Best Practices: Six steps to success with your Answer Bot Follow
When working with Answer Bot, many customers asked for a best practices guide, something to help them set up and use the feature in the best way possible. So we’ve documented six steps that will help you set yourself up to be as successful as possible using Answer Bot.
Write the right articles!
In many cases, we find that customers haven’t done enough prep work before they start wanting to use Answer Bot. The first step should be a no-brainer, but we urge you to do it anyway: Take some time and do a quick audit, maybe of the last 7 days, to find tickets where customers really could have self-solved their problem.
For instance:
- Review tickets where a macro was used to solve the problem in one-touch. Could there be an article for this? Go write that article.
- Review common tickets that were just answered in one touch: Could there be an article for those that would have allowed customers to self-solve? Go write that article.
Write succinct, single-problem articles
Answer Bot really works best when the question and the solution to the problem is stated succinctly at the start of the article. You don’t need to have a very long article just for the sake of it. Keep it short and sweet, and keep it about just one thing. If you write a single article that addresses every question about a feature or topic, you might find it doesn't work very well at all.
Use article labels to reduce the "noise"
We implemented the ability to whitelist a subset of articles to ensure that only the right articles were being used when searching for the best suggestions for those customers. Writing your articles is one thing, but using labels on your articles will ensure that the only the selected articles you choose get surfaced to customers. Take a look at the best practices for using labels with Answer Bot to learn more about how best to use labels and for information on creating articles and adding labels, see Adding articles to the knowledge base.
Segmentation (trigger conditions) + article labels = winning
The power of article labels really comes through when you have a situation where you can segment your customers when tickets are created. A good example of this is a game developer supporting iOS and Android platforms. The developer has two channels for incoming tickets: one for iOS issues and one for Android. Using the trigger conditions, you can create 2 Answer Bot triggers (one for each channel) and then use article labels to serve up only iOS or Android articles where they make sense. So, stop for a second and really think about how you could best segment your customers and create useful conditions around them - and then serve them the best articles possible! This is described in more detail in the best practices for using labels with Answer Bot guide. For information on working with triggers for Answer Bot, see Creating new triggers for Answer Bot.
Check your placeholders!
It almost goes without saying, but if you don’t put gas in the car, the car isn’t going anywhere. The Answer Bot placeholders are the key part of the this feature working - make sure you set them up properly using the {{answer_bot.article_list}} and {{answer_bot.first_article_body}}. Make sure you check out the full Setting up Answer Bot, which goes into more detail.
Insights is insightful
We’ve built all the key data points into Insights to allow you to really monitor, report and slice away at your tickets and their deflection data. There is so much you can tell if you spend the time building a few reports (or just digging into the default ones).
About 48 hours after you begin using Answer Bot - that is, after it's been turned on and used to respond to tickets - the Deflection tab is added to your Insights dashboard. This tab is where you can see how Answer Bot is working for you - take a look through our Analyzing your Answer Bot activity article to learn more, and for information on using Insights, see Setting up Zendesk Insights.
Comments
0 comments
Please sign in to leave a comment.