The Warehouse Task

Anybody can throw objects into a building. A warehouse business becomes successful because they can:

  • Intake objects into a facility efficiently
  • Find those objects again, and
  • Send the objects back out again to where they need to go.

While none of the tasks are particularly difficult, there is a considerable requirement for data accuracy. If something is lost in a warehouse, it’s very difficult to find it again.

The User

Warehouse work is work. A few do it because they love it, but the majority do it because it’s a job and they need one of those. The issue of user motivation is an important factor to consider when designing for this user population.

Design Methodology

Ease of use is important. Ease of making a good decision is critical. When designing for this population, you have to make it easy for them to do the right thing.

Interface Idea 1: Persistent Zoom Window

On the front end of the workflow there’s data entry. Newly arrived products must be categorized, serialized, assigned to projects/accounts, etc.

In the middle, there’s data entry. Objects must be assigned to pallets and storage locations. These values change as objects are moved, removed, and reorganized.

In the back end, there’s data entry. Shipping address must be entered. Shipping payment must be assigned to the appropriate account.

Workers are supposed to be checking this information for accuracy. Accuracy is very important.

Due to its data entry nature, warehouse UI is typically cluttered with data fields and drop down menus. This clutter results in small text. The font should, of course, be chosen carefully to increase character discrimination, but no matter which font you choose it’s always going to be more difficult to read when it’s smaller.

You could make the text larger for everything, but that has its own drawbacks. Much less information would be present on the screen at one time, there’d be more scrolling around, and that would slow down everything while reducing overall awareness of the product details.

The idea is to add a space to the UI which magnifies whatever information is in the active field. It would just be a constant magnifying glass that the users would have at their disposal. It’s important to note that this idea is not to simply increase the size of the active field (user testing could flesh this out, but having a field with a variable size would mean that the interface would jostle every time that a new field was selected and 1 field shrank and another field grew. Items would shift on the page, increasing the need for visual search, and making the job harder), the zoom window would be in a persistent location and the data in the active field would actually be presented in 2 locations at once.

Other design approaches to fix the problem could have the user clicking on something to bring up an enlarged review screen, but then they would also need to click again to back out. You could provide a zoom-in button, but then the user would need to zoom back out again before completing their work.

When dealing with a user with low motivation, the big concern is not that these click-required solutions would slow down the work. The main concern is that the user would choose to not utilize those options at all, and instead decide that “the smaller text is fine, I can read that” which will result in increased error rates.

Interface Idea 2: Color Coded Data-Entry Protection

As I’ve said before, workers need to perform tasks that require data entry, and accuracy is important.

However, workers also need to perform tasks that don’t require data entry, such as looking up inventory and tracking progress of an order. When completing these tasks, the chance remains that an errant keystroke will change some value in the system and cause unhappy results.

There are several ways to address this issue in the interface.

  • One is to limit write-access to only trusted employees, but it’s difficult to limit access efficiently when limiting access also limits an employee’s capability.
  • Another way is to add a save button and force the user to make a deliberate action in order to upload any changes. This is a good idea and shouldn’t be discounted, but after a couple of times of forgetting to save work (and therefore needing to redo work) there is the chance what the user will learn to habitually hit save from time to time.
  • Time-out user access. If the workstation is inactive for a period o f time, it loses the ability to make database changes. This creates a different interface problem where a user has timed-out, but not noticed it, then done some data-entry work before trying to save it, only to discover that all the work needs to be redone.

My idea would be to have two modes for the system: one mode where you could change data values, and one mode where you couldn’t. The way that the user would identify which mode they were in would be by the colored tint to the screen. Normal tint = no data entry. Colored tint = data entry possible.

  • This would allow users more relaxation when in a non-data-entry mode, because they know that no matter what they do or what accident happens, there’s no way that they can modify the data fields.
  • The colored mode could time-out, providing a visually apparent way of letting users know that they can’t make changes anymore.
  • Because the means of communicating system mode is so visually apparent, it would allow for self-policing within teams without the personal-space violation of reading over someone’s shoulder (people don’t typically react positively to people reading over their shoulder). Instead of looking closely at someone’s screen to determine what they are doing in the system, supervisors/peers could instead say “Frank, why is your screen yellow? I only asked you to look up the tracking number for that shipment. You shouldn’t be changing anything.”