or finance
You can only sweep what you’ve named, so make the list complete. (Check your capability audit for which of these the assistant can actually reach.)
2. Sweep each surface for open items
The assistant looks for:
- Open requests addressed to the team or a team member.
- Messages that contain an implicit ask (“the numbers look off again”).
- Questions awaiting a response.
- Items assigned to the team that nobody on it created.
3. Cross-check against the tracker
This is the core of it. Compare what the sweep found with the team’s source of truth:
- What’s here that isn’t tracked? Dropped or never-captured work.
- What’s tracked that’s already done? Stale items to close.
- What’s tracked as active but is really waiting on someone else? Items to mark as blocked, with the owner named.
4. Produce one consolidated list, with new items flagged
The output is a single view of everything open, with the new or dropped items flagged clearly apart from the ones the team already knew about. This is where the reconcile pays off.
5. Decide, then update
For each item: do it, schedule it, hand it off or close it. Then update the tracker and the shared memory so the next reconcile starts from a more accurate picture.
What a reconcile typically turns up
In practice, a good reconcile tends to surface a familiar mix:
- A real request that came in sideways, such as a routing change for a new segment asked for in a DM and never ticketed.
- A decision someone is waiting on: an unanswered “can we turn this workflow off?” on a ticket.
- A mislabel: work tracked under the wrong campaign, object or ticket number.
- Confirmation that several worries are unfounded, which is its own relief.
The last one matters more than people expect. Part of the value is catching dropped work; part is the confidence of knowing the team is on top of everything, rather than carrying a low-grade worry that something has been missed.
A diagnostic benefit: seeing the pattern
Run the reconcile a few times and a pattern usually appears: which stakeholders send requests across several channels, and which channels leak most. That’s useful management information for whoever runs the team. If one sales leader routinely asks through three tools and then escalates when nothing happens, that’s a case for a proper intake process, and the reconcile has given you the evidence. It doesn’t just catch dropped work; it shows why work gets dropped.
When to run it
Run a reconcile whenever incoming requests have outpaced the team’s certainty. For most ops teams that means weekly, plus an extra one after any chaotic stretch: quarter-end, a big launch, a CRM migration cutover, or someone returning from leave. The more fragmented your intake, the more often it’s worth doing.
The reconcile turns “I think we’re on top of everything” into “we know we are”. That shift is worth the few minutes it takes.
Frequently asked questions
What is a “reconcile” for an ops team? It’s a periodic sweep across every channel where requests arrive (Slack, email, tickets, CRM comments, shared sheets), cross-checked against the team’s tracker, to surface work that was dropped or never captured.
How does a marketing-ops team stop dropping requests? List every intake surface, sweep each for open items, compare against the tracker to find what’s untracked, then consolidate, decide and update. An AI assistant makes this fast because it’s mechanical cross-referencing.
How often should we run one? Weekly suits most teams, plus an extra reconcile after quarter-end, a launch, a migration cutover or a stretch of leave.
Doesn’t a proper intake form solve this? It helps, and the reconcile’s pattern data is often what justifies building one. But requests will still arrive sideways, so the sweep stays useful even with good intake.
This is Part 5 of The Ops Operating System. Next: the review gate that lets AI draft real ops work without anything touching production CRM data unchecked.
If the same requests keep resurfacing because the underlying systems are broken, our GTM Systems Audit finds the root causes in routing, sync and lifecycle logic.