One of the common frustrations in IT support ā other than customers who donāt realise their hardware isnāt plugged into the wall ā is spending 45 minutes trying to figure out what the person is actually trying to do. This happens because, often, people will ask for help based on their attempted solution (āIāve tried X, and itās not working.ā) rather than the underlying problem (āY has gone wrong. Do you know what might be causing it?ā)
They call this the XY Problem. It was coined by American software developer Eric S Raymond, itās an easy trap to fall into. Both for users and IT professionals.
What is the XY Problem?
The best way to explain the XY problem is to offer an example. Imagine your car wonāt start. Instead of calling a mechanic and getting them to diagnose the actual problem, you assume itās a flat battery, so you call your friend and ask for a jump-start. Only it turns out the problem wasnāt the battery after all. Youāre back at square one, and now your friendās involved too! If anything, youāve just made things worse.
The XY problem is when a user thinks they have a good understanding of their problem (Y), and try to fix it by doing X. When X doesnāt work, they ask for help with X, not Y. See the issue?
Eric S Raymond put it this way:
āIf what you want is to do Y, you should ask that question without pre-supposing the use of a method that may not be appropriate. Questions of this form often indicate a person who is not merely ignorant about X, but confused about what problem Y they are solving and too fixated on the details of their particular situation.ā
How do I get around it?
The first step is not to assume anything. Remember, youāre trying to find the underlying problem, then a solution. Never the other way around. If a user comes to you and says āMy Y wonāt start, so Iāve tried X, and itās not working. Can you come and help me with X,ā thatās a red flag that something else is going on.
Author Laurence Enderson has a good quote thatās relevant here: āWe can be too quick to blurt out what we believe are the correct answers, when more value can be gained by searching for a better question.ā What he means is, you must go in with an open mind, and a curious mindset. Donāt take anything for granted.
So, whatās the one, important question you need to ask?
The magical IT support question
The magical IT support question isnāt āHave you tried turning it off and on again?ā (which, weāll admit, solves more tickets than youād expect). The question is: what are you trying to do?
Simple as that. What are you trying to do? Whatās your actual goal? By getting users to explain their fundamental objective, you can quickly trace things back to their roots and start again. Here are some follow-up questions you can add to help the process along:
- How did the problem first occur?
- What were you doing?
- What were the initial symptoms?
- What happened?
- What steps have you taken to solve the problem so far?
- What was the outcome?
- Can we reproduce the problem now?
These questions are all useful, but without the first one ā āWhat are you trying to do?ā ā they can all lead you down the XY path. You need that original objective because thatās the real problem youāre trying to solve.
Try and get people to be objective here, too. The only useful information is whatās actually happened, not what a user thinks might have happened. In other words, you need symptoms, not guesses, because symptoms donāt lie.
Itās also important to get the user to frame their problem in detail. The more detail the better. Donāt just settle for āThe printerās not showing, I canāt get it to work.ā A better IT call might go something like, āIām trying to connect this printer to the network, but itās not showing up on the list of suggested devices. Iāve refreshed the list and re-started the WIFI, but nothingās changed.ā
See the difference? The first call lays out a problem, but the second call gives the support worker context, detail, an actual objective, and a couple of solutions that have already been tried. By getting the user to frame the question correctly, youāve already saved yourself time ā and future headaches.
