Marko Pavicic, CEO

January 5, 2023

Why is 'Mean Time to Resolution' (MTTR) the most essential metric in Customer Service?

With over a decade in SAAS businesses, and most of that time in services roles deploying sales and services processes for large enterprise customers, I have been witness to several customer service jumpstart strategies, many of which I'll talk about in this article.

To begin with; if we deconstruct Customer Service into its atomic parts - the only thing customer service needs to do is help the customer with their problems - nothing less, nothing more. And unless you are in a romantic movie where you're calling your soulmate that works in customer service your aim is to spend as little time as possible talking to an agent. This means that we've boiled down customer service to two things - the ability and speed to resolve. 

Granted that you can quickly and easily resolve customers' problems, there are very few other metrics or statistics that the customer really cares for. They will disregard only a few options to contact the company, they will disregard that there is no social network page, and they will even disregard if you're not the most approachable person with perfect pronunciation when they talk to you. Of course, all of these things matter, but imagine the opposite; You have all the means available to raise a problem, but it takes the company weeks to resolve. Customers are spending their most valuable resource 'time' by repeating themselves in a ping-pong match between various departments all while their frustration slowly bubbles. 

That becomes even worse if the request is simple in nature (low-value demand); Name change, user/email change, address change, order status update, etc. If you are a company where these kinds of requests take more than 60 seconds on your self-service portal or through an agent you either are, or should be allocating a budget for a new Technology/Process change. If these kinds of requests are taking a long time to resolve, what are more difficult or unique requests doing to your overall MTTR metric, bottom line or NPS score?

Yes, there are tools out there to help alleviate this problem, but before embarking on a journey of buying software/hardware, try to analyse what is holding you from excelling in customer service. Look at your data, map out your processes, ask your agents what is distracting them from doing their job, asses what your operation considers good or bad demand and analyse your costs - i.e. create a customer experience map that will form the base of your transformation endeavour. This is paramount as it will:

  1. Greatly speed up the technology/process implementation project when you start one
  2. Create a baseline metric that you can track over time and report upon
  3. Help you correct your trajectory when you see statistic fluctuation after you implement any changes

The inverse of that is that you will:

  1. Spend a lot more in Professional Services when implementing any technology/process (and prologue the project)
  2. Not be able to track what enhancements were made, what metrics changed and where should you focus next
  3. Not be able to fail fast - while continuing down the wrong path costing considerably more 

When looking at any process Dr. Eliyahu Goldratt in his 'Theory of Constraints' says that the most important thing to identify is the limiting factor, i.e. the bottleneck in the process. 

Now why is that important I'll try to explain through an example.

Let's imagine a process where your customer service is a pizza place, there are 4 people (stations/steps) needed to deliver the pizza to the customer (or 4 steps to solve a case).

  • 1 person taking orders
  • 1 person rolling the dough
  • 1 person putting the ingredients on the pizza and cooking it
  • 1 person getting the pizza to the table 

And for simplicity let's imagine that they all can do their 1 job in 1 minute. This means that the pizza place can make 1 pizza in 1 minute (let's disregard the first one which will take 4 min) - throughput is 1 pizza/min and utilisation is 100%. Great, this means that MTTR is 1 minute.

Of course, that is atypical and usually there is a job in the process that takes more time than the others. As an example let's say that rolling the dough takes 2 minutes - Now, this means that it takes 2 minutes to deliver a pizza (MTTR = 2min), throughput drops to 0.5/min but the most important thing is that you start taking orders you might not deliver & stations after the pizza rolling station are 50% utilised due to not having enough pizza doughs to work with.

We have now discovered our bottleneck and this is limiting us in reducing the MTTR - Improving any station except the bottleneck is just an illusion and will not reduce the MTTR. To elaborate why; If you hire 5 more people to take orders you will just have a lot of orders that you cannot deliver (and add stress to the dough roller person), and if you hire more people to put the ingredients or deliver pizzas you will just have more people twiddling their thumbs as they do not have any work while the MTTR stays the same.

One of the solutions for the pizza problem is a technology/automation one - an automated pizza dough roller machine that reduces the time to 20s. This now moves the bottleneck to some other station/step where you can improve efficiency in some other way. Improvement in its form is an iterative process where there is no end - how Sensei Mike Rother puts it: "almost does not matter what you improve, as long as you're improving something. Why? Because if you are not improving, entropy guarantees that you are actually getting worse".

This is why it's crucial to understand your process, map it out and discover your bottlenecks.

If we move away from the pizza analogy and back to the real world, I've seen bottlenecks in; technology, senior knowledgeable agents, approvers, managers, external systems, email inboxes, field technicians, salespeople, so the bottleneck can be literally anything - find yours and try to decrease your MTTR.

I've recently transferred to the world of ServiceNow working for Sneeyeg where I'm helping companies transform the way they serve their customers and motivate their employees. ServiceNow technology offers exactly that with automation and operational efficiencies in places where many thought it is not possible. I hope that this article starts the thinking process for your transformation, and if there is help needed in discovering bottlenecks, analysing processes, or implementing the ServiceNow technology, reach out.

Every good article leaves you with something to think about, so:

  1. Do you have all your processes documented?
  2. Have you identified your bottlenecks?
  3. Do you have a vision of where you want to be?
  4. Do you track everything you need to track (MTTR, FCR, NPS, CSAT, CES) and all the constituents that form them?
  5. Does your technology allow you to do it?

Get in Touch