Another DAM Blog

Blog about Digital Asset Management


1 Comment

How do I address user error?

Many people do not read instructions. They may enjoy reading what they want to read, but instructions are not one of those preferred works of non-fiction that come to mind. When was the last time you read instructions? Why? There is often the assumption and expectation things will be easy to understand and easy to use. When we download a new software application to one of our devices, do you see piles of instructions, guides, or manuals? Not likely.

And are we expecting other people to read instructions as well? Newsflash: many do not read instructions even if their lives depend on it. Some manufacturers in certain sectors include instructions to supply warnings, notices and legal documentation attempting to limit their liability in the case their product or service is (was) used incorrectly and disaster occurs. An example of this was a product mailed to homes as a free sample. This sample came in a small packet and included an image of a lemon on it. People instantly assumed it was lemonade mix, so they mixed it with water and drank it without ever reading what it said on the packet. The organization received countless calls and complaints from potential consumers who got sick. Most of them did not read the fact that this was dishwasher soap with a new lemon scent and an added citric acid cleaning agent. Reading is either clearly overrated for the masses or a means of separating the people who want to be informed from those who are too lazy/busy to bother.

Some manufacturers do not even include instructions anymore. Why? The product or service should be easy enough and users should just want to use it. They expect user adoption magically happen with some high hopes.  Maybe that works with some mobile devices and some of their respective apps through simple, smart design. Never mind the precautions, warnings or issues that could arise. What could possibly go wrong? Users are smart enough to just know how to use it, right? Well, if you add some humans to any equation, you will get some inconsistencies, variables, and yes… errors. Sure, we can blame the:

  • poorly designed user interface (usability testing can help identify these issues)
  • lack of forethought in the system implementation, so everyone must think like the person who created it (user testing can help identify these issues as long as they walk through all the processes and note what/where is a miss)
  • the whoops on the real world which avoids the end-to-end walk-through of a solution to be sure it works

One organization had a lot of user errors, so they started focusing on the tasks that caused the user errors and tracked them.  This could identify system flaws needing correction.  Here is how to start on the path of accountability. Every instance an error happened:

  • the user was tracked by name (who)
  • the type of error was tracked (what)
  • the frequency of the error was tracked (when)
  • where the issue was occurring in the system was tracked (where)
  • what the user did incorrectly to cause the error was tracked (why)
  • the recommended changes to the process were communicated and documented (how)

Every time an error happened, a template email was sent to that individual user and their supervisor which included:

  • their specific error
  • the impact of this error to other users and the system (there often was one)
  • a recommended fix (for the user to complete)
  • a set time frame to fix the error properly (one to two business days)
  • a follow-up to be sure it was completed and the error log to close out that particular instance of that error.

Before implementing this error correction, the policy was fully documented and shared openly. As soon as this process started, error rates dropped significantly.  That is the effect of accountability. Prior to this, accountability was not visible. Note that errors do not completely disappear because “perfection” is not a realistic goal for any organization. There is room for improvement for users, processes, and likely the system.

How to have MORE errors

There are the counterpoints to all this…

  • Assume too much or just assume everything will work the way you expect it to (just like the world will continue to revolve around you)
  • Ignore all issues you encounter. Do not verbally mention nor document in writing the issues for anyone to know about.
  • Do not test thoroughly or just ignore all testing completely  (The testing fairy is coming soon. Just don’t wake up from that dream)
  • Do not verify any information down the exact character. In fact, just do not check on anything at all
  • Do not follow specific instructions. Do not have clear, up to date instructions. In fact, do not have any instructions at all (see assumptions for similar results).
  • Do not explain how nor why something works nor even IF it actually works. People are just supposed to know this simply by osmosis or being born with this information.
  • Do not have a simple, easy to use GUI. If you really try, you could skip having a GUI completely. 
  • Ignore all usability experts and their literature. Why would you want anyone to actually use the system your company paid for?
  • Believe everyone works and thinks like you (revisit assumptions again)
  • Be sure to have extra slow processors to make people believe the system is frozen or non-functional. It might be acceptable in some people’s mind if a simple process with a few bits of data take a half-hour to two hours to yield the results requested.
  • Be sure to blame the end-user when the system is not working, but it is best if the results are inconsistent just for that added bonus.
  • Confusion is always welcome. With open arms.
  • Do not document anything. When working with other companies, trust everyone freely and believe that they will document everything for you, understand it all your way, and do not share this documentation openly.
  • Believe everything (including coding) is really easy and it will automagically be completed overnight flawlessly. Every day. With no documentation nor specifications. Nor testing.
  • Every IT department can read minds. They have an app for that.
  • Eventually, everyone can read your mind.
  • Trust everyone. What could possibly go wrong? You do not need any verification either.
  • Do not plan ahead.
  • Do not train users. Ok, maybe once and believe they will remember it all. 
  • Do not supply any ongoing support for your user community. They will figure it out.
  • Errors go away if you ignore them enough. Errors do not multiply when you do this. Errors are so much fun. Dream of getting more over time and it will happen in reality.
  • Never take any vacation nor breaks. It will not catch up with you in any way.

I only wish these were all so ridiculous that these did not ever happen nor were even thought of. Sadly, they actually do. Too often.

How do you address user error?