Omnichannel Manager, the other half: headquarters.
- Daily users
- 10,000
- Issue resolution
- 25% faster
- Spreadsheet work
- −40%
- Kind
- Retail order management, web
The same order, seen by everyone who has to do something about it.
Head office, customer service and the shop all need the same order and all want a different screen. This is the one place where an order, the stock behind it and the person waiting for it are in a single view, so three departments can argue in front of the same thing. About 10,000 people use it every day.
Three audiences who disagree about what a dashboard is.
Retail operations want throughput. Customer support wants one order, right now, with its whole history. Finance wants the numbers to reconcile. Build for the average of those three and you get something none of them opens twice.
The answer was role based views sitting on one consistent set of numbers. Different front doors, no competing definitions of what “fulfilled” means. The hard part was not the interface. It was getting three departments to agree on the arithmetic underneath it.
Four jobs, one set of numbers.
- One timeline
- Status, routing logic and fulfilment progress for any order, whichever channel it came from.
- Exceptions first
- Cancellations, returns, split shipments and stuck orders surface before a customer notices.
- Insight
- Throughput, fulfilment speed and issue trends, not a data dump inherited from a charting library.
- Roles
- Tailored views for operations, support and finance over the same underlying numbers.
The first screen anybody opens is the one listing what went wrong.
A bag arrives damaged. The last one on the shelf is not actually on the shelf. A payment does not authorise. The order goes on hold, and somebody at head office has to decide what happens to the person still waiting for it.
So the on hold queue is not an error state bolted on at the end. It is the screen the product is judged by, because it is the one a retailer opens first and works through before anything else gets attention. Designing it well is most of the job, and no dashboard screenshot has ever shown it.
“In a perfect world we would have zero on hold orders every day, but that’s not realistic because things happen.”
Five screens
Direction from me, the work from the team.
I designed the first version of this, the way I did with every NewStore product, and by this point there was a team to hand it to. I worked with the product managers and engineers on the workflow model, defined how analytics should be read before anyone built a chart, and the product designers took each area from there.
Consistency across the Associate App and this one was not an accident. Both sat on the same design system and the same vocabulary, so a number meant the same thing on the shop floor as it did at headquarters.