Here at Olark, we’re committed to building integrations with complementary analytics, CRM, and help desk platforms that will boost our users' business.
Google Analytics (GA) is one of our most popular integrations, and is arguably the most popular web analytics service in the world. Data from BuiltWith shows that nearly 76% of the 100,000 top websites in the world use GA. And here at Olark, we found that more than 40% of our paying customers use the Olark and Google Analytics integration.
However, for a long time GA was a major customer support bottleneck for us. We knew enough to handle most basic GA questions, but none of us were GA experts or used GA on a regular basis, and so as a team we would struggle with deeper or more specific customer questions.
Whenever a customer would come to chat with a GA question I used to cringe. Segments, custom dimensions, and metrics - it’s like they were speaking a different language. I had a choice: to live in constant fear of Google Analytics; or to queue the montage music and train up.
Luckily I did not have to go on this journey alone. Peter Shin, my colleague who is also a master of movie/TV references and all around great dude, decided that together we would become Olark’s GA experts.
Our plan was as follows (in case you want to follow in our footsteps):
- Put our noses to the grindstone and study! (Specifically we studied Google Analytics course 1 and course 2.)
- Make it official and get GA Certificated
- Educate our fellow Olarkers by doing an internal show-and-tell
- Revamp our GA Help Guides for our customers
What is the Olark and Google Analytics integration?
So you may be asking yourself, why get so involved in GA when we work at a live chat company?
Olark provides its own robust reporting metrics around the chats that take place on your site, but by combining and viewing your Olark and GA data in one place, you can get more actionable insight than you could from either one alone, and answer questions like:
- What % of your visitors chat with you?
- What % of your visitors are on mobile?
- On which pages do visitors start chatting with you most (and least)?
- Do customers who chat with us convert more? If so, how can we make it easier for them to start a chat with us?
For example, you can use the GA integration to identify key sales pages that are generating too few chats, and experiment with adding “Click to Chat” buttons to pages and sending automated proactive messages to customers with Targeted Chat rules to boost chat rates.
We decided to build this integration because without data it’s hard to pinpoint what areas of your site need improvement. Do you have a page that is failing to generate sales? Try putting a targeted chat rule to engage your customers. Are support pages generating a lot of chats? Try modifying and improving the content so that answers are easier to find.
What documentation we offer
We understand that GA can be a bit intimidating in the beginning, so we've tried to provide enough documentation to get you started:
- How to turn the integration on
- Reporting insights such as, "How busy is chat right now?"
- Conversion tracking like an eCommerce revenue report
- A detailed case study on creating a live chat revenue report
We also have a 10 page whitepaper from our friend David Kutcher over at Confluent Forms that provides step-by-step instructions for creating live chat segments in GA (when you're ready to level up).
What’s new?
At the start of this article we mentioned that part four of our plan was to revise our help guides in order to pay it forward and share what we learned with our customers.
In particular we created two new guides which highlight some of the simpler, more high-impact use cases for our GA integration:
- Insights to help administrators determine some basic chat metrics, such as “How busy is chat right now?” and “What % of our customers are chatting with us?”
- And Conversions to help you determine if live chat is helping you convert your visitors and increase your revenue.
Let us know if you have any questions we can answer, or if there is additional documentation you'd like to see. Thanks! :)
