Delivery Systems Migration

Go-live contextual enquiry

TL:DR

A contextual enquiry on the go-live date for a software migration from an old to new system for deliveries (both driver and office systems). The team divided attention with some out on the road with drivers and some observing the operations office. We were able to report the migration was largely successful and were able to identify key issues for fast-follows that prevented longer-term problems and limited impact cost of the migration

How did we approach it?

The UX team spent the go-live day on site with the drivers and driver coordinators to see first-hand the day-one problems with the systems.

The dev. team left their sprint largely free to allow for quick turnaround of any potential service breaking issues or key fast-follows.

  • 3 UX graduates went out with a delivery driver each for a full 10-hour route to observe the impact of the new driver application.

  • Myself and the UX designer for this space spent these 10 hours shadowing the operations teams in their office.

What were we trying to understand?

As part of a larger migration to a new platform, one set of routes from one Ocado warehouse was selected to launch with the new delivery systems first.

We wanted to determine:

  • The overall usability of the new system in the field.

  • Any large usability issues with the new solution.

  • Any small teething issues that could be fixed within the current sprint.

  • What the cost to the business this first small scale roll-out had on the business and prevent these costs as the migration scaled.

What did we learn?

  • The new applications worked and got all user groups through a full shift with no fundamental issues arising.
    We were able to confirm with our remote dev. and product team that the systems worked on the ground as expected.

  • We identified a key missing element in the driver app around the ability to leave notes for other users for particular drops
    The designer was able to mock-up a proposal within the week and provide to the dev. team for the second week of the sprint.

  • A few fast-follow issues primarily around data logging and data surfacing between the drivers and office staff were identified.
    The devs were able to rectify these issues within the first week of their sprint.

Impact

  • The core impact of this work was it gave us targeted insight into necessary fast-follows to make this migration successful as soon as we launched. Despite the development and product teams being based remotely.

  • Issues around data surfacing caused a two-fold increase in support call time with the driver office, which we were able to return to normal levels within 2 days of launch.

  • Driver frustration with missing elements was mitigated within 7 working days of migration by delivering on this fast follow.

  • 3 of our graduates got their first experience of a contextual enquiry and also an awareness of how successful collaboration between user, UX, product and dev. looks.