Metrics Revolution: Customer Satisfaction Based

Traditionally metrics in customer care and engineering based teams have revolved around the performance of those teams internally. These were usually centered around looking at the micro level performance areas to measure velocity / throughput / output of certain tasks being performed. The problem with this approach is it does not focus on the mission or success of the company. Does the number of lines of code you wrote or the number of calls you were able to answer accurately paint a picture of the overall happiness of the customer base? Not really.

Evolve your metrics to look at all active customers in the system: are they able to use your service to meet their needs, is the service responsive and have accurate data?

Paradigm Shift

To start with lets do a review of what traditional reporting around care and engineering has looked like. When these methods were developed they did serve a purpose for measuring productivity and success. However as competition in your industry grows, and people are easily able to publish reviews online of their experiences with your service, the ability to maintain your customer base has evolved into needing to keep that customer base as happy as possible.

Various factors come into play when you are looking at this new approach that will influence how important these types of metrics are weighed:

• Ease of your customers to switch to your competitors

• How connected the user base is for your service

• Pace of change in your industry

If you are heavily contract based where there are large fees for anyone to attempt to switch services, your users are disconnected without an online community, or the industry you work in is not changing, than using these metrics may not matter as much to you. However if you have many competitors where it can be easy to switch, get a bad review for the world to see, or lose market traction because of lack of innovation: your future success may depend on these numbers.

Where to Focus

Lets dive deeper into what these new metrics will look like for your company so you can properly instrument. They are going to focus more around the actual customer experience as your customers are not paying for your internal efficiency, they are paying you to do something for them with your service. They expect that service to be easy, fast and accurate, or they would be doing this work themselves. You lose value as a service when they have to go out of their way to contact you to get problems solved.

Percentage of your customer base experiencing problems as measured by the backend errors being reported. The most effective area you can measure in this process is the number of actual customer errors, as only a portion of customers will actually reach out to you to get the problem addressed. The types of errors in this report should probably be paramount and on your "top 5" list for engineering / product to address as defects.

Abandon rates of transactions within your system. For the customers who had errors in the system, did they keep trying? Or did they immediately give up? This might show you how many customers are defecting. If these numbers are high you have a problem as your competitors are winning when this is happening.

Percentage of your customer base experiencing problems where they reach out for resolution. Knowing the proactivity and engagement rate of customers who have issues is key as it tells you how loyal they are to your service. If this number is high, then great! As this tells you customers want to work with you to solve the problem. However if this is low, it likely means your customers are going elsewhere.

For the customers who had errors, which methods did they use to contact your organization? This is an important metric for customer care as it tells you about reducing barriers to entry for your customers contacting you. Depending on the demographics of your customer base, they may need that phone call. Whereas others might hate being on the phone and only be comfortable with chat / email. Is limiting your choices for contacting support hurting your ability to effectively serve all of your customer base? Making your support phone number difficult to find or hiding it on your site may bring down your call center costs, but those unable to easily reach out may defect to your competitors.

Percentage of critical errors you were able to solve before customer contacts. This is the holy grail of automation and being agile when responding to production errors. This tells your engineering / operations teams how good their tooling and processes are at timely and effective responses to outages, data issues, etc...

Breakdown of customer contacts where there are defects vs blocker problems. This metrics helps you to know the average severity and therefore temperature of customers calling in. Are people contacting you about minor issues or are most calls about critical errors? If you have a higher percentage of customers unable to actually use your service you need to get these defects prioritized with development.

Average time it takes from initial customer contact to actual resolution of a problem. This is where legacy systems fall down the most. Traditionally measured at a person or team level for success, the effectiveness of the entire process is what matters most to the customer.

Deployment defect rate. High defects post deployment mean it might be time to do some review of quality process with development to ensure software being released is error free.

Correct the Process

Once you have completed your analysis of the above issues, it might be time to complete a lean value stream mapping exercise to understand where the inefficiencies lie in your process to be better able to serve your customers in the future.