Learn how to build a customer service Balanced Scorecard step-by-step. Start with a general business model, analyze the main business challenges, and convert them into a strategy map with KPIs.
Customer support should be fast, good, and inexpensive – a classical project management triangle, where one can pick any two options. With a right approach to the strategies and KPIs, an organization can balance on these three options, and have excellent customer service at a reasonable price.
Customer Service Balanced Scorecard vs. KPI Scorecard
In this example we create a Balanced/Strategy scorecard. As its name implies, this addresses the challenge of strategy execution. If you plan to use it for the customer service in your company, make sure that it is aligned with the overall strategy of your organization.
If your goal is to be more specific and address the operation issues of customer service, then you will find some examples of indicators for a KPI scorecard in the final section of this article.
Draw a Customer Service Model
As we do recommend in our KPI system: have a look at the business process, find its inputs/outputs, activities/results, and bottlenecks. If you decide to build a Balanced Scorecard for your customer service, then I’d recommend starting with a model of how it is working now.
Here is how it might work:
- A customer or a prospect client has a question about your product or service.
- A customer goes to the website and searches for the solution in the knowledge base, or
- A customer is contacting your company via a contact form, social networks, or phone call. First-line customer support specialist is trying to answer the question. If the problem is a tough one, then the issue is transferred to the second-line agent.
- A company is learning: the product is improved; the knowledge base is updated if needed.
What might be the possible goals/response plan for this model?
- Addressing Step 1: Make the product more user-friendly so that the customer would have not faced this problem, and respectively would have not contacted you, at least not with this issue;
- Addressing Step 2: Making the knowledge base easier to use so that customers could solve more problems themselves.
- Addressing Step 3: Train your first-line agents so that more of the difficult problems could be solved without transferring them to the 2nd line specialists.
- Addressing Step 4: Make sure that there are no information silos between customer service and other business units.
Let’s put these goals on the strategy map to reflect the cause-and-effect relationship between them. We can also add more details, and align them with the goals’ proper metrics.
We need to review the model described above once again and locate the points where the main costs are produced.
- Maintaining a knowledge base
- 1st line agent’s costs
- 2nd line agent’s costs
Goal: Reduce costs for the customer service
The highest costs are related to the agents. We can start with:
- Leading indicator: First-line resolution rate
- Lagging indicator: Cost per inbound contact
In this context “First-line resolution rate” is a leading indicator, but the same metrics can be a lagging one for a goal like “Improve agent training.”
Managing the Costs of Broken Appointments
Let’s discuss an example of how the costs of customer service can be managed. Broken appointments or FTA (Failure to Attend) is one of the typical examples. Clients forget about their compromise, and can affect the tangible costs. The costs might vary from insignificant, like in a case of no show for a online demo call, to important, like a dental practice, for example.
FTA KPI: (The number of not attended appointments * 100%) / (The total number of appointments)
- Contract costs (like cost of treatment in healthcare)
- Specialist time (time cost estimation that a specialist would dedicate to another task)
Calculate the FTA cost to understand if FTA is important for your business. This will also help to get a buy-in from the stakeholders.
- Develop response time. A simple recall system proved to be effective in most cases.
- Do client profiling. Analyze the profile of the clients where FTA is typical. For example, a study for the dental care domain (“The Reduction of Broken Appointments in General Dental Practice”) shows that the majority of patients that were failing to attend an appointment were unemployed.
In the Customer perspective we need to map what customers want a client service to be like. A typical mistake is to map here the behavior that the organization wants to see from its customers.
Another important nuance is that besides external customers of the organization, there also are internal ones! Your employees and your partners will also need to get in touch with your customer service. You might want to have specific goals and indicators for the internal customers as well.
Goal: Make knowledge base easy to use and up to date
What can be a possible improvement initiative for this goal? Many organizations have their knowledge base available for the public online, but not all of them invest in keeping those knowledge bases up to date. The action plan for this goal might be as simple as “Find and remove 404 pages,” or as detailed as “Do a behavior analysis of the website visitors and suggest better ways to organize and present data online.” For this example, I’m using an umbrella initiative: “Maintain knowledge base up-to-date.”
Here are the indicators that I suggest starting with:
- Leading indicator: % of the up-to-date materials in the knowledge base.
- Lagging indicator: “self-service completion rate” or its spin indicator “% of the “typical” questions asked of the customer service specialists.”
The lagging indicator in this case is very subjective. You might have an excellent knowledge base, but your clients will still prefer to ask “typical” questions in person. Nothing bad about this; to qualify the value it makes sense to compare it with the historical data.
Measurable by Design
How can you find out the value of “% of the materials that are up-to-date”? I’d suggest following here a “measurable by design” principle like we discussed in the KPI system. Instead of trying to measure something afterwards, be proactive and incorporate the measurement into your business processes. For example, when some feature of the product are, getting updated, make sure that the respective section of the knowledge base is marked as “Update is needed.”
Goal: Quick and effective conversation between a client and the company
In simple words, this goal means that a customer is asking a question and gets a detailed answer within a reasonable time. A typical action plan in this case is to train your customer service team.
- Leading indicator: First contact resolution rate
- Leading indicator: Average speed of answer (for each media)
- Lagging indicator: Satisfaction by the customer service
Net Promoter Score Indicator
What about another popular indicator: Net Promoter Score? I saw some customer service scorecards where this indicator is used together with “Satisfaction by the customer service,” but the problem is that NPS is strongly influenced by the company’s product/service, so it might be hard to distinguish the specific contribution of the customer service. The same might be said about “Satisfaction by the customer service,” but the degree of influence is less than in the case of NPS.
As you can see on the strategy map, the goals from the “Customer” perspective are linked by cause-and-effect connections with the goal in the Financial perspective. Active usage of the knowledge base and effective answers to the client’s questions will lead to lower costs for the company.
One might argue that this model pushes customer service specialists towards giving fast, but less useful answers. That’s not true. First, we have “Customer satisfaction” indicator to make sure that we are leading customer service in the right way; second, there is an implied goal in the “Finance” perspective – “Increase organization’s profits,” so whatever we do, it should help a company to earn more in long term, not just to cut short-term costs.
Internal Business Processes Perspective
Let’s have a look at the costs structure again. One of the biggest costs that we want to address are agent costs, which actually consists of:
- Hiring costs
- New agent training costs
- Regular training costs
- Managing costs
- Agent’s salary, bonuses, etc.
In other words, the company invests a lot in preparing new agents, and it is important to keep the turnover rate of high performers low. A business goal in this case might be formulated as:
Goal: Maintain Excellent Conditions for Agents
What might be an organization’s initiatives to achieve this objective? A common sense approach is to give your agents excellent training/coaching, necessary tools, and make sure that the work load is appropriate:
- Leading indicator: Agent training hours
- Leading indicator: Agent occupancy
- Lagging indicator: Agent satisfaction
- Lagging indicator: Turnover rate among top performers
You can go more into the details by adding more sub-goals and indicators related to the training efficiency and effectiveness.
With “Agent occupancy” we have an interesting case:
- On the one hand high “Agent occupancy” will lead to lower costs,
- On the other hand, high “Agent occupancy” means high work load and as a result a high turnover rate.
Where is the golden mean? Do controlled experiments to answer this question! Here are the hypotheses that you might use:
- Hypothesis 1: “Happy agents produce better results.” Keep agent occupation low and make sure that an agent has enough time to answer a client’s questions in detail.”
- Hypothesis 2: “High agent load doesn’t affect their results that much.”
- Hypothesis 3: “Improve knowledge base and train agents better to decrease their workload naturally.”
Your metrics will help to qualify the results obtained. According to some reports you can expect a dramatic growth of “Turnover rate” when “Agent occupancy” hits 70%.
Learning and Growth Perspective
A question for this perspective is: where should an organization focus their learning efforts in order to serve clients better and achieve goals from the financial perspective.
Here you might add the goals related to understanding clients’ needs better; the goals that will help to improve agent training and coaching programs; and improve the way the customer service team is communicating with other business units.
Goal: Learn from mistakes
If the problem was reported by a customer, was it reported to the developers? If there is some typical issue that customers struggle with, was a knowledge base revised?
Goal: Keep knowledge aligned with new customer behavior habits.
Is the knowledge base mobile-friendly? Do we have a version for print, an online version? Do we have guides in a video format?
Goal: Provide agents with better training and coaching programs
Do we provide enough tools and technical knowledge to the agents? Do we help them to improve their communication skills?
Goal: Break information silos
What needs to be improved in the communication between customer service, IT, Developers, Marketing and Sales? Find and eliminate possible information silos.
Indicators for a KPI Scorecard
The approach that we have discussed is appropriate for the definition and execution of a customer service strategy. What about more specific operational metrics? Here are some suggestions:
- Average Answer Time
- Average Resolution Time
- Tickets backlog
- First-call resolution
- Abandonment Rate
- Calls handled
- % of calls transferred
- First-line resolution rate
- Satisfaction by the customer service
- % of customers given satisfaction surveys
- Number of complaints
- Self-service completion rate
- Knowledge base completeness, %
- Cost per inbound contact
- Agent costs (hire, training, managing)
- Agent training hours
- Agent occupancy
- Average number of requests per agent
- Agent satisfaction
- Turnover rate among top performers agents
The list goes ahead. My recommendation is that there is no need to track all these metrics, the best approach is to understand the current business goals, and find metrics that will help to track progress towards these goals.
I guess many of the readers would like to create a similar scorecard for their own organization or business unit. Here is a 1-hour long video where I go through the stages once again and explain my steps:
6 Examples Where Plain and Clear Customer Service Beats Complex Service
Following up with the ideas about complexity metrics discussed in the previous article, let’s now focus on some examples of complexity applied to the customer service domain.
Complexity is all about perception, so before we start, here is how my perception of customer service evolved. For a long time, I was skeptical about the importance of customer service excellence: Finally, we buy a product and its qualities, but service is not among the main factors. After having some negative and positive experiences with customer service, I have to admit:
- The importance of customer service is crucial to retaining clients.
Let’s analyze some cases from the viewpoint of complexity. Our goal is to understand how, in those cases, the complexity can be:
Coffee Machine vs. Laptop
One month, I had to deal with two repairs services covered under warranty – one for a Nespresso coffee machine and another for my laptop. The laptop experience was “OK,” but the Nespresso experience was excellent. What was the difference?
- The coffee machine was picked up by a transportation company (that time I was living in Barcelona) that brought a special hard case for that very model. I’ve also got a replacement for the time of repair.
- The laptop experience started with packing a laptop in several layers of protective paper (my responsibility to ensure that it will safely get to the repair shop) and finding some old laptop as a temporary replacement.
How can the complexity in the laptop example be detected? Well, it’s easy – there was an instruction text which mentioned several times the importance of packing the laptop well before sending. I could deduce that many people failed to pack their laptops well and that resulted in that long instruction. When you have to write a two-page text just to explain how to send the product – it feels like a complexity issue.
Measurement of Complexity
How can we quantify and measure the impact of this complexity? How about:
- Time a client needs to pack the laptop well (including buying the materials)?
This complexity for laptop repair could be managed in the way Nespresso did it for their coffee machines, or they could simply allow the customer to get the laptop directly to the repair shop.
Car Renting Case
Another example that can demonstrate a different approach to managing complexities is car renting in the Canarias. The best practices in this example come from local company, Cicar. The negative experience comes from one of the world’s known car renting brands.
- At Cicar, you don’t pay upfront, and they have all reasonable insurances included in the price (so you don’t need to note all minor scratches in the car renting contract before signing it) they allow you to travel between islands; they charge you a deposit for the fuel and refund immediately when you return the car. If you return with more fuel than you had when you picked up the car – they will compensate it for you.
- The “big brand” renting company did not bother about eliminating complexities for the customer. The insurance by default is with a significant franchise, so most clients prefer to mention all the scratches in the contract to avoid problems when returning. You have to return the car with the same level of fuel, and they needed four months and a number of calls to customer service to get money back to the card.
How did Cicar learn to be so good? I guess this is a great “wearing customer shoes” example. They simply eliminated things that normally make you crazy when you have to rent a car, plus, their focus is on the Canary Islands so they can use 100% of the specifics of the market.
Measurement of Complexity
How can we detect and quantify similar situations?
- What are the most frequent problems mentioned in the testimonials and asked to customer support?
How about the complexity on the side of the renting company? We can also measure it using this effectiveness metric:
- Time from greeting the client to giving him/her the keys
Besides recording negative feedback, one needs to develop a response plan to detect and eliminate similar situations in the future.
Look at First-Time Customers
It’s important to focus your attention on those who are clients of your business for the first time.
Dentist Practice and a Doorbell
One dentist practice was located in an elegant building with high tech equipment and professional personnel. The only visible issue that they had was the doorbell – it was located three meters from the entry door. So, any first-time client was somewhat nervous when he/she saw the closed door, but no doorbell.
Those who came for the second time already knew where to find the doorbell, and it was not an issue for them.
- Look at those who have just started to use your product or service, and they will help you to detect “doorbell” issues.
Car Renting from a Big Brand
A similar problem happened to me like the car renting case described above. It was clear from my confirmation email that the office of the renting company was supposed to be in the Arrivals zone of the airport, but it wasn’t there. Neither was a sign or a shuttle bus to get to their office. Not an issue, if you know that you have to do this, but it was a kind of negative experience for the first time.
How can you detect complexities that first-time customers face? The most obvious idea works well – observe the behaviour of new customers, try to talk to them and get fresh impressions. In the 10 Step KPI system book, we compared observation and direct interaction, so don’t trust just one method.
Measurement of Complexity
How can you quantify complexities for first-time customers? Most likely, you will start with similar metrics as for normal customers, so the first step is to copy and paste the standard metrics (in BSC Designer, you can copy the whole scorecard if needed) but apply them specifically to the new customers.
- What is a good metric for a doorbell case? We could come up with some, but I´d say that this is the case when it’s easier to fix the problem than trying to understand if/how it impacts the business.
- In the case of car renting, the simplest quantification would be the number of negative feedbacks. Should there be 10 negative feedbacks in public, consider that there are x10 times more unseen.
Metrics That Matter for Customer Service. Hosting Provider Case
Let’s discuss another case that is a good illustration of how potential customers quantify (intuitively) service quality. I experienced this when we had to choose a hosting provider for a dedicated server. The market offered a number of options that satisfied budget and technical requirements, so the choice was not about price/features.
Filtering out marketing testimonials that companies usually put on a website, we noticed that one company appeared many times on our radar. It was Future Hosting. They were really good in customer service. We learned that from several forum posts by their current/old customers, then we tested it ourselves with some quick questions and finally, we confirmed this during a test period.
Measurement of Complexity
What metrics are applicable in this case? Let´s look at the situation from the customers perspective:
- Time to answer a support query. A benchmark depends on the nature of your business. I mentioned Future Hosting responded quickly (“We’ve got your message, and we are looking at your problem,” and it was not an auto response) and get most problems solved within a few hours.
- Query resolution success rate. e.g., how often do you have to tell your clients that you cannot help with this because it is out of the scope of your business?
Fighting Complexity – BSC Designer Software Example
A quick example of complexity fighting at BSC Designer. There are three typical cases in how prospective clients become a paying user of BSC Designer:
- They buy right away without even trying (we provide a free 30-day trial, but someone could buy directly because they used the software before working for another company).
- They sign up for a trial version and buy within a 30-day trial period, or
- Their trial period expires (as there is a need to get approval to buy new software), and they buy afterwards.
We were looking at how customers were getting through the registration process, and we found a complexity – different users had different ideas about what to do with the registration code that they received.
Measurement of Complexity
The metric that we used to detect this complexity was the number of customer support tickets related to this issue.
The solution was simple – instead of providing new customers with long instructions, we programmed a script on our side that detects the situation of the specific customer and does one of two things:
- Enters the subscription code automatically into the trial account of the customer, reactivates an expired account if needed and informs the customer that everything is ready now, or
- Gives instructions to a customer (that are specific for that very case and translated to the customer’s language) about how to use the subscription code.
The mentioned script was nothing special from a technical point of view, but our metric (the number of customer service queries of a specific type) showed that this was no longer an issue.
Complexity – Even Small Improvements Matter
From the previous cases, it might look like complexity is something difficult to manage. In many cases it is; it’s hard to organize a pickup of a laptop for repair while taking into account the variety of models and sizes; it’s hard to create a clear offer of car renting when there are insurance and bad driving involved. However, there are always small improvements that could make a big impact on how complexity is perceived.
Complexity That Can Be Managed in 1 hour – Parking Reservation Case
Here is another case – a parking reservation at an airport. The whole process was relatively painless, except the final step – the confirmation email. The email said that it is recommended to print it “just in case.” The issue was that the email was four pages long.
The need to print, and moreover, to print four pages is enough complexity for me (see the thoughts about the definition of complexity in the previous article).
Measurement of Complexity
Pick the one you like more: time spent on printing, the number of trees killed, CO2 emission.
The solution? Remove unnecessary rubbish from the reservation (or put it on the second page so that I could print just the first one)! Changing the email template can be done in less than one hour.
Do you want to create an even better experience? Then get the reservation delivered on a smartphone in the pass format. It’s not a big deal from a technical point of view, but it will make a tangible impact on how the whole service is perceived.
8 Steps to Manage the Complexity of Customer Service
To summarize the thoughts of this article, let’s focus on practical steps one can follow in order to detect unnecessary complexities of customer service:
- Create a strategy map. It’s not about complexity – it’s about understanding where your business is going to, the expectations of the customer and how you are going to satisfy those expectations. If you are new to strategy maps, use this free Strategy Map Wizard to get started.
- Find contact points with your clients. If you don’t have enough (one is not enough) – make sure to find more; make sure you are listening to those sources that you already have.
- Analyze the questions people ask. Sort the questions by the occurrence and impact. Looking at the most frequently asked questions is a good idea.
- Pay special attention to new clients. The challenges that they face will influence whether they come back to your company or not.
- Spot the small improvement points. Fighting complexity is not always about big changes; ongoing, small improvements are perceived by customers much better than big and long-term update plans.
- Find the metrics to track the complexity. You can say that you understand a certain aspect of complexity when you know how to measure it. Users of BSC Designer software can align those metrics directly with their business goals and use different types of metrics – the leading and the lagging ones.
- Develop an action plan and execute it. Users of BSC Designer can put those ideas into the “Initiatives” for each goal or indicator.
- Track complexity metrics. They are supposed to show you that you are on the right track with managing complexity. Repeat steps when needed.
A Final Word. Think About Behavior Induced.
I heard a story about a Scandinavian company that wanted to improve their call center. They implemented a number of KPIs and one of them was “Time to handle a call.” They pushed too hard, and soon they found out that the agents started calling each other to game this metric. This indicator was always in the green zone…
The main recommendation in this context is to think about the behavior that your new indicators will induce and if your current measurement culture can support these changes.
Do you have any thoughts? What are your favorite metrics for customer service? Feel free to share.