Client case study · 3 months Illustrative

96→6

Hours of manual data entry per month, over 3 months.

Elapsed: 3 months
How it was measured: Hours logged by the two staff who were doing the work, before and after.
Running alongside: No other operational changes in the period.
Published with permission: Anonymised at client request

The constraint

Three systems, three answers, none of them right

Stock lived in the point of sale, on the website and across two marketplaces, each updated by hand.

Overselling happened most weeks and two staff spent their mornings reconciling.

Four systems with no shared source of truth
Pricing changes took a full day to propagate
Refunds and apologies for items already sold

What we found

The audit, including the unflattering parts

We logged the actual reconciliation work for a fortnight.

96 hours a month across two people
11 overselling incidents the previous quarter
27 hours average between a price change and the last channel updating

The plan

The 90-day roadmap as it was actually written

Pick a source of truth and push everywhere from it.

Point of sale becomes canonical for stock and price
Webhook-driven push to the website and both marketplaces
Hourly reconciliation to catch anything a webhook missed
Alerting when a channel disagrees with source

Month by month

What shipped, and what it moved

The line extends as you scroll. Every step below is a real month, including the one where nothing happened.

01Month 1

Baseline

96 hours a month of manual reconciliation and eleven overselling incidents the previous quarter.

Baseline: 96h/mo

02Month 2

Website and POS synced

The largest pair first. Reconciliation more than halved immediately.

-57%

03Month 3

Marketplaces added

Both marketplace channels on the same push. Overselling stopped.

-85%

What went wrong

The part most case studies leave out

The first build synced hourly, so an hour of overselling was still possible during a flash sale. We rebuilt on webhooks with the hourly job demoted to a safety net.

Batch frequency is fine until traffic is spiky, and then it is not.

Our first build synced on the hour, which during a flash sale still leaves a window where four channels can sell the same unit. We moved to webhooks and kept the hourly job only as reconciliation.

The numbers

Where it landed, and how we know

Hours of manual data entry per month: 96 → 6 over 3 months. Hours logged by the two staff who were doing the work, before and after.

Hours are as logged by the two staff doing the work. Nothing else about the operation changed.

Manual reconciliation: 96 to 6 hours a month
Overselling incidents: 11 in a quarter to zero
Price propagation: 27 hours to under a minute

Results reflect this client's market, licence status and starting position. We publish the method alongside the number so you can judge whether it transfers to yours.

In hindsight

What we'd do differently next time

Build on webhooks first; the hourly version was a wasted iteration
Add disagreement alerting in week one, not month three
Agree the source of truth before writing code — that conversation took longer than the build

Before you ask us to do this for you

Does this actually apply to your situation?

This worked because of specific conditions. If yours don't match, say so on the first call and we'll tell you what would change.

Market: Any retailer selling the same stock through more than two channels.
Starting point: A point of sale or inventory system with an API.
Timeline: Six to ten weeks depending on channel count.

Start with the audit. It's free and it's specific.

Send us a URL. You'll get back the technical issues, your AI-search readiness score, what competitors are doing that you aren't, and what we'd fix in the first ninety days.

Get a free audit → WhatsApp us instead