Chat Key Performance Indicators

https://www.klipfolio.com/resources/articles/what-is-a-key-performance-indicator#KPIeffective

1. Number of Chats-non QA form source

2. Agent Utilization Rate-non QA form source
3. How Long Visitors Are Waiting in the Queue-non QA form source
4. Average Handle Time-non QA form source
5. First Contact Resolution- QA Form
6. Invitation Acceptance Rate- non QA form source
7. Sales Conversion Rates- Non QA form source
8. Visitor Logs and Wrap-Up Notes-QA form
9. Customer Satisfaction Scores-QA form

I. Number of Chats

Here’s some good news: calculating the number of chats that your department has received is
pretty straight forward, and it can tell you a lot. Here are some things to look for when studying
this metric:

1. How many chats are agents accepting as opposed to rejecting or passing off to
other agents? Consider what the numbers reveal about agent work ethic and performance.
2. Is your live chat count lower or higher than what your company is aiming for? Ask
yourself what changes could be made to your live chat experience to make your numbers go up
or down. It is important to consider how visual elements such as the visibility and placement of
the live chat button may impact how many chats your company is receiving. You may also want
to consider any changes in marketing that might be needed to generate more traffic to your
website. If you want more customer engagement, consider switching to a proactive live chat
strategy.
3. What is your number of missed chats? This number can show you whether your
department is adequately staffed during peak busy times. If not, consider taking on more
agents, seasonal hires, or just shifting the schedule to have more agents available when your
team is busiest. A high number of missed chats may also indicate that you need to check your
average handle time (more on this metric later) to make sure that your agents aren’t spending
too much time on each chat.
4. What is your number of offline chats? If there are any times where you are receiving a lot
of offline messages or chat requests, consider adding to or shifting your operating hours to
accommodate this.
You can analyze this metric by viewing your chat volume report.

II. Agent Utilization Rate

The best measure of how work time is being used is agent utilization rate. This key
performance indicator reveals the percentage of time that agents are spending in live chats,
wrap-up, and other productive functions, as opposed to in “away” mode or offline.

Agent utilization rate can be measured as follows:

Amount of live chats per month x Average Handle Time / Hours worked in a month x 60
minutes
Typically, a 50% to 60% benchmark is good to aim for. A utilization rate that is too low might
be indicative of problems like overstaffing and poor agent training. Similarly, a utilization rate
that is too high may result in rushed chats, mistakes in documentation during wrap-up,
negative customer feedback, and agents who are stressed out.

Consider investigating the following when studying this metric:

1. What is your agent’s occupancy as opposed to their utilization? Occupancy is the

percentage of time an agent spends logged into the system and ready for or engaged in live
chat. Utilization is the percentage of occupancy that agents spend actually busy—either in a live
chat or doing post-chat wrap-up work. This differentiation will help you identify dips in live chat
volume and problems with over staffing.
2. Does your agents’ log-in time correspond with the number of hours they have
worked? By checking both your remote and in-house agents’ log-in times, you can ensure that
you have agents logged in and available when their schedules say that they’re supposed to be.
Noting any divergences can also be telling of agents’ work ethic (such as when an agent clocks
in but wait to log into the system) and their level of commitment to the team. By making sure
that agents are clocking in and logging in on time, you can also identify any potential need for a
schedule rearrangement (such as if an agent always logs in late because he or she doesn’t have
enough time to make it from a class to the office before his or her shift starts).
3. How much time are your agents spending in “away” mode? The percentage of time that
your agents are making themselves unavailable can speak for their work ethic, and can also
identify a lack of team morale.
4. How many chats are your agents handling at once? Check how many chats your agents
are handling simultaneously and if that amount is impacting their performance. This number
can reflect an agent’s experience, knowledge of company policy, and efficiency—but only so
long as the content of their chats is quality. Investigating this will help you identify any training
deficiencies among agents who are handling a low number of chats, and can also help you
identify problems with understaffing if agents are handling too many chats.
You can check out this metric in the agent workload report.

III. How Long Visitors Are Waiting in the Queue

Wait time has a huge impact on customer satisfaction. How long visitors are waiting in the
queue is an important metric that is available to managers in the wait time report. The wait
time report shows details on visitors’ average and longest wait times within a specific time
range. Another valuable report, the queue report, shows the number of visitors who waited in
the queue, abandoned the queue, switched to message, or were refused by agents within a
specific time range.

You can compare data from the wait time and queue reports to see how wait time affects
visitors’ actions when they are waiting in the queue. If the wait is long and abandonment is
high, this can be a sign that you need to take on more agents or increase the maximum chat
allowance per agent. If the wait time and/or abandonment rate is high during specific times of
the day, consider having more agents work during the busiest hours. If the queue time
fluctuates heavily with the seasons, consider taking on seasonal agents to help you tackle the
holiday website traffic.

IV. Average Handle Time

Average handle time, or AHT, is a classic measure for evaluating agent performance and refers
to how long each agent spends on a chat on average. By measuring your agents’ average
handle time, you can help enforce a speedy, concise resolution of customers’ issues.
The average handle time can be measured as follows:

AHT= (Total Chat Time + Total Wrap-Up Time) / Number of Live Chats Handled
In order to balance customer satisfaction and a speedy resolution, you may want to try aiming
for an average handle time of 14 or 15 minutes. Keep in mind however, that average handle
time varies greatly by industry and by types of queries handled. We have seen some companies
work with an ideal AHT of as low as 4 minutes, and others who aim for 20!

A low average handle time might indicate that an agent has the skill, efficiency, and knowledge
needed to help his or her customers fast. At the same time, it might also indicate that an agent
is rushing through chats, and is not fully solving his or her customers’ issues. Meanwhile, a high
 average handle time might indicate that an agent needs additional training to be able to solve
issues more effectively. However, it might also show that an agent is taking on complex
customer issues and that he or she is dedicated to solving these issues the first time, no matter
how long it takes.

Here are some additional things to consider when measuring your average handle time:

1. Are you putting speed first? Speed isn’t everything. If your live chat agents are rushing
through live chat conversations simply to get customers in and out, then their needs are going
to be in direct opposition to the customers’ needs. While this metric has its uses, it is important
to never drive agents to value speed over issue resolution.
2. Are your agents’ chats simple or complicated? Another problem with the average handle
time metric is that it does not differentiate between complex, involved cases, and simple ones.
This means that it is important for you to use this metric in conjunction with an investigation
into live chat transcripts, or other metrics that measure the quality of an agent’s live chat
abilities.
3. Are your agents taking advantage of all of the live chat tools that are available to
them? Maybe your agents’ average handle time is higher than you would hope, but they are
doing a great job at taking care of customer inquiries and issues. If so, ensure that your agents
are properly trained in areas such as how to use live chat shortcuts, canned messages, etc. to
save time.
You can check this metric in the agent efficiency and the agent performance reports. For more
information on how agents can reduce average handle time, check out our blog post: How to
Reduce Your Average Handle Time Fast.

V. First Contact Resolution

First contact resolution, or FCR, indicates whether a customers’ issue has been resolved during
their first contact with your company. In other words, this metric measures what percent of the
time your agents are solving customer’s issues in a single live chat session.

Increasingly, FCR is being recognized as the number one of the most (if not the most)
important metrics to watch in customer service. According to TELUS International, a study
conducted by Customer Relationship Metrics found that “CSAT (customer satisfaction) ratings
will be 35%-45% lower when a second call is made for the same issue.” This shows just how
important first contact resolution is to customer satisfaction.

Making first contact resolution a priority, not an option, can be extremely effective in avoiding
repeat contacts and reducing queue wait times. Encouraging agents to prioritize this metric also
helps increase customer satisfaction, and reduce the amount of effort that customers have to
put forward to find a solution.
 First contact resolution can be tricky to calculate. If the issue is resolved as a transfer, does it
still count as first contact resolution? The consensus is generally no. What if an agent marks a
customer issue as resolved the first time, but the customer disagrees? Again, no; but this
discrepancy can make first contact resolution hard to measure precisely.

In the end, the customer is the ultimate expert on him or herself. One way to measure FCR
effectively is to ask customers in the post-chat survey if their issue was resolved the first time.
If you are relying on the agent checking a box in wrap-up mode that indicates whether the
issue was solved the first time, you will need to conduct periodic audits to make sure that
agents are marking accurately.

For information on how your agents can maximize their FCR rate, take a look at our blog post:
One and Done: How to Optimize Your First Contact Resolution Rate.

VI. Invitation Acceptance Rate

If your company uses a proactive live chat strategy, then the invitation acceptance rate metric
will show you how well that strategy is working.

You can measure your invitation acceptance rate with two different reports: the auto-invitation
report and the manual invitation report. The auto invitation report reveals the number of
invitations triggered by predefined rules as well as the invitation acceptance rate within a
certain time period. The manual invitation report reveals the number of invitations sent
manually by agents as well as the invitation acceptance rate for any given time period.

Consider the following when analyzing this metric:

1. Are enough customers responding to your automated or manual chat invitation in

general? If your numbers are low, consider whether you are targeting the right customers and
whether your default chat messages greeting is enough to get your website visitors to interact
with you. You may also want to consider whether your proactive chat invitation is visible
enough and well-designed.
2. What does your chat source report tell you? The chat source report shows how many of
each type of chat has been accepted, and can help you gain insight on the effectiveness of your
proactive chat strategy. Consider whether your manual or your automated chat invitations are
more effective, and what you can learn from each to more effectively target the right customers
at the right time.
 VII. Sales Conversion Rates

Are you using live chat as a lead generation tool? If so, it’s good to be able to track whether
your company and agents’ efforts are working. This number is very much influenced by your
automated and manual invitation acceptance rates.

Every company’s sales rate goal is going to be different, depending on the intensity of their
proactive chat strategy. A low sales rate could mean that agents need additional training in
good sales and upselling practices. It could also reveal a need to revise and rework your
automated and manual proactive chat invitation strategy for maximized sales capacity.

Check out our blog post, Sales Tips and Tactics: How to Up Your Game When Selling to
Customers for tips that will help your agents make more sales.

VIII. Visitor Logs and Wrap-Up Notes

Visitor logs and wrap-up notes are important for categorizing your chats, and for adding any
important details to customers’ cases. Agents whose visitor logs and/or wrap-up notes are left
incomplete, mismarked, or excluded could cause problems in the future with customers who
need to make a repeat contact.

Consider the following when studying visitor logs and wrap-up notes:

1. Are your agents adding the appropriate notes and wrap-up to the chats? Perform an
audit to ensure that your agents are marking their live chat messages appropriately, and not
marking a junk message as an inquiry just to leave it open for 15 minutes.
2. Are your agents remembering to wrap-up every time? Wrap-up notes help you better
categorize and manage your company’s live chats. This information makes it easier for you to
identify any repeat issues or frequent problems with products or processes. Therefore it is
important that your agents always wrap-up a chat once it has been completed.
3. What can you learn from your wrap-up composition? What is your percentage of
complaints as opposed to inquiries or suggestions? This information can help you see what area
needs the most improvement.
You can analyze this information with the wrap-up report.
 IX. Customer Satisfaction Scores

Customer satisfaction is a metric that is vital to determining the success of your live chat team
and the quality of company procedures, policies, and products.

Customer satisfaction can be measured several ways. Some options include the following:

1. Net promoter score (an index that measures a customer’s willingness to recommend the
company’s products and services to others),
2. Customer effort score (a score built by a single question survey that asks if the company
made it easy for the customer to handle his or her issue),
3. Loyalty measurements (whether customers have remained loyal to your brand or are
starting to shop with the competition)
4. Customer satisfaction surveys (surveys about the customer experience delivered via post-
chat pop-up, email, customer interviews, focus groups, and more)
One of the most comprehensive of these measures is the post-chat customer satisfaction
survey. By having your agents encourage customers to take the customer satisfaction survey,
you will receive a more consistent, accurate spectrum of feedback. This is because if your
agents do not encourage customers to take the survey, it is possible that only customers who
had an especially great experience or an especially awful experience will fill it out. This results
in data that is not quite reflective of the customer service experience as a whole.

You can access the customer satisfaction survey results under the post-chat survey report. To
ensure that you maximize the amount of customers who are taking your post-chat survey, try
keeping the survey short and sweet. If you aren’t receiving enough customer feedback, click
here to learn more about agents can promote the customer satisfaction survey.

How To Use Your KPIs As Part of an Ongoing Performance Plan

Set Clear Expectations

In order for your KPIs to work as they should, your agents must first know what is expected of
them. As much as you might love all of your metrics equally, don’t try to give them all equal
emphasis. Instead, think about what customer service values are most significant to your
company and how to go about attaining them. This way you can establish which metrics—such
as first contact resolution and customer satisfaction—should be emphasized as your top
priorities.
Once you have identified your top priorities, make sure that you are setting expectations that
aren’t in direct opposition to one other. For example, telling agents to prioritize both quality and
speed might make striking a balance tricky for them—what’s success on one measure is bad
performance on another. This can result in role ambiguity, which can be stress-inducing
(unreasonable expectations typically are!) and detrimental to staff.

You should also remember to make sure that the information and benchmarks that you give
your agents for each measure are as clear and specific as they can be. For example, instead of
telling agents to keep their average handle time “low,” tell them to keep it under fifteen
minutes or less (or whatever number is your company’s standard). One agent’s definition of
“low” may be different from yours, and it’s best to keep things as clear as possible up front if
you want your company’s standards to be met.

Assess KPIs Throughout the Review Cycle, Not Just at the End

To truly meet the goals that your company has set for your KPIs, it is important that you assess
your key performance indicators throughout the review cycle, and not just at the end. This is
because sometimes your agents are not thinking about what metrics they are hitting or
missing—they’re just thinking about getting through the day.

Checking your KPIs periodically will help you know where you stand as far as which areas your
agents need to work on. With this knowledge, you will be able to remind your team of the goal
as appropriate.

Catch Performance Slips Right Away

By reviewing your KPIs, you will be able to detect and correct performance slips right away. You
can best catch performance slips by using the live chat reports and tools that are available for
management.

One key auditing feature that managers have access to is live chat transcripts. By appearing
offline and reading agent transcripts live (or right after they happen), you can track metrics
such as first contact resolution rate, customer satisfaction rate, and whether average handle
time is being achieved at a running pace or with quality insurance in mind.

By being attentive, you can notice any slips in the quality of agent performance and address
them right away, so that customers are not affected in the future. This helps agents learn and
grow by avoiding bad behavior.
 Set Goals Going Forward

Moving forward, it is important to set goals according to your KPI findings. Consider the
following questions when deciding on your next steps:

 What metrics do you want to improve by your next review cycle?

 What changes need to be made to improve any of these metrics?

 Is any additional training required? If so, what sort of training, and for which agents?

 How can you get all of your agents on board?

Conclusion

By keeping track of your must-watch key performance indicators and adjusting accordingly, you
will be able to provide a live chat experience that is easier, faster, and more satisfying for your
customers, as well as more profitable for your business.