I’m so excited to have you back! Initially this was all supposed to be one blog post, but as it turns out, KPIs have become so important to customer service that it warrants spending more time discussing them than I could fit into just one post.
If you’ve somehow found yourself here first, I recommend you start by reading Part 1.
Changed customer expectations & behavior = New KPIs
First, a quick recap of the basic premise: your KPIs are important because they usually play a large part in determining how your contact center is run. Using KPIs that aren’t the most important ones for the customer will ultimately create a worse customer experience - something that few businesses can afford in the current competitive landscape with customer loyalty decreasing and expectations increasing.
I’ve seen many contact centers revolve around their service level, and even though service level has its uses as a measure of performance, it generally doesn’t create any value for the customers in and of itself. If you want the entire discussion and my experience in working with service levels, get yourself over to Part 1.
Now, let’s continue by looking more closely at one of the really important metrics: FCR.
First Contact Resolution
Time is one of the most precious and important things for consumers and customers today. It follows that having their issues resolved quickly and with as little effort as possible creates direct value. And that’s First Contact Resolution in a nutshell. In fact, it’s one the most important drivers of customer satisfaction:
Think HDI - Metric of the Month: First Contact Resolution
A thing to note here is that an FCR rate represents one of the few win-wins in customer service. Both the customer and the company spend less time and resources with a high FCR rate. It simply takes longer and costs more to resolve an issue the more people are involved in it.
But there are more benefits than pure cost saving. Answering fewer conversations with a high FCR rate delivers a better customer and agent experience. Improving the agent experience is also important since we often have to deal with high employee churn compared to most other job functions, and this part of a better agent experience has a significantly positive impact on employee churn.
So how do you improve FCR? There are a few things that you can do, and most of them revolve around skilling up agents. Here are a few of the things that I’ve found most helpful:
- Tracking FCR for each agent creates a fantastic feedback mechanism. It makes it possible for the manager to give very specific and targeted feedback, or if your systems are set up for it, for the agents to collect it themselves. This takes a bit of time on an ongoing basis, but it’s usually time well spent.
- Give customers more information than they ask for. After handling several similar issues, you quickly learn what information the customers need and you can anticipate exactly what they need to know and then supply the necessary information in your first answer. In customer service, it’s generally difficult to over-inform customers and therefore much better to err on the side of giving too much information than too little.
- Invest in training. Sounds simple, but we all know it’s not - especially if you find yourself with very high employee churn. You don’t want to invest too heavily in employees that aren’t going to be around next year. In that case, I’d instead suggest two things: 1) invest in externalizing knowledge through Knowledge Centered Service (KCS) or a similar methodology and 2) read my post on the basics of how to manage a contact center. None of the two are quick fixes, but with some time invested, you and your customers will see results.
Tracking FCR can be tricky business, and most companies will only have the option of tracking it for emails. To track it across other channels, or even all your channels, requires a complex system and often a sizable investment. Before you even contemplate such a venture, you should consider whether or not you get substantially different types of contacts on other channels, or if email covers most of them. If the latter is the case, wait until IT systems catch up in the coming years and invest in cross-channel tracking later.
Average Handling Time
Average Handling Time (AHT) is best used as a way to make sure that you’re staffed to deliver the quality of service you want. AHT, like service levels, isn’t a KPI that delivers any value for the customer at all. An issue should take as long as necessary to resolve - and that’s the important part; that it gets resolved, preferably always the first time.
A lot of contact centers use AHT as part of measuring an agent's performance, often in conjunction with the number of conversations handled per hour or per day. What this does, is simply make the agent become conscious of spending less time with customers and it’ll invariably impact the customer service that he or she delivers. A good customer experience comes from the agent being able to be fully present for the customer and that doesn’t happen if agent finds her- or himself under pressure to have a short AHT and answer as many calls as possible.
Once again, this doesn’t mean that AHT doesn’t have any value to a contact center - it’s merely for internal use. Besides using it for workforce management optimization, it can be very beneficial to see what kind of contacts take longer than others. That information can often be used to change a product, service, process or just the communication to customers.
Transactional customer ratings
I distinguish between two types of ratings: transactional and relational. Transactional customer ratings are based on the customer’s rating of a single interaction that he or she has had with your company, while relational customer ratings try to capture how the customer feels about his or her relation with your brand.
Example of a transactional rating in a chat:
Transactional ratings have a great advantage over relational ratings since they provide instant feedback to the agent who handled the issue. When the agent gets positive feedback from a customer it feels good and greatly increases motivation. Getting negative feedback allows the agent to take a closer look at what happened. This makes it possible for the agent to handle a similar issue differently next time around and maybe get positive feedback instead. Agents are also able to reopen a case and find a better solution with the customer.
For the manager, transactional customer ratings make it possible to get an overview of the agents' performance as perceived by the customers. As with FCR, I consider this an excellent basis for the manager to provide feedback to each agent. Apart from that, the manager is also able to reopen cases that could have been handled differently and get back to the customer. This can turn a bad customer experience into a really good one.
As a side note, getting bad ratings overall for a type of contact can also help uncover problems with certain products or services. This is exactly the type of information you can bring with you to other parts of your organisation to instigate important changes to your business.
A note on Customer Effort Score
Another KPI that can be used transactionally is Customer Effort Score (CES), in which you basically ask customers to consider the statement “the company made it easy for me to handle my issue”. I won’t get into the details of how smart I actually think this is, except to say that in my experience it’s generally a good idea to track, and it works better the closer you can get to real-time and connect it with a transaction.
Remember, one of the best things you can do is respect your customers’ time.
Net Promoter Score
Net Promoter Score (NPS) is one way to very efficiently measure customer satisfaction with your company. It gives a holistic view of the way customers feel about your product, service and company as a whole by asking them to consider the statement “how likely is it that you would recommend our company/product/service to a friend or colleague?”. In that sense it’s a great measure for the entire customer experience.
What’s important about NPS and similar satisfaction metrics is that you act on the feedback you get. Not changing the way you do things renders the entire survey process moot and practically worthless (to your customers and therefore also to you).
I cannot stress this enough: all the time and energy you spend on things that don’t improve the customer experience is time that you want to spend differently. One of the pitfalls of surveying is not doing anything afterwards. Don’t fall into that deep, dark hole.
If you have ever visited a rating site, be it for a seller, a hotel, a shop or company, I'm pretty sure that you, just like I do, find the negative reviews and see what went wrong and how the company responded.
There are sites like Trustpilot, Yelp and the review section on Facebook, and they represent an opportunity for anyone serious about their customers’ experience. Know that your customers will find these sites and rate you whether you want them to or not.
A negative review is a great chance to show the public how you react when you make a mistake or give a customer a bad experience. And it’s an opportunity to rectify a bad situation with an individual customer. If you have to apologize - and chances are you should from time to time - don’t be shy about it, show that you take responsibility and then find a solution with the customer.
Try to think of review sites of a good way to get external and objective feedback for your company. Therefore, I recommend that you establish a presence to answer their questions or complaints and do it as soon as possible. A negative review on sites like these have a big impact and can spread like wildfire through social media, and that’s just one reason a customer review without any response doesn’t look good.
Getting positive reviews
Having a lot of reviews and a high average rating can make existing and especially new customers much more confident in your company and will benefit you in the long term. As most organic reviews tend to be negative, you’ll often have to do something actively to get positive ones as well. This is where it gets a wee tricky. You probably don’t want to invite everyone of your customers to rate or review your product (if you can pull that off and get exceptional reviews, let me just say this: you rock! Keep it up! Please share what you’ve done).
More than likely, you’ll want to identify customers with a positive experience. This does cost a few resources, either because you have to do it manually, or because you need to invest in a bit of development to get it done. I suggest two things: 1) Invite those of your core customers who have bought something that you know you generally handle well, and 2) empower your agents to invite customers to review your product manually.
At one of the previous companies I worked at, after a long while of doing this, we found that we received the most positive reviews on two specific days of the week. For us it was Mondays and Fridays. To this day, the reason for this trend still evades me. If anyone has any ideas as to why this could be so, please share them. It does, however, emphasize the importance of slicing your data as these are the kinds of insights you wouldn’t get intuitively.
Refocus your KPIs on the customer
In case it isn’t clear by now, customer satisfaction is extremely important to a business’ well-being, and a large part of the customer’s satisfaction is their experience when contacting the business. In a recent report, Forrester compared direct competitors in the UK and Germany with big differences in their customer experience, and not surprisingly found that companies that deliver a better customer experience retain more of their customers, sell more goods, and in general has a more positive word of mouth.
In order to get an edge, I really do suggest you look at the value that your KPIs create. To summarize:
KPIs that DO NOT measure customer satisfaction:
- Service Level (ie. 80% of calls answered within 30 seconds)
- Average Wait Time
- Average Handling Time
KPIs that do measure customer satisfaction:
- Answered calls
- Transactional customer ratings per conversation
- Review site reviews and scores
I advise all contact center managers to look at their current KPIs and consider what perceived customer satisfaction they measure. If you don't have KPIs that actually reflect customer satisfaction, you should probably reconsider your KPIs and change a few or even all of them.
It’s going from asking questions such as:
- How long do our customers wait in queue/for a reply?
- How long time do we spend per case?
- What’s the index on our service levels?
- How many customers had their problem solved immediately?
- How many customers are currently waiting for us?
- What do the customers think about product or process X and why?
If you, like many, have to get support for such a decision from your organization and/or upper management, I’d advise you to make the case for customer satisfaction and its business value and then tie in the new set of KPIs to that single metric.
Exactly how to go about it will differ from company to company, so I can’t really go into more details without sounding vague. But a nifty trick is to label it as ‘refocusing on the customer’ rather than ‘what we’re doing is wrong - we should do this instead’. If you do have specific questions, don’t hesitate to hit me up on Twitter or even just contact me directly.