For a new company, there's no better channel than Facebook to help you scale quickly. But it's not the only channel you should be relying on.

Facebook is a volatile platform. Today, the amount of competition is massive compared to the start of the decade. In 2010 and 2011, the CPCs (cost-per-click) and CPMs (cost-per-thousand-impressions) were unbelievably low, and it was easy to get traction using Facebook. Now, you have to fight for your audience with increasingly higher spend.

And that means you'll eventually need to diversify into new and emerging channels if you want to continue driving growth without ruining your marketing budget.

It can be risky to enter new channels, but if you take a smart, data-driven approach to testing and scaling new channels, the potential rewards far outweigh the risk.

I spoke with our Senior Director of Growth, Nisho Cherison, about the approach our marketing team takes.

Here's what he recommends:

1. Look for three key traits when choosing a new channel.

The first trait is scale. Is there enough potential here to scale your efforts and drive traffic and sales? If it's a small-level display channel, you could end up spending a lot of time and money trying to make it work, but you may not get the scale because the inventory is too low.

"The demographics of the channel are also important," said Cherison. "You have to ask: Does your target demographic really use this platform? Or are you testing it simply because it's popular--without much representation from the people who will actually purchase your product?"

Lastly, think about the possibilities for iteration. How quickly can you iterate on ads and creative concepts? It should be easy to tweak messaging and targeting so you can see what content resonates with the audience.

If the channel has the potential for scaling, exposes you to your target demographics, and allows you to iterate on your message quickly, then it's likely a good option for diversifying your marketing.

2. Expand beyond the basics.

Always stay on the lookout for new marketing channels to use.

"At ThirdLove, we've found success with Pinterest, direct mail, Instagram, and even podcasts. We're currently testing both Snapchat and television commercials--two channels with little in common," said Cherison.

It's important to continue testing new and unusual channels if you want to maximize your reach.

For instance, channels like Snapchat or Pinterest don't take vast resources to test. They may not have the reach of Facebook, but if the CPA (cost-per-action) is comparatively low, you can still end up getting a lot of value out of your spend.

Pinterest isn't often seen a top channel for companies, but we're seeing a lot more success on this channel than some other advertisers. The performance surprised us, but it just goes to show that different platforms may work better--or worse--for different companies.

You'll never know until you start testing, so don't be afraid to try something different.

3. Move fast and take risks.

If you don't take risks, you're not learning. If you're not learning, you're not moving forward.

The truth is, you have to take risks in order to make progress. You don't have to jump in feet first, but you can test a channel using a smaller audience to get a good idea of whether it will work for you.

When we launched direct mail last year at ThirdLove, we started off very small. We were relying heavily on Facebook advertising, and we knew it was time to diversify into something else. We started out by sending a few thousand pieces of mail the first month. And we doubled it the second month.

The results were amazing. We actually acquired higher-quality customers through our direct mail than we did through Facebook. Once we saw the initial results, we decided to move fast. We didn't want to miss the boat on direct mail, so we took a risk and went all in. We made the decision based on the sound data we got back from our initial efforts. It was a huge jump, but it paid off. Direct mail is now our second-largest channel after Facebook.

Remember, just test small, get some results quickly, and if you like what you see--go full steam ahead.

Published on: May 23, 2018