Last week, we had an insightful conversation with Emre Tekoglu, Vice President of Customer Support at Zywave, a leading insurance software company. Emre's unique approach to scaling customer support and using data to drive improvements offers valuable insights. Here's what we learned:
Last week, we had an insightful conversation with Emre Tekoglu, Vice President of Customer Support at Zywave, a leading insurance software company. Emre's unique approach to scaling customer support and using data to drive improvements offers valuable insights. Here's what we learned:
At Zywave, my department size is about 70-plus people, about 40% of which are in the Philippines. They're our employees on our payroll – we don't do BPO. We service our customers over phone, email, and chat.
We've found that servicing our customers over chat is very effective. When you take a phone call, you can only handle one at a time. But with chat, we can do two simultaneously. We've found that two is the optimal number. It's twice as effective as phone support. From chat, we can also easily transition to a screen share if needed. This year, I've been focused on growing the chat channel adoption. We're now at about 40% of our incoming cases coming through the chat channel.
As a software-as-a-service company, we need to be on the front lines to communicate when there are login issues or outages. We've been focused on updating our status page within 15 minutes if there's an outage. Year to date, we've had 443,000 views on that status page. That's potentially 443,000 customers who didn't have to call us because they got their updates there. It's another way we're trying to scale support efficiently.
We've created a career path in customer support where we promote people to stay in support. One of the criteria we look at is people's ability to recognize patterns. If they see cases coming over and over, I want them to start asking how we can prevent those types of cases from coming in. It doesn't have to be a defect with the software; it could just be a volume driver.
I expect them to work with our R&D department and tell that story through data. We use AI tools to measure customer sentiment, and we present this data to R&D to help lower case volume. We can't just add more people to scale support – that's not the smart way.
We have 30-plus products that we service our customers with. One of those products is a learning management system. We gathered details using our tools in Salesforce and another tool called Support Logic to look at case volume related to performance issues. We presented this data to R&D, showing a higher volume of issues relating to one particular product.
Development worked on fixing those issues in the month of June. As a result, we had no performance-related issues that month. This was through us gathering data, presenting it, and explaining to R&D what issues they needed to fix.
We also track in-product NPS surveys. Our scores had been trending downward from March through June, but now in July, it's been going up because we were able to identify and fix those issues. We expect July and August to continue to see improvement in the NPS score range.
Right now, we're working on a more systematic approach, but we consider a few factors. One, we look at the case volume. Two, we look at the sentiment – how much pain is this causing our customers? We can measure these easily in our systems.
We've divided our teams based on products, so they align with the product team. This makes it easier to prioritize within each product because they're not competing for the same resources. Each month, I have my team leads and managers determine the top one, two, and three priorities for their product area.
For me, the most important part of this is the people side. We promote from within for leadership roles – 100% of my leaders came from support roles. I take the time to coach them on how to be effective leaders. Twice a year, we run an anonymous survey where each person rates their manager. In our last survey, we received a 90 NPS score for the question 'My manager is a highly effective leader.'
We're also looking into implementing a large language model type of framework to analyze support interactions. I want to focus on case prevention and whether we're educating the end user to be better. We're not there yet, but that's on my wish list.
Our motto is to trust people to do the right things, give them a growth mindset, and let them fail. This leads to people liking their managers, which means we take better care of our customers.
When I first came to Zywave 3 years ago, one of the challenges was employee churn. We adopted the slogan 'How can we be an award-winning customer support department?' In the last two years, we've won several Stevie Awards, including for use of technology, use of data analytics, customer service leader of the year, department of the year, and thought leadership.
I'm also proud of our focus on case deflection rather than case closures. I bet you won't find any support department that has a better revenue to cost ratio than ours because of this focus.
Our conversation with Emre highlighted several key principles for building an efficient, data-driven customer support organization:
1. Focus on efficient channels
2. Proactively communicate
3. Use data to drive improvements
4. Prioritize strategically
5. Invest in people
6. Embrace technology
By applying these principles, companies can create a more efficient, effective customer support operation that not only resolves issues but also drives product improvements and business growth. As Emre put it, "We focus on case deflection, not on case closures," demonstrating a strategic approach to customer support that goes beyond traditional metrics.