Grafton Robinson is CMO at Kin, a Chicago-based home insurance startup. Read on for a transcript of his presentation from our September 19, 2017 Here’s How Startup Marketing Conference, where he describes how he made every marketing channel a growth driver for his company.

My name is Grafton Robinson and I’m with Kin. We are a homeowners insurance startup based in Florida. We were basically founded out of how horrible it is to buy home insurance. If you’ve ever tried to buy home insurance there are always some obscure things you need to know, for example, a fire class protection code, which nobody ever knows. Our founders were trying to buy homes at the same time and realized how terrible it was, so they thought, “What if we could just get all of this data online and make it so that we can give you a quote for how much your homeowners insurance would cost with just your address.” And now, here we are.

Basically, since our launch we’ve more than doubled our client base in three consecutive months, we’ve got a partnership with Guaranteed Rate, a lending company, and we’ve been able to grow a lot via word of mouth because everybody that starts with this sort of thing knows how terrible it is. People know somebody that buys a home or is buying a home and they can start to tell their friends who will tell their friends, etc. We started marketing in earnest in June, when I came on, and then announced our launch on August 1st.

We had a pretty ambitious goal at the start – we wanted to double our growth month over month.

We’re in Florida exclusively at this point so geography is limited and customers often pay via their mortgage, so our product is something that they might consider for about 15 minutes every year. Since there’s no bill or credit card charge that they’re getting every month, we needed to be at top of mind during that 15-minute window when they realize how much they don’t like home insurance and then try to swoop in at that point.

We also needed to build a attribution model and tracking mechanism because we hadn’t been marketing and were starting all of our marketing at the same time. We were also building the backend to determine whether or not everything that we’re doing is actually working. It’s like building the plane while you’re flying I guess, but that’s the nature of what we’re doing, and here’s how we did it.

  1. We evaluated every potential marketing channel

We looked at everything: digital, paid search, affiliates, partners, how much it cost to get somebody a Kin tattoo, etc. There’s one street in Florida that has a bunch of different homes that represent the type we’d be insuring and we used a bunch of them for all of our testing. We looked at a lot of different things, like how much it would cost to throw them a block party that they didn’t know about.

  1. We determined when these channels can help, from “immediately” to “2018” to “probably never”

We determined when they could help because we needed to get something up basically the day after I started. We looked at timing and targeting, so how granular our targeting could get. We’re only active in Florida and we’re obviously only looking for adults that own a home. Ideally, we’re looking for what their deed date is, or when they’re buying a home, or if they’re in the process of buying the home at that moment. We could do Florida Public Radio, but how great is our tracking going to be if everybody just goes to the website and we have no idea where they came from? And we also need to know tracking because we’re doing all of this at once and we want to be able to know both what works and on what channel it’s working. If we know this, we can go from that channel and apply the learnings to another channel, so what works in a text ad or a text message might then go into an email subject line, etc.

  1. We decided on KPIs – not all channels can be measured the same way

And then we started looking at KPIs for each channel. We looked at KPIs by the channel, the audience, and the tactic. A social media campaign might have different tactics depending on whether or not we’re reaching out to potential partners or end clients. Frankly, we were brand-new at this. We didn’t know if it would be worth it to go for an email or to try to get them to sign up right away. There’s a lot of data filters, we had to ask “Is it worth it to start softer and push at the end or just push right from the get-go and try to get them to sign up?”

  1. We focused on building the best possible experience for each channel

Because we were starting from brand new, we said, “Somebody’s on Facebook. What is the best experience for someone coming from Facebook, for example, to ultimately buy insurance? Do we want to send them a text message?” Spoiler alert, the answer is yes. We never want to send anyone an e-mail if possible. We found out with hurricane Irma that over 99 percent of our clients had given us their phone number and it was a cellphone number, not a landline, so when we texted them, we only got two bounce backs from going to a landline. Everything else was received. It’s extraordinary the degree to which everyone uses cell phones for every piece of information that they could possibly want and how receptive they were when we texted them. I would have guessed that perhaps they’d be skeptical about getting a text message from their insurance company, but it turns out that they loved it.

Once we built the KPIs and thought about how we wanted to do this, we needed to create this memorable experience that would work when they were trying to go from four insurance companies down to two, then down to one. They have a local agent calling them, they have us, and they have anybody else that they’ve found online. If we know when they bought their house or that they’re buying a home, somebody else in Florida knows that they’re buying a home as well. The best part about doing this with a one person marketing department is that I was able to share all of my insights across channels and take what we learned with one and apply it to the other, then take it and apply it to the product that’s on the actual website so that we can integrate it into the marketing copy.

The Results

We’ve had 126 percent sales growth in the first month, 140 percent from July to August, all the cross-channel learnings. We have a pretty clear path to get to our end-of-year goals, which, like I said, are pretty ambitious. We built a tracking apparatus, so now not only do we know where everybody’s coming from, but we’re ready to scale up when we can get other channels built up. We also can start to build cross-channel attribution because we have that 15-minute window where we have to be top of mind. It’s not going to be one touch point for these customers, we need to know how several touch points work together to build a marketing funnel and to do it across devices, which is also an excellent capability for us to have. And we have data on all these different channels that can help us figure out and go back and revise what our issues were with the first channels and push through and grow more channels.