New

Herald helps teams take a data-driven approach to user feedback. Learn more

How OpenPhone is Dogfooding Itself to Product-Market Fit

With Daryna Kulya, Co-founder

OpenPhone is a business phone app for startups, small businesses, and individuals. They make it easy for any company to manage direct lines of communication with their customers via phone or text.

OpenPhone has grown to eight people with two full-time support representatives. The team is partially remote, with the rest working out of their San Francisco office.

Daryna Kulya is co-founder of OpenPhone. Jay and I recently caught up with Daryna to learn about how they dogfood OpenPhone for collecting and organizing customer feedback.

Tell us about how OpenPhone got started.

My co-founder Mahyar ported his phone number to VoIP as an experiment. He quickly realized how great it was - the quality of calls was way better compared to his regular carrier, it was convenient to use his phone number around the world and it was cheaper too. However, the service he used felt incredibly clunky. We saw a big gap in the market for a business phone service that people actually enjoyed using.

The newly released OpenPhone web app was the result of numerous customers filling out surveys describing why they'd like to use OpenPhone on their desktop and what problems they believe it will solve for them.

These days, the vast majority of founders and professionals still use their personal phone numbers for business. With OpenPhone, we are making it incredibly easy for anyone to own a business phone number, collaborate on messages and calls with their team, connect it with their tools and use it anywhere they are, on any device they own.

What major channels do customers provide you feedback?

  • Phone - calls and texts (using OpenPhone, of course)

  • Chat widget

  • Email

  • Social media

  • Surveys

  • Phone interviews (also using OpenPhone)

Who talks to customers?

  • Founders

  • Customer Support

  • Product

What happens once a customer reaches out and asks for a feature or brings up a bug?

It's very important for us to get insights to the team. We think about it a lot. We care about it a lot—getting insights from customers and keeping the team in the loop.

We use Notion for our internal knowledge base and have extended it to include a board for customer insights. There, we have a ranked list of features that people want, and for each feature, we have a list of all the people that have wanted it in the past year. It's nice for us to know the number, to have a sense for how popular a request is. This is key. For now, we just want to know how many people want something and who they are.

As we grow, this is probably going to get out of hand. Our number one feature request for example, there are so many people asking for it. If we want to email all the people who asked for a feature when it's ready, it's very hard. The process is also a bit manual. Our customer support team, they have to remember when they are talking to someone to put the insights they're getting into Notion. There are definitely people and data points missing as a result of this. For example, if I'm talking to someone directly through email about a feature, I might forget to add it there. I'm proud of the fact that we're logging everything, but it falls on a team member to make sure we do this.

Seem like you're getting a ton of volume. How does your team manage to process all of it?

The text medium is great for us, better than live chat because there's no expectation for an immediate response. With live chat, you really need to be there immediately and at least respond. With text, it's OK to take a half hour or so to dig into a problem before providing a response. A huge bonus is that it let us dogfood our own product.

Volume is a good problem, but it is a problem. I still handle quite a bit of processing feedback myself. But as a founder, I do want to focus more on other parts of the business

There is a difference between how people talk about a feature and what the actual feature is. For example, you might say I want a shareable UI for my phone messages. But really the feature is that they want a shared phone number. So they might be asking for X, but we have to understand that they want Y.

For our most requested features, we actually just put a link to the feature on our web site, but that just leads to a Typeform for customers to request access. We're doing this as knowing who wants something is not enough. We want to know about the company they work for. Are they a current customer? What do they want out of this feature?

This survey is connected to our customer insights Slack channel. Every response gets pushed there, so the entire team is able to see it.

Ready to take a data-driven approach to customer feedback?

Herald is the best tool for teams to collect, tag, share, and analyze user feedback to make data-driven decisions.

What value does your team get out of this feature request survey?

There is a difference between how people talk about a feature and what the actual feature is. For example, you might say I want a shareable UI for my phone messages. But really the feature is that they want a shared phone number. So they might be asking for X, but we have to understand that they want Y.

For smaller features, we don't need a survey. The feature is pretty straightforward, it probably has a much smaller scope and is well defined. The other thing I like is that when customers ask for one of these big-ticket features over Intercom, we can add a "block" the the response message that embeds Typeform directly in the chat window. That makes it super easy to get this information.

I also want visibility about requests and insights—I don't want to be the only one with this information. The thing is me and my co-founder, we read these responses all the time. The reason we have this in Slack is because before that, I would have it in my head, but no one else did. Now everyone has access, and our team really enjoys reading the customer's direct feedback.

How does your company prioritize product development?

To be honest, I wish I could tell you I'm analyzing everything. The truth is, at least at this stage, what we're finding is that there's a disproportionate amount of people wanting 2-3 things. So that's what we're prioritizing.

This board helps us keep track of requests and get a rough count of how popular something is. I care a lot about reaching out to these customers when we release these features. It's still tedious, getting the email addresses out of Notion and into our CRM, but at least the information is there.