Redesign the Workflow Before You Add Another Tool

E-commerce teams rarely suffer from a shortage of software. The harder problem is that the work has accumulated around the tools: exports between systems, spreadsheets that reconcile other spreadsheets, Slack messages used as approvals and manual checks added after the last mistake.

Adding another tool to that environment can make one step faster while leaving the whole workflow just as fragile. Before buying, start with the operating outcome and work backwards.

Start with one outcome

“Improve operations” is too broad. Pick one outcome that can be observed without a new dashboard: reduce the time required to resolve a delivery exception, publish a product accurately, approve a promotion or understand why returning customers have dropped.

Then follow one real case from start to finish. Record every handoff, repeated decision, wait state and place where someone reconstructs context. The resulting map is usually more useful than a list of feature requirements because it shows where the work actually breaks.

Remove work first

For each step, ask four questions:

1. Does this decision still need to be made?
2. Does it need a person every time?
3. Is the information already available somewhere else?
4. What happens when the answer is wrong?

The first useful change is often deletion. A duplicated approval disappears. A report is replaced by an exception alert. Ownership moves from a shared inbox to one accountable role. These changes cost less than software and make any later automation easier to govern.

The first useful change is often deletion.

Design the exception path

Happy-path demonstrations make almost every tool look convincing. E-commerce operations are defined by exceptions: partial stock, failed payments, damaged parcels, conflicting promotions, unusual return behaviour and supplier data that arrives late or incomplete.

Before selecting a tool, define which exceptions it must detect, what it may do automatically and when it must stop for human judgment. If the team cannot agree on those boundaries, the workflow is not ready to automate.

Buy against the redesign

Only then turn the workflow into requirements. Test vendors against real cases, including the awkward ones. Measure the full cycle rather than the speed of the automated step. A tool that saves ten minutes but creates a new review queue has not improved the operation.

Happy path

Demo-friendly flow that makes almost every tool look convincing.

Exception path

Partial stock, failed payments, damaged parcels, conflicting promotions — where the operation is actually defined.

The goal is not a smaller software bill or a larger one. It is a workflow with fewer handoffs, clearer ownership and faster recovery when reality refuses to follow the plan.

Work the redesign with people who run the same workflows — then decide what to buy.