Subscribe to the blog and receive recommendations to boost your CX
It's no longer news that contact centers manage social networks. Facebook, Instagram, WhatsApp, Twitter, TikTok... New media demand that brands update their contact center management to satisfy consumers.
In fact, more than 70% of the projects we manage at inConcert incorporate social channels.
The pandemic and the increasing representation of Generation Z in markets have come to definitely transform customer communication with brands: more and more users and natives of social networks are spending more and more time on them. And that's where they prefer to meet their favorite brands, learn about their offers, interact with them, make their purchases and manage their services.
48% of consumers expect a response to their social media inquiries within the first 24 hoursStatista
42% of social media users use them to research productsData Reportal
In this article we want to guide Workforce analysts with tools to monitor in real time and measure the quality of service on social networks in the style of inConcert. A simple but infallible 4-step method for keeping your social channels under control.
4 steps to monitor social networks
1) Interaction flows
2) Key KPIs
3) Agent states
4) Reports and alerts
1) Interaction flows
The first thing an analyst must do is identify the different cases of interaction flows, that is to say, The step-by-step that every social media interaction takes. Knowing what happens from when a customer starts the conversation on social networks until the agent ends it allows us to detect possible bottlenecks and opportunities for improvement.
An example of a flow of interaction in networks:
In the flows of interaction on social networks, two particular situations stand out:
Using bots:
In some cases the interaction is fully resolved by an intelligent bot, without ever going through an agent. In others, the bot provides a first line of attention (as in the example) and then transfers the conversation to an agent. The Workforce analyst must segment all the time spent at each step of the interaction in order to correctly size the agents and ensure accurate monitoring. If the contact center platform doesn't segment care between bots and agents, it will be very difficult to control the indicators accurately.
Absence of abandonments:
The most relevant difference between social networks and calls or chats is that there is no possibility of abandonment, so interactions can create Backlog (accumulation of unresolved interactions) and affect indicators such as level of service, waiting times and customer satisfaction. The omnichannel solution we choose should, at a minimum, allow us to easily identify these interactions and the time they have been waiting for a resolution.
2) Key KPIs
The function of any analyst is to take care of the operation's indicators.
In order not to lose focus and focus on what is truly important, we recommend define KPIs that are directly associated with the type of operation and commitment to both internal and external customers.
What we should define about our KPIs:
The formula for measuring them, the target result, the ranges of non-compliance, who to alert when the indicator is off target, and the matrix of actions to follow in the event of a deviation.
As an example, we list some of our favorite KPIs for social media campaigns:
- Forecast accuracy: Difference between the actual volume of interactions and the forecast, expressed in%.
- TMO: Average operating time. This KPI can be custom built based on the data that our platform provides us. Generally, social media service eliminates classic contact center times such as Hold (wait) and the ACW (after-call work, post-call work), since the interaction does not imply an instant response to the customer, and the management work is done in parallel with the chat. We can then measure the TMO with the resulting time between the first customer interaction and the last of the agent, or the first response of the agent and the last interaction of the agent, or even account for automatic closures due to inactivity after a certain amount of time.
- Time between messages: Time that elapses between the customer's message and the agent's response. Ideally, your contact center platform should allow the measurement of these times, presenting a consolidated summary per contact, and detailed summary per message.
- Concurrence: The number of cases that an agent resolves simultaneously in a given time.
- Cases per hour: Number of interactions answered by an agent in an hour.
- First Contact Resolution or FCR (First Contact Resolution): This indicator refers to whether the customer's query or case was resolved in the first contact, and it is not necessary for the customer to write to us again to follow up.
- ASA (Average Response Time): It measures how long the person who contacts us must wait to be served by an agent (in other words, the average speed of response).
3) Agent states
Real-time analysis wouldn't be complete until the agents' connection was taken into account. From the agents' states, we can obtain multiple productivity reports.
Agent metrics are established according to the needs of each operation, although they are all based on the sum of time and the number of times agents spend in certain states. The two main metrics in trading are utilization and occupation.
Although the use of both is already quite widespread, they still tend to be confused.
La utilization is formulated to measure the time an agent spends at work or available for work, over the total (paid) time of their connection. This helps us to compare the use of productive vs. non-productive states.
La occupation focuses on how much of their available time the agent actually attends to interactions. Occupation does not take into account non-productive intervals such as rest times or training. For this reason, occupancy is more recommended as an indicator of productivity in digital channels, since it is sought that the agent is actually working most of his available time: that he always has assigned cases, that he has the Queues (work queues) correct, etc.
4) Reports and alerts
Finally, once the real-time analyst has defined the indicators and goals to manage the operation, he must agree with his team the format of the report to be delivered, its frequency, to whom it will be addressed and by what means.
For this, we recommend building a scaling matrix. This one details:
- the indicators
- the goal
- the alerts to be configured by result ranges
- the recommended actions for each case
- to whom to direct the information
Let's look at an example of a scaling matrix for the ASA (Average Response Time) indicator on a social network, for example Facebook.
INDICATOR: ASA - Average Response Time
TARGET: 120 seconds
The frequency of delivery of reports and alerts can be determined by the capacity of the contact center platform. That is, how often it provides data updates, either through integrated dashboards, automatic data downloads or email automation.
Following these four keys, The Workforce analyst will have tools to contribute to maintaining and improving the quality of service in the contact center, considering all channels.
👉 Discover all the InConcert capabilities specialized in Workforce
SEE MORE ABOUT WORKFORCE IN THE CONTACT CENTER:
