Michele Hansen: Learning from your audience

May 13, 2021 · 24 minutes reading time

Transcript of episode 15

Michele's photo

Matthias: Good day everyone. Matthias here again with the next episode of The Audience Explorer. Today, I’m happy to have Michele Hansen here as my guest. She’s an entrepreneur, she lives in Denmark, she’s American. And I’m so happy to have you here, Michele, good morning!

Michele: Hi, how are you?

Matthias: Fine, thanks, hope you too!

Michele: I’m so excited to be here.

Matthias: To introduce yourself a little bit to our listeners, tell us a little bit what you typically do. And how you have started to do this what you do?

Michele co-founded a geo-coding SaaS

Michele: Yes, so my husband and I run a small SaaS called geocod.io, which is a software to convert addresses to coordinates and coordinates to addresses for the US and Canada, and then also to add additional data on top of that. And the basic reason why this exists is because a computer does not understand an address, it only understands the latitude and longitude coordinates.

So what we do is make it easy to convert from an address to coordinates so that a computer can understand it. So for example, you could make a map, and then also to convert coordinates to addresses so that a human can understand them. And then also to make it easy to get other pieces of information that you can only get when you have the coordinates such as timezones.

Matthias: Interesting and … who would use this, the SaaS, what are your typical users or customers.

Michele: So we are a horizontal SaaS, which means that we sell to a wide variety of industries and customer types. It’s everything from university students who are working with geographic data for the first time to fortune 50 companies using it as part of their data intelligence. It’s really a huge breadth of customers, which makes it so fun.

Matthias: And how much time For how long have you been doing this?

A side-project that started back in 2014

Michele: We started the company in January 2014, as a side project, grew it on nights and weekends for about three and a half years. And then I went full time. And my husband went full time, about six months later. So we have both been full time on it for coming up three years now.

Matthias: That’s amazing. That’s already a long time for a SaaS startup!

Michele: It’s, it’s been wonderful. It’s we really never dreamed that it would be our full time jobs. And here we are.

Matthias: So nice. So when you started all this, how come that you got the idea to start geocod.io?

Michele: It was a problem that we experienced ourselves. So we had a mobile app where this is back in 2012, where you could open the app, and it would show you the grocery stores and coffee shops and convenience stores that were currently open. So for example, if you needed milk at midnight, or a coffee at 3am.

At the time, you couldn’t just type into Google and it would tell you, you know, where’s the nearest you know, Starbucks, for example, right? You would instead have to know, okay, I think there’s a Starbucks near here, let me go to their website, let me go to their store locator. Like it was like five clicks deep. And if it’s, you know, if it’s midnight, and you have a screaming baby and need diapers, like you don’t have the brainpower for that. So you just need the nearest thing that happens to be open. So, um, so you would open up and there would be a map.

And the problem was, so we I mentioned that the coordinates can be used to put addresses and locations on a map. We used Google for that. And the problem was you can either get 2500 locations for free per day, you couldn’t store them. Or if you needed more than that you could pay like $20,000 a year for an enterprise contract to get 100,000 a day. And we’re like, we just need 3000 a day like we don’t need, we don’t need 100,000. This is a little mobile app making $400 a month. What are we going to do to keep this going? So we ended up having to build our own very rudimentary geocoder.

And as we just kind of talked about this with friends of ours who are developers, it turns out they had the same problem, which was that they weren’t allowed to store the data from the major providers, and that the pricing was unaffordable for them. And somebody was like, “Hey, why don’t you just like slap a paywall in front of this. And then like, you know, other people can use the API, and maybe it’ll cover the cost of the servers and then you can keep your app going and like this will cover its costs” or like, “Oh, that sounds awesome!”

A $31 success in the first month!

Michele: (continues) So we put it up. In January of 2014, after a couple of months of testing with other developers and two little digitalocean droplets for $20 a month. So our goal was to make $20. That was that was like success was “$20 a month in revenue”. And we ended up with $31, which was a huge surprise and way more than we had ever planned!

Matthias: On which month was that? Was it already in the first month? Or was…

Michele: that was the first month, so we didn’t pay as you go model. And so we charged everyone on February 1, for the first half of January. And it was $31. And it was we were so surprised that anyone wanted to pay us that we actually hadn’t written the code to tell Stripe to bill people.

Matthias: Ah yeah, so geocod.io worked, but the Stripe connection didn’t yet work…

Michele: Stripe, but we had Stripe set up, but we hadn’t, you know, written the code that actually like, you know, did the actual charging. Um, and you know, from the very beginning, um, you know, we lucked out and ended up being on the first page of Hacker News, pretty much the whole day we launched which was a huge surprise to us. Wow. Now, we never ever replicated that traffic. I mean, if I were to show you a chart of our website visits for cross the whole last seven years, you just see this massive spike on the first day, and it’s like a mountain and then everything is flat in comparison.

Matthias: So there was a real need, right, that people were waiting for kind of thing that can…

Michele: Yeah, yeah, the terms of service and pricing pain with the major providers turned out to be, you know, not just something that the people we knew and the people that we had, we had talked to, but you know, a pretty big pain point.

Extending audience and feature set

Matthias: And what would those people call themselves? Would they all call themselves developers, or other other types of audiences who are using your service?

Michele: So when we launched, we were initially only focused on developers. So it’s API only just forward geocoding, which is address to coordinates. And then as we went on, we got lots of feedback from people, you know, we’re listening to them about what they were trying to do and what they were, what they were currently using, and why it wasn’t working for them.

And we, I think, maybe a month later, we added a reverse geocoding API. So that’s coordinates to addresses. And then a couple of months later, I also had people who wanted to upload spreadsheets. And these are more like marketing people. And I remember asking them, like, what do they currently use? And someone’s like, “well, there’s this guy that you can, like, send him your file, and then he’ll get it back to you in a couple of days”. It was like, “oh, that seems like an opportunity for software to just automate all”

Matthias: Yeah, to do it more quickly. Yeah,

Michele: yes, exactly. So um, so our customer area has a huge range. We started out with developers. But um, yeah, I guess the developers told told other people about it, and then they wanted to be able to use it too. And we have a lot of developers who use the spreadsheet upload tool as well. Especially if people might have you know, a backlog of a couple 100 million addresses that they need to run. Wow. Yeah.

Hacker News and other digital watering holes

Matthias: And how did you get in first contact with your audience? You already mentioned Hacker News, what was that the main channel over the other channels also?

Michele: So from the very beginning, I think, since we had, um, you know, my husband is a developer and I was a technical project manager. So we had access to developers just within our social networks. And before we launched, we actually went down to, like, I don’t know, what you call me is sort of a combination between like an incubator, a hackers space and coworking space, that kind of thing that very hot in that early 2010s period

Matthias: Right!

Michele: Um, and kind of, we had some friends who had a small company there. And so basically went there and said, like, hey, like, you know, play around with this, let us know what you think. And then just talk to other people in that in that coworking space. Yeah. And then and then once we launched, we went on Hacker News. And we, we also started kind of posting anywhere where we could find where people were talking about geocoding, whether that was on Reddit or on StackOverflow. You know, Tableau community forums like, you know, those sort of, as Amy Hoy calls it, the “digital watering holes” right there, where people were congregating. And then you know, people were like, hey, like, I have this spreadsheet of addresses, how do I geocode it? And then jumping into say, you can use geocod.io for that and linking to our site.

Matthias: That sounds pretty easy for this type of audience. Really good.

Michele: Yeah, yeah, we’ve been pretty SEO driven from the very beginning. And so, so it’s just a matter of being there when people need it, because people are, they’re already aware that they have the problem. And we just need to make them aware that we exist.

Matthias: Mm hmm. So it wasn’t the the other way round, where people sometimes put out a product and wait for people to show up, right? This this old … what was it? … “Field of Dreams” thing, right? So it will see the opposite, you built something and you went to the people where they already were hanging out?

Michele: Yeah, Yeah, I did. that’s accurate.

Matthias: And what did you do then, after you met those people who were interested in your product? How do you, how do you say that in English? How do you keep in touch with them? And keep the conversation going with with them? What do you do for your audience these days?

Michele: Not much, honestly, you The thing about having a product that people know that they need is that we we can’t really convince them that they need it, you know, like they either, you know, need to know the locations of, for example, the customers in their CRM, or one of my favorite examples is companies that have tractors and tractors now are incredibly smart. And they send back data reports every night on the health of the tractor, and that may need to be time stamped for the specific location so that they know that that one tractor is in Delaware, and one is in Texas, and that it’s all at midnight on their local time.

And, and so when you when having a product that is just something that people need, like I feel like it’s like the equivalent of selling wood into the software industry, just like everybody needs it. We don’t really do a lot of lead nurturing or anything like that, because we can’t control whether they have 500 tractors to get reports from or 500,000. So instead of focusing on trying to sort of convert and upsell individual people, what I do a lot of is interviewing customers and listening for use cases that are both frequent and painful. And also, you know, do have those higher volume needs. And then it’s a matter of speaking more to those higher volume needs, writing landing pages, things like that, basically, so that when people with those use cases are looking for something, we’re already there saying hey, you know, you can use us to play for the geocoding on your real estate listings website, for example.

Matthias: Okay, so you really were into b2b right? You you you’re not so much into b2c space.

Michele: No, we have some b2c but it’s it’s very, very small. The closest we get to b2c is really I don’t know if you called b2c but sort of business education with your you mentioned students and academics like that’s really the closest I did have someone once who told me how they used geo coder to to make a map of all of these locations in a some sort of game they played that was like an imaginary game across the United States with treasures I’m sorry, I don’t I’m probably some somebody is listening and knows what game this is. And they know that I’m totally mangling this. But you know, so they use it for their gaming purposes. But like, that’s pretty much the only true b2c use case I’ve ever.

Matthias: Yeah. Someone the other more b2b? Yeah. Yeah. And how do you find those customers who you want to talk to, to get to talk about new use cases?

Setting up as many customer interviews as possible

Michele: So I set up try to set up as many interviews with customers as I can. And the primary ways we do that are first that we have interviews that trigger that go out to new customers five days after their first purchase, just asking them

  • hey, like, you know, why do you even need this date in the first place?
  • And how did you find us

and then I either try to talk to them over email, what their what their use case is, or set up a call.

And then we also have surveys that go out that’s like just a little one question survey pop up where people rank us from one to 10 for their satisfaction, and then I asked what they use before they use Geocodio. And again, that’s a way to learn more about their use case, what they’re using before, what led them to switch.

I also, there’s a wide variety of research that we do, those are sort of the ongoing research, the sort of I call it sort of “maintenance research”, which is just part of everyday work.

But once a year, I do an analysis of our top 80% of revenue. And then I analyze all of those customers, like an investor, would a portfolio, and then I try to speak with as many of those customers as possible to understand why they’re happy, because it’s much easier to find more happy people more you know, with with use cases that are good fit for our product than it is to try to make our product fit use cases that it’s not designed for.

We will also try to talk to people who have cancelled if we’re launching something new, where there are specific questions about a particular experience or landing page or whatnot. We’ll also do sort of screen share tests. But those are less frequent than those “maintenance interviews”.

Talking to people who are happy with the product

Matthias: That’s an interesting idea. I love that. “It’s easier to talk with someone who is happy and the use case fits your product.” It’s easier than to make someone happy who hasn’t got the right use case. That’s an interesting idea. I love that!

Michele: In many ways, it’s inspired by the concept of loss aversion, which is this concept that comes from Daniel Kahneman and Amos Tversky, which is basically that people are inclined to try to avoid losing something than they are to try to gain something, which is why a lot of custom people who do interviews they will, they might focus on people who have cancelled, and then try to prevent those people from canceling or try to win them back.

The way I use cancellation interviews is if I hear about a use case that we’re not a good fit for, then it’s like okay, maybe we should dial back that marketing because we’re not a good fit for what they’re doing. And instead, what do we learn from the people who are happy, it’s much easier to take someone from neutral to happy than it is from unhappy to neutral, nevermind, unhappy to happy.

Triggering interviews by automated email

Matthias: Hmm. Amazing! And the trigger for the interview is … you said, it’s part partly automated by email?

Michele: So I have an automated email through Intercom that goes out to people five days after their first purchase. So um, so I mentioned you know, earlier that we have a pay as you go model for one of our one of our plans is a pay as you go plan. And so someone may have started using it, the paid version of it, say, you know, the month beforehand, and then their charge the first of the following month, so then that would trigger five days after that. But if they signed up for one of our subscription plans, you know, it’s much sooner. But what’s always really interesting is that just because they added a credit card and started using it as a paid level, very often they were using it for a long time on the free tier before upgrading!

Matthias: Speaking of the free tier, that’s interesting, because I’ve got a problem to solve in my own startup with a free trial. I’m not so sure whether it’s better to ask for a credit card right away and start a free trial nevertheless, or ask for the credit card at the end of the free trial. How is it in your case?

Michele: So we have a free tier, which allows people to run up to 2500 addresses are really lookups (so including all those other data appends) per day, and there’s no credit card required, they only need a credit card if they want to go above that. 2500.

Matthias: Okay, that’s also an interesting approach because it’s related to the value to the to the volume that they get, right. It’s not time related. So it’s, you can use it for several weeks?

Michele: You could use it for years! Actually, I always love when I see that, like, you know, I see someone come in through intercom and on the on the sidebar, it has their stripe payment history. And I always love when it happens when someone has a fairly low user number like you know, under 10,000. And then, you know, it turns out that their payment history is like, you know, $2 in, you know, February of 2016. And then and then it was $885 in January of 2017. And then it was $50 in October of 2018.

And then it’s like oh yes!, like this model really works because you know, people can just use what they need when they need it. And they’re not there. Like they don’t have to be locked into something like we do have a subscription subscription plans, but it’s just, it’s always so satisfying to, to see that nice because you know, we were that customer, right like, like we had that situation where we just we only needed what we needed. Yeah, and no more than that and couldn’t pay for more than that.

Matthias: So what really sounds through is that you do really care for your audience you wanted to, to them, you want them to get value and to be happy!

Empathy for the customer

Michele: Yes, empathy for the customer is the cornerstone foundation pillars of everything that we do we always make decisions from the perspective of “if we were the customer, how would we want this to work”?

Matthias: Mm hmm.

Michele: And often, that means, you know, getting outside of our own perspective, and understanding how things work from their perspective, but not making any decisions that we feel are “hostile” to the customer. Which in many ways is it’s sort of a “goes against the grain” of, you know, we intentionally don’t lock people into things and whatnot.

Matthias: Yeah, it could seem attractive to lock people in but I think it really isn’t. Because if, if everyone is getting forward, including the customer, and the company, then everything is much better!

Michele: If, there’s incentives for the company to do that. Like, it’s very valuable for the company if people are locked in, but not necessarily for the user and a lot of cases and that, you know, that requires a set of circumstances where we’re not incentivized to growth.

But if we had investors, if we needed to grow, if we needed to show revenue growth, then you know, it would make sense for us to to have more aggressive tactics. But since we don’t have any investors, and it’s just the two of us, we can you know, that that gives us the freedom to make the decision that’s in the best interest of the customer. And we really value that.

Matthias: And gives you more peace of mind, yeah!

Michele: Yeah, yeah. It’s just how we like to run things.

What happens in a customer interview

Matthias: About these customer interviews, when when they really begin and get and you get started, can you tell me to have two examples:

  • one where you got an interesting, surprising response to your questions,
  • and when you get got one that you already expected?

Michele: Oh, so interviews always have a combination of those! I always I love it when I talk to someone, and I discover this use case that I didn’t even know existed. And it’s always so excited me like, Wow, I didn’t even know that happened in the first place. And they’re using our product for it. That’s so cool.

Yeah, um, just the other day, I was talking to someone who works for an electric utility. And they were trying to optimize the routes that their technicians take when they are assessing meter readings. So in the US, often this is this is done manually, they don’t have a way to do it automatically. And so they were trying to optimize all of those different routes. And they were they were moving to a new like a new piece of software that they had built that needed the coordinates in order to make these routes.

And, and they’re using geocoding for it, and you know, are the thing I mentioned about tractors, like I just I love that one! Because, you know, just imagining all these little tractors talking to each other. (Matthias laughs) I mean, the farming industry is just amazingly, technologically advanced now.

Delighting in understanding people’s perspectives

And so I always learned something that I didn’t know, whether it’s about why they use our product, or how they use it, or, you know, the different use cases or how they found us, like I just, I genuinely delight in understanding other people’s perspectives and their journeys towards things like I just, I just really find it fun. So, so I love it when I find something new. And then and then also when we find things that we already knew.

That’s validating and the thing I always remind myself is, okay, even if they say, you know, if someone else were to come to me and say, you know, we’re, we’re using you guys, because we are a let’s say we’re a university, and we need to be able to to know which major cities our big donors are near so that when we go and have meetings in those cities, we can make sure we talked to the right people.

So we need to know the Metropolitan Statistical Area to know that someone is in. Yes, they’re in Massachusetts, but maybe they’re in the Boston area versus they’re in the New York area, for example. And so when I hear a use case that I’m already familiar with, reminding myself to continue to ask Questions and ask for clarification, even though I don’t need it because there is always something to be discovered.

Matthias: Ah, so it influences your next question, so to say! When somebody tells you something familiar, you’re familiar with, you still remind yourself, you, you need to ask for more, or you need to ask for more information.

Michele: Always ask for more information.

The future of the SaaS: Optimizing for stability

Matthias: And how do you plan for the future? Now, when you think geocod.io, you’re, let’s say, in one year or two years, what what do you want to make happen, then

Michele: we just want to keep on doing what we’re doing. You know, we’re at a stage of the company where we are optimizing for stability. We don’t really need to pursue growth, it does happen, like we grew over 50% last year, but it wasn’t intentional. So, you know, we’re really focused on making you know, you know, things, you know, just the right we just we just optimized for stability, which is just leads to a different set of priorities than it might if you are optimizing for growth.

Matthias: That’s fine. That’s right. So you don’t plan to hire tons of people and and conquer the world, but you’re keeping things stable and for your peace of mind, right?

Michele: Yeah. I mean, we are excited about turning the the old hen house on our property into a “shed quarters” in the backyard.

(Matthias: “What is that?“)

But yeah, so is the kind of I’ve heard other people with indies as is talking about having a shed quarters, which is basically you have a little a shed in your backyard. That is your quote, unquote, headquarters. Oh, and so we call it Yeah, the the hen house is going to be a project for us so that we can more separate our personal life from work a little bit easier. And you know, if the computers are in a different building, maybe it’ll be harder for us to check in on things at 10 o’clock at night.

Matthias: That’s right. That’s right. Amazing, thank you, Michele, it was so, so nice to hear about your approach to work with your audience. And I wish you very much success for geocod.io, for you and your husband.

Michele: Same to you. It was really nice talking to you!

Matthias: So that’s a wrap for today, my dear listeners! If you want to hear more from Michele, you can listen to her podcast Software Social, which is one of my personal favorites! She talks about running her small SaaS geocod.io and her co-host talks about getting one off the ground, too. You can also find her on Twitter at @mjwhansen.


Outro

Thanks for listening to The Audience Explorer podcast, today.

You can find me on Twitter at @GetTheAudience and you can check out the blog at gettheaudience.com

If you have any questions about this episode, reach out on Twitter or send an email to matthias@gettheaudience.com

If you want to support this podcast, please leave a rating in your favourite podcast player app. This will help other founders or creators to find this podcast about developing an audience for their product or service.

Comments welcome:


Be the first to know.

Do you like this post? Get articles like this one delivered to your inbox, automatically. Enjoy my weekly insights about how to systematically develop an audience for your product or service!

Email
matthias@gettheaudience.com