Social customer service: The measurements and metrics to monitor

Social media may be a new set of channels over which customer service can be delivered, but that doesn't mean that some of the old rules don't apply. For instance, the old adage 'you can't manage what you don't measure' is still highly relevant.

In fact, with the jury still out on social customer support, you could even argue that nowhere is it more appropriate than in this field, as businesses are still feeling their way and desperately need visibility into the effectiveness of their efforts to be able to steer future investment. 

There are, of course, a great many key performance indicators (KPIs) that can be applied to your social customer care efforts.

For instance, in its popular Socially Devoted statistical reports into social support, Socialbakers measures the speed of the brand’s social response, and the quantity of the responses as well – although the caveat here is that obviously it’s easier to respond when the volume is lower, so this statistic favours smaller organisations. Nonetheless, this provides a useful point of comparison.

IBM’s social customer care consultant Guy Stephens advises that businesses don’t fall into the trap that traditional service measures have in the past, however. He says: “A lot has previously revolved around process efficiency – how quickly calls are answered – they’re not about the experience or resolving an issue.

Part of this is the result of the great production lines that came in at the start of last century. A lot of the metrics reflect that focus on processes, and this is an outdated way to look at something. You still need to understand the processes but now experiential metrics are more important. So now you should see a different type of metric, in addition to the process metrics, that put the resolution or the issue first. You still need to understand the processes but now you need more experiential type of metrics coming in, which are more important.”

Supporting this view is a Gleanster study which suggests that the most successful brands are going beyond simple response time on social, to focus on problem resolution time. As reported on Our Social Times, 94% of ‘top performing’ brands are monitoring – and actively looking to reduce – problem resolution time.

Jeremy Taylor explains: “Not only does this reduce costs, but a social media user that has their complaint quickly and efficiently resolved is likely to use the company again in the future and publicly recommend them to others.”

He suggests that as more organisations invest in social media monitoring for customer service, they alongside marketing and PR are reaping the benefits, with one being able to identify trends. “This means that you can identify and resolve recurring problems to ensure that the same thing doesn’t happen to other customers. Furthermore, social media can provide an early warning when there is a specific problem with a product or service,” he says.

In a presentation at Gartner's recent Customer 360 Summit, analyst Jenny Sussin outlined a host of social customer service metrics that she suggested brands should use to achieve greater visibility into the success of social customer care: 

• Time to first response     
• Average handling time (overall and by agent)
• Number and percent of serviceable posts (actual customer/consumer inquiries)
• Percent of inquiries managed by social channel (such as Facebook and Twitter)
• Cost savings on call deflection
• Changes in sentiment
• Changes in customer satisfaction
• Net promoter score (NPS)
• And flush rate (posts that are no longer relevant or serviceable).

Walter Van Norden from TELUS International suggests that broadly speaking there are three categories of social care metrics:

1. Service measures

“There are a number of different types of service measures – some are associated with demand such as listening volume, whereas others use the speed of response as an indicator of service level,” he explains.

“Service level is one of the more important indicators of success. An industry best practice is to measure service level every half-hour and report it as a weighted average over the entire day. However, companies should start by finding a feasible increment of time for measuring service metrics and move towards the ultimate goal of measuring in half-hour increments. In addition, contact centres should calculate which posts are not answered in the outlined time and calculate the percentage of posts that are abandoned.”

2. Quality measures

Measuring the quality of social care responses is a little more complex than calculating service measures and requires both a qualitative and quantitative evaluation of the responses, says Van Norden. “The key performance indicators that comprise quality measurements provide a more overarching examination of how social interactions are handled. In addition, quality measures include an evaluation of the impact of ‘channel redirection’ on metrics calculation.

“Channel redirection occurs when a customer asks a question on a channel that is not able to support an appropriate response. For example, on Twitter an agent can only respond in 140 characters, and may ask the customer to move the conversation to another channel in order to answer the question fully. Although channel redirection ensures that customers’ questions are answered, it raises problems for measuring performance and customer satisfaction.”

3. Effectiveness measures

Effectiveness measures provide contact centre managers with an evaluation of how the social care conversations affect overall brand perception. Depending upon the type of business the contact centre is engaged in, managers may wish to measure the posters’ sentiment, how likely they are to evangelise the brand, as well as the total reach for each post.

Van Norden says: “The clearest indicator of customer satisfaction is first post resolution (FPR), which is adapted from the traditional contact centre metric, first call resolution (FCR). Using a customer survey, FPR measures the percentage of posts that are answered on the first response. In addition to FPR, contact centres should also use this survey to measure the quality of an agent’s response, and should calculate the number of conversations that are redirected or transferred to ensure they capture the overall picture.”

Calculating ROI

But whilst it’s important for companies to measure service, quality and effectiveness, it’s also vital for leaders to be able to demonstrate the actual financial return of their social operations.

Kevin Bottoms, from TELUS International, explains that two of the most tangible gains are cost reduction and revenue generation.

Cost reduction activities

Examples of cost reduction activities include:

• Eliminating the number of calls to agents by facilitating the resolution of customer questions and problems via social care channels. How to calculate? Look for year-over-year declines in the volume of other channels. It’s important to survey customers who resolved their issue on social channels to determine how many would have contacted the service department via voice, chat or email if their issue wasn’t resolved in social channels.
• Reduce the amount of time customer service agents spend addressing inquiries by building an online platform where customers can search for answers to similar problems they are experiencing and solve their problems without interacting with the service department. How to calculate? Determining the cost savings of this indirect benefit is much harder to calculate than a direct transaction but can be estimated by looking at the reduction in utilisation of other customer service channels.
• Prevent returns through topical and immediate help.

Revenue generation

Examples of revenue generation activities include:

• Converting sales through direct transactions with sales agents. This type of interaction occurs when an agent directly intervenes while a customer is asking questions on a social venue about buying a product. How to calculate?Subsequent revenue generation can be measured by matching the agent interaction with the customer’s purchase.
• Converting sales through indirect transactions with a sales agent. These types of transactions happen as a result of the social community having visibility into direct sales transactions. When an agent helps one customer to convert a sale many other potential customers see that transaction and a percentage of those will act on that information. How to calculate? These transactions are more difficult to capture but can be measured with the help of post-sales surveys.
• Improving brand equity through proactive customer service shows customers that a company is willing to invest in their satisfaction. When a company is proactive in responding to complaints and negative comments, customers notice and brand equity increases. People are more likely to buy from a brand they trust. How to calculate? This type of gain is difficult to quantify but can be measured through overall increases in sales and improved customer satisfaction.