🎯 How to measure your support team’s performance

Do not index

This is the second in a series of blog posts that we will be releasing. It’s geared towards founders who are trying to set up a ticketing system for their business.

At this point, you’ve now got a framework for how to handle your customers requests. (If not, jump back to the first article in this series.) But any framework is only as good as you’re able to put it into practice.

So how do you know if you’re doing a good job at supporting your customers?

To understand your team’s performance, you have to start by measuring what matters: the speed and quality of your support.

🏎️ Speed

How to measure

The best ways to measure the speed of a response are:

time to first response is the time from ticket creation to the first agent response

time to next response is the time from any subsequent customer message to the next agent response

time to close is the time from ticket creation to the ticket being closed

Set goals for each of these numbers and keep track of your performance for each metric so that you can figure out where you need to improve the most. If you’re a B2B startup, you’ll need to give your customers SLAs (service-level agreements) when you get larger clients, so it’s important to know what you can reasonably achieve here.

You have the most control over first response and next response times, so these are the two for which you should set SLAs. Internally, you should also monitor the number of times you need to respond to a ticket (known as the number of touches or contacts).

Benchmarks

Your SLAs should be based off of your customer expectations. At this stage, it’s good to keep these fairly simple and consider them by channel only.

Here are some benchmarks for response times by channel (median times shown, 90th percentile times shown in parentheses):

ㅤ	email	Slack	WhatsApp / SMS / Chat
first response	1 hr (12 hr)	5 min (4 hr)	1 min (1 hr)
next response	4 hr (24 hr)	10 min (8 hr)	5 min (4 hr)

Strategies for improvement

If you’re not consistently meeting your SLAs, here are a few ways to fix that:

recruit teammates to help with support (everyone on the team should be doing support)

turn on notifications so you see new messages wherever you are most (eg, Slack, browser notifications, and phone notifications)

set expectations for each channel with auto-responses telling customers how long they should expect to wait for a reply

create chatbot workflow to deflect easy tickets

write help center articles to provide customers with a self-serve option

switch to a lower frequency channel if you’re still not able to meet response time expectations at this point

✨ Quality

How to measure

At this stage, you should measure the quality of your responses using the customer satisfaction score (CSAT). CSAT is measured as the percentage of users who give a positive reply to the question “how would you rate your overall satisfaction with this conversation?” (the scale is typically either 1-5 or 👍 / 👎). Send these as survey questions right after each ticket is closed.

Benchmarks

Aim to at least meet the startup average CSAT score of 85%. The CSAT response rate is around 19% for chat, 5% for email, and 5% for phone, so it can take a few weeks to establish your first score.

Strategies for improvement

The most common reasons customers are unsatisfied with support are long wait times and agents not understanding their situation.

Understanding a customer’s context early on is fairly natural: you likely know most (if not all) of your customers and probably talk to them frequently enough to be aware of their issues.

Here are a few additional things you can do to help you understand your customers:

use a CRM to track calls, meetings, and vitals for your customers

use a support ticketing system to track customer queries

record user sessions to see how your customers are using your app (especially helpful when you need to understand exactly what their problem is within your app)