Short-term problem solving can work. In the short-term...

It's Monday morning. You've been at your desk for an hour before anyone else arrives because you've got a lot to do this week.

Situation normal.

You're approached by a member of staff, with a problem. Actually, and more specifically, it's a client who has the problem. Because something that should have happened, didn't. You're annoyed because it means unnecessary interruption - unnecessary because it means someone somewhere didn't follow procedures. So, what to do - because the problem needs fixing. Before it becomes an issue.

Your first instinct might be to investigate the problem. Is it individual performance, or systemic shortcomings, for example. Whatever you do, it's tempting to focus on the problem first.

Not for me - there are three other questions to consider before then. Firstly, I'd be concerned about how important the issue was, and who will be affected by the outcome; secondly, I would decide how quickly a solution needs fixing, and who had the resource and capacity; and lastly, I would determine who should fix the problem, based on who had the capability and authority. Because there's a client waiting for a response - and you need to decide how quickly that should be coming; and because there are staff who, if treated and coached correctly and given the appropriate authority, can make the call. Quickly. You never know - sometimes the client might be impressed by quick responses. I frequently use to judge organisational capabilty by the way in which it would deal with problems.

Which is why, sometimes. a short-term fix can work. So, here are some thoughts on how...
If it's an everyday problem, I let my colleague fix it. I figure that it's a great delegating and learning tool, so whenever possible I will let them  get on with it (I have to be confident they have the smarts, and sufficient awareness to determine whether a short-term fix is the ultimate solution. I use Situational Leadership as a model to make that decision)

1. If it's fixed immediately, then good. Mostly - at this stage, though, I don't know whether it's an unusual incident, or symptomatic of something deeper. So I take note. I put some kind of recording system into place - you might find it works for you, too. It allows you (and your managers) to take a more considered view at a more suitable time, because...
2. ...you should want to take a cool look at the areas involved in the issue. It's where the complaints register can really help - although if that sounds too 'procedural' or your culture treats mistakes as an excuse for a witch-hunt, you should probably call it something different! Identify the areas and/or departments in which you're experiencing consistent complaints. Investigate these problem areas.
3. If you have delegated the problem for a quick fix by the individual or manager concerned, be sure you've given them the authority to proceed. And if they get it wrong (which can mean, they've fixed it differently, rather than wrongly - I found it's interpretative, often), train, guide, coach and mentor them. I once made the mistake of bawling someone out for fixing something differently. It took a while before they would make a decision again, and it cost me credibility and an awful lot of time!
4. Review. Because you now need to make a decision. Is this short-term fix enough, or is there something deeper. This is where I used to add value to my former businesses. Treating the symptoms rather than the cause would both prevent future problems, and cost less time. And money... So, I would then...
5. ...talk to the staff and department heads or managers to get an overall picture. Ask them about the problems that they most often experience with clients. Make a list and divide it between those things over which you have limited or no control (I call these external factors), like the quality of suppliers' products; and those which you have more  control, like IT or process issues (internal factors). Some of the businesses I've run got more analytical, using tools such as Six Sigma or other data flows - you might want to use a more simple tool such at Root Cause. Whatever you choose - and gut instinct has it's place - make sure you're not confusing cause and effect
6. Decide whether the data and responses show the matter to have been systemic, or individual. If the former, then there's a whole bunch of things you can do to put procedures (systems) right; if the latter, then review the training and supervision regimes. Think about whether there is a pattern to the individual error, and decide whether there's a skill or a will problem
7. Review procedures
8. Review training
9. Review controls and checks

Then, most importantly, move on. Too much time spent here will divert your time and attention from other problems. There always are other problems...