9 out of 10 customers will switch to the competition if they are not satisfied with their customer experience. This is why it is essential for companies to cherish their customers throughout their relationship. But how can you know if the customer relationship you envisaged is actually working? Measuring satisfaction can be even more complicated than you might think if your company simply relies on a customer relationship centre for sales prospecting or after-sales service.

This article describes five ways to measure contact quality and the level of customer satisfaction.

• Mystery shoppers
• Quality Monitoring (QM)
• Customer feedback
• Rate of claims or complaints
• Analyzing interaction content

1. Mystery shoppers

The mystery shopper conducts a satisfaction survey of your customer relations. They are not a real customer, but one who is generally sent by a specialized provider. This “customer” contacts your teams via the channel you wish to evaluate (e-mail, social networks, chat…), or even physically at a point-of-sale.

You will be able to find the information you are looking for based on a selection of criteria that you had established. The advantage is that you obtain a totally independent evaluation of your ‘customers’ experience.

However, there are a couple of downsides:

• The unit cost is relatively higher especially for store visits, which typically means that the number of evaluations performed may be low. As such, evaluations can be less relevant overall.
• It is not always possible to identify the brand’s remote sales representative or agent, which can sometimes lead to taking collective actions (more costly and harder to implement) when individual action would make more sense.
• The chosen scenario may not reflect the reality of your customer relationship. It is very challenging for the mystery shoppers to remain anonymous while carrying out their task, as their behavior is too different from your regular customers.

2. Quality Monitoring (QM)

Quality Monitoring (QM) is an evaluation based on the recordings of real conversations between a remote representative and the brand’s customer. This type of evaluation provides a realistic context for assessing customer relations. Of course, the criteria selection is based on:

• The type of interaction being considered, such as an information request, reservation, cancellation, problem resolution, etc.
• Respecting your processes, notably the overall expected behavior.

For example, evaluated criteria could focus on the use of tools made available to the representative, such as business applications, CRM systems, knowledge bases, as well as the representative’s knowledge on the topic. It is essential to identify these points in order to grow your representatives’ skills rapidly. Unlike with the mystery shopper, it is much easier to adapt individual action plans for each representative.

This method, however, may present a few limits:

• Our experience has shown that it is possible to evaluate an average of 3 to 5 contacts per representative per month, given the time it takes to perform the assessment, debrief, and set up a personalized action plan.
• A good evaluation of a Quality / Monitoring representative through Quality Monitoring will can show that they adhere to the customer relationship process intended by your company, before considering whether the customer experience itself was of good quality.
• The Quality Monitoring approach is often viewed negatively by representatives, once their skill level has reached the expected level of excellency. At this point, it is necessary to then pay particular attention to proximity management.

This type of evaluation is mostly performed internally at the contact centre. The metrics provided by Quality Monitoring are nonetheless interesting, and particularly relevant if you outsource your customer relationship.

3. Customer feedback

Customer feedback is an evaluation provided by customers themselves following their interaction with your company. To be effective, feedback must be captured as quickly as possible, ideally in a short extension of the contact with the representative and on the very same communication channel. For example, you could ask the customer to answer a satisfaction survey on your website after making an online purchase

Feedback must always:

• Be as brief and contextual as possible
• Be captured in a prolongation of the customer’s interaction with your company
• Ask relevant questions adapted to the type of contact

The notion of one contact per call requires questions to be asked at the end of that call. It is considered as the same call, without interruption, so that the entire context of the customer’s perception can be captured. 65% to 95% of customers are indeed influenced by the human interaction that occurs during contact.

The strength of this method is its high response rate: 85% of the customers we contacted agreed to provide feedback. To give you an idea, we generally receive feedback from 1,000 customers per year per agent, compared to 40 Quality Monitoring evaluations.

Nonetheless, two limitations should be considered:

• The customer must always feel or be able to ensure that their feedback is useful and that service will be improved thanks to their opinion. Otherwise, providing feedback only serves as a statistical purpose.
• Feedback can only work with a process control system, both to check whether the agent follows those processes correctly, and to detect oriented feedback requests (for example, an agent who only transfers those interactions with positive feedback).

4. Rate of claims or complaints

This metric used to be the only method for monitoring the quality of customer relationships. It is hard to say whether this is actually an assessment of customer satisfaction, as it is intrinsically based on their dissatisfaction. It is said that 1 unhappy customer shares their experience with 11 other people. According to Jeff Bezos, CEO of Amazon, social networks boost that number to 6,000 people.

It has therefore become a major issue to take immediate action for unhappy customers before your brand can be subject to bad buzz. This has even become one of the major tasks for many contact centres and customer service departments.

5. Analyzing contact content

Contact content analysis (or text/speech analytics) is an analysis of the content from a customer exchange performed by software. This method enables you both to define all types of received contacts and their reasons, and also to identify areas and criteria of satisfaction or dissatisfaction. 

With respect to other satisfaction metrics, this monitoring tool can analyze 100% of all interactions between remote agents and customers they communicate with. Contact content analysis can therefore measure your customers’ global satisfaction. The greatest strength of this approach is that it identifies all the company’s dysfunctional areas prior to the customer relations centres, notably those that generate unpleasant contact experiences that the company will have to work on to eliminate.

Still, it does have a few limits:

• You need at least 20,000 contacts to create a relevant model based on machine learning. Analysis lacks accuracy without that volume, only enable an occasional use rather than as a permanent option
• Human interaction is mostly evaluated on non-verbal elements, so the results of such analysis must be considered carefully. Generally speaking, the quality of these results is far less accurate than customer feedback.

Overall, these five techniques each have their own specific uses, advantages, and disadvantages. They are even complementary to each other! There is no reason not to use several of them simultaneously in order to maximize the benefits. The reality is that the more you know about your customers, the easier it is for you to meet their expectations, improve their satisfaction, and build their loyalty. It is a way to help you get into the virtuous circle of Customer Love.