What a Month of Running the Account Actually Looks Like

Near the end of most scoping conversations somebody asks what the ongoing commitment looks like. The question sounds like it wants a number of hours. It almost never does. What the person is actually trying to picture is a month on their own calendar: which days have something on them, whose desk it lands on, and whether the whole thing collapses the week the season starts.

Fair thing to want before signing anything, and answerable — the work has a shape, and the shape is more useful than an hours figure that would vary by operator anyway.

The short version: the month has three layers running at different speeds. A daily floor of answering, which cannot be batched or deferred and is the only part that genuinely breaks things if it slips. A weekly block of publishing, which can be moved but not skipped for long. And a short monthly review: reading what happened and deciding one thing. Most of the load that surprises people sits outside all three — in approvals and in supplying the raw material, and both fall on the client.

Three speeds, not one workload

Treating channel work as a single ongoing task is what makes it feel unmanageable. It behaves better once you separate it by how often each part comes round, because the three layers have different owners and different consequences for slipping.

The asymmetry in that right-hand column is the whole planning problem. The layer with immediate consequences is the smallest one, and the layer with no visible consequence at all is the one that determines whether the other two were worth doing.

The daily floor

Every day, in Chinese, within a window you publicly keep, the enquiries have to be answered. No other part of the month has slack in it, and none gets underestimated as reliably, because the volume lands in clumps instead of spreading out.

Two things make it survivable. The window is a promise you choose — a stated reply time you can actually hold beats a faster one you cannot, which we worked through in staffing enquiries across a large time gap. And most of what arrives is repetitive: the same dozen questions — pickup, weather, children, luggage, a flight that lands late. Once those have good written answers, the daily floor becomes short. Before they exist, every message is a small piece of drafting.

The honest version of this layer, then: heavy for the first several weeks while the answer bank gets written, then a genuinely small daily task, punctuated by the occasional enquiry that needs a real person to think.

The weekly block

Publishing wants a block, not a trickle. Fifteen minutes a day scattered through the week produces less than one deliberate session, because most of the cost is context-switching into the work, not the work itself.

What goes into the block is ordinary: one or two pieces out, replies cleared under the last ones, next week's material queued. The part that fails is never the writing. It is supply — somebody has to keep producing raw material from the actual operation. That means filming and photographing while you work instead of booking a shoot day, and getting a year of posts out of one trip's material.

When a channel stalls, the stall is almost always here, and it almost always looks like a content problem when the shortage is upstream.

The monthly review

Once a month, somebody reads what happened and changes one thing. Not a report nobody acts on — a short look at where enquiries came from, which pieces are still being found, and what the questions people ask reveal about what the storefront failed to answer.

It can stay short because most of the dashboard is noise at small volume; the useful signal is narrow — the argument in which numbers mean anything at small volume. The discipline that matters is changing exactly one thing as a result. A month that produces a decision is doing its job even if the numbers were flat.

The part nobody budgets for

Here is where the month actually goes wrong, and neither item appears on anyone's plan.

Approvals. Every piece of Chinese copy that needs sign-off from somebody who cannot read it will cost a round trip, and rounds compound. One named approver with real authority turns a loop measured in days into one measured in hours. A committee turns a weekly cadence into a fortnightly one and takes the life out of the writing on the way.

Supplying the specifics. Prices change, a route closes, a vehicle is replaced, the season shifts. Nobody outside the business knows any of that until somebody inside mentions it, and a storefront that quietly keeps stating last year's facts is the failure mode described in what it costs to keep a Mini Program running — nothing breaks, it just stops being true.

Both of these are client-side by nature. No agency arrangement removes them; what a good one does is make them small and predictable instead of ad hoc.

Month one against month six

The shape changes, and expecting that helps, because otherwise it reads as things going badly.

Month one is front-loaded and mostly feels like homework: the answer bank does not exist yet, the first pieces take longer than they will, and every policy question surfaces for the first time. Enquiries are few, and that is normal; discovery has not accumulated yet.

Month six is the opposite. The answering is quick because the answers exist. Publishing is a habit. Older pieces are still being found, so the month's results are no longer tied to the month's output — which is also why a quiet month is not evidence of anything on its own.

The middle is the easiest place to give up: the work has become routine and the results have not compounded yet.

Where CN1X fits

The daily answering, the weekly publishing and the monthly review are ours to run — the operating half of what we do, alongside the build. The two things we cannot absorb are the two that decide the pace: one person who can approve Chinese copy without a committee, and somebody who tells us when something in the operation changes.

If you want the version of this shaped around your actual season and team instead of the general case, tell us how the year runs and we will map it against what the channel would need.