Skip to content
SaaSQuillSaaSQuill

Live selling · Fulfillment

Live-Commerce Operations & Shipping System

Anonymized study of a real client system — no client names, no invented numbers.

The problem

A live-selling operation compressed hundreds of orders into short broadcast windows. Labels were generated one by one after each show, packing was error-prone, and reconciling sales against inventory took the rest of the night.

The approach

The constraint was pace: everything had to keep up with the stream. We designed the workflow around capturing sales the moment they happen and pushing each one straight through to a printable label and a packing queue.

What was designed and built

An operations tool that captures live sales as they occur, prints shipping labels in real time during the show, feeds a packing queue with per-order manifests, and reconciles sold items against inventory when the stream ends.

Technical approach

  • Node.js
  • WebSockets
  • Carrier/label APIs
  • Thermal-printer workflow
  • Inventory ledger

Before and after

Labels after the showlabels during it

Verified outcome

Label printing moved from after the show to during it. Packing runs from a queue instead of a handwritten list, and post-show reconciliation became a report instead of a project.

One real broadcast, captured live: 660 sales in just over four hours, recorded and labeled by this system while the show ran. Real aggregate data from software we operate — buyer, seller and revenue details deliberately omitted.
2700h00–0h05: 3 sales0h05–0h10: 13 sales0h10–0h15: 12 sales0h15–0h20: 11 sales0h20–0h25: 12 sales0h25–0h30: 15 sales0h30–0h35: 15 sales0h35–0h40: 14 sales0h40–0h45: 13 sales0h45–0h50: 18 sales0h50–0h55: 13 sales0h55–1h00: 7 sales1h00–1h05: 11 sales1h05–1h10: 13 sales1h10–1h15: 8 sales1h15–1h20: 9 sales1h20–1h25: 5 sales1h25–1h30: 7 sales1h30–1h35: 5 sales1h35–1h40: 7 sales1h40–1h45: 7 sales1h45–1h50: 5 sales1h50–1h55: 5 sales1h55–2h00: 15 sales2h00–2h05: 19 sales2h05–2h10: 16 sales2h10–2h15: 17 sales2h15–2h20: 6 sales2h20–2h25: 14 sales2h25–2h30: 26 sales2h30–2h35: 14 sales2h35–2h40: 27 sales2h40–2h45: 20 sales2h45–2h50: 20 sales2h50–2h55: 19 sales2h55–3h00: 21 sales3h00–3h05: 13 sales3h05–3h10: 14 sales3h10–3h15: 21 sales3h15–3h20: 19 sales3h20–3h25: 21 sales3h25–3h30: 9 sales3h30–3h35: 13 sales3h35–3h40: 3 sales3h40–3h45: 13 sales3h45–3h50: 15 sales3h50–3h55: 24 sales3h55–4h00: 27 sales4h00–4h05: 6 sales27 sales / 5 minshow start4h05 later
View as table
Sales captured per five-minute interval
IntervalSales captured
0h00–0h053
0h05–0h1013
0h10–0h1512
0h15–0h2011
0h20–0h2512
0h25–0h3015
0h30–0h3515
0h35–0h4014
0h40–0h4513
0h45–0h5018
0h50–0h5513
0h55–1h007
1h00–1h0511
1h05–1h1013
1h10–1h158
1h15–1h209
1h20–1h255
1h25–1h307
1h30–1h355
1h35–1h407
1h40–1h457
1h45–1h505
1h50–1h555
1h55–2h0015
2h00–2h0519
2h05–2h1016
2h10–2h1517
2h15–2h206
2h20–2h2514
2h25–2h3026
2h30–2h3514
2h35–2h4027
2h40–2h4520
2h45–2h5020
2h50–2h5519
2h55–3h0021
3h00–3h0513
3h05–3h1014
3h10–3h1521
3h15–3h2019
3h20–3h2521
3h25–3h309
3h30–3h3513
3h35–3h403
3h40–3h4513
3h45–3h5015
3h50–3h5524
3h55–4h0027
4h00–4h056

More studies

All case studies →

Facing the same kind of problem?

The system behind this study can be shaped to your operation. Start with a free workflow review — an engineer replies within one business day.

Get a Workflow Review

Replies within one business day · no mandatory calls · or email [email protected]