What is QA and why do we do it?
When you Google ‘Quality Assessment’, you will find few hits related to QA in Customer Support. You will find information on how to quality test medication, software, cookies, and anything else humans produce. Whenever we create something, there will be a process to determine whether the product lives up to our expectations.
Simply put, QA is the process of gathering relevant data and analysing that data according to a set of predefined criteria.
When it comes to baking cookies, that process is relatively straightforward. You’ll have a recipe to follow:
- Add flour, sugar, and butter to a bowl
- Mix for a specific amount of time
- Bake at a specific temperature for a specific amount of time
- Eat cookies while they’re still too hot (optional)
If you do not follow the recipe exactly to the specifications, you’ll end up with a failed batch and disappointment. It will also be relatively easy to trace back where mistakes were made: was the temperature off, or did you not add enough butter?
It helps you to:
- Repeat the quality of previous batches that have sold well.
- Prevent a bad batch from accidentally being sold.
- Teach new bakers to do a good job.
Having good conversations with customers is more complex than baking cookies; we can all agree on that. But the reasons for keeping track of conversational quality are the same.
Customers who interact with your brand typically talk to only those in Sales or Customer Support. That means the quality of their work will be seen as a direct reflection of whatever it is that you’re selling – especially when that product is digital. You can smell and taste a cookie to find out whether it’s to your liking. Rating your accountancy software through taste and smell is ineffective 🤷🏻♀️
You will want a team of support agents who:
- Can all communicate in a similar fashion so they create experiences that are recognisable for your customers.
- Can teach new hires to follow the same expectations.
While many conversations in support will follow similar structures, they will differ for each customer – unlike cookies, which look exactly the same. That makes it harder to come up with a ‘recipe’ of what good quality looks like when providing customer support.
An additional problem is that everyone has a different opinion on what the quality of a conversation should look like and has some form of bias (we’ll talk more about that in a future lecture). That makes it even more challenging to get an accurate view of the team’s performance.
Implementing a QA program for your Customer Support team is a way to objectively determine if the work done is up to standard.
The purpose of this course is to provide you with a framework. By the end of the course, assuming you have followed along and did all the exercises listed in your workbook, you will have a complete script for your QA program.
The QA framework that I have developed follows a typical data analytical structure, consisting of 6 steps:
- Initiate – making sure you have everything you need to start off well.
- Gather – how to collect data
- Clean – sanitising data so it is as objective as possible
- Analyse – making sense of the collected data
- Present – how to pour your findings into human language
- Redefine – how to restart the process and make sure it continues to work
Throughout the course, you will be directed to fill in your own findings and definitions in the workbook. Feel free to reuse the document whenever you start a new QA program, or are planning to revamp what you’ve got.