Building on Slack Saved Our Startup

The perils and perks of building on a platform

“This isn’t working… we’re screwed.”

Those are not the words you want to be saying out loud three months after you quit your jobs to start a company.

You especially don’t want to hear the reply back from your cofounder, “Yeah, I think you’re right…”

From the start, we did everything right with Tettra. We validated our assumptions through customer interviews and fake mockups. We got 10 teams to agree to use our product before we ever touched a line of code.

Our idea was to build a simple wiki for sharing knowledge on a team, and everyone we talked to seemed really excited to use the product.

Then we launched our MVP…and no one used it.

When we reached back out to those team to figure out what had happened, we found that they all still had a problem sharing internal knowledge. They still believed they could use Tettra solve it. The problem was that no one remembered to use the product.

Even after all our research, we missed an important truth about how the world has changed: The way people access a tool is just as important as what the tool does. People are overloaded with information, logins, and interfaces. And here we were, asking them to create a whole new habit.

Around the same time we were trying to figure out how to save Tettra, Slack launched their app directory. This felt like the right move.

A year later, it’s been one of the best strategic decisions we’ve made as a new startup. If you’re considering this path for your business, here’s what you should be taking into consideration.

The Perils of Building on a Platform

There’s been an explosion of new platforms in the last few years. Facebook Messenger, Slack, Medium, and many others have opened their ecosystems, letting third-parties hook into their massive user bases. Jump-starting your growth by building on another company’s platform is enticing, but not without its risks.

Platform can shut down or cut you off

A few years ago Twitter opened its doors and let third-party developers build on their platform. New companies sprung up building all sorts of tools on top of Twitter’s API, including dozens of alternative clients to Twitter’s native web app.

Then when Twitter had grown large enough and needed to start monetizing their audience, they locked down their APIs. Many of those companies, including most of the alternative Twitter clients, shut down because they couldn’t access the data they needed to make their products work.

Another more recent Twitter-related example is Meerkat. Meerkat launched a live-streaming video app in early 2015 and used Twitter for distribution. It took off like wildfire, growing to millions of users and raising millions of dollars soon after launch. But unknown to them, Twitter was quietly working to acquire a competitor in the live-streaming space called Periscope.

Twitter eventually disabled Meerkat’s access to their social graph and promoted Periscope instead, cutting off Meerkat’s primary distribution. The blow proved fatal and, six months later, Meerkat shut down.

And Twitter’s platform strategy is not an isolated event. More recently, LinkedIn, Netflix and ESPN all locked up their public APIs to approved partners only, shutting off access to open third-party developer ecosystems. Unless you can pay, you can no longer play on these platforms.

Betting your business on an open API can be risky business.

Our risk assessment

We talked about this risk a lot at Tettra. For us, this decision came down to two things:

1. Platform investment: Slack is investing heavily in their platform and even has a dedicated investment fund to support 3rd party developers.
2. Revenue model: Slack’s platform is focused on businesses. Its likely monetization strategy is subscription revenue vs. ad revenue. 3rd party revenue from external developers lines up with their business model and is good for their bottom line.

The platform can clone your entire product

Earlier this year Reddit finally added native image hosting to its service, ending a long-standing alliance with Imgur. Now Reddit users have a choice — they can keep using the third-party app Imgur or they can use the built-in option when posting an image to Reddit.

It’s yet to be seen what the effects of Reddit offering their own image hosting service will be on Imgur, but it’s probably safe to say that the majority of users will choose the path of least resistance and go with Reddit’s option.

When you hook into a platform, the owner of the platform gets to see all your usage data. Information like how many times your product’s been installed or how much money you’re making through the platform’s billing system is realistically accessible to the platform owner.

In other words, if you have a popular app, the platform knows it’s popular. In the age of ever-growing juggernauts vying for attention and wallets, this is bad news.

Our risk assessment

This is a big risk for us. Slack’s actually an acronym that stands for Searchable Log of All Conversation and Knowledge. As a knowledge sharing product, you can see why that’s worrisome.

At SaaStr Annual 2016, Slack’s VP of Product, April Underwood, gave some good advice to developers:

‘’If the end goal for what you’re building is to make Slack using better, … then I would recommend trying to think more broadly and to try to solve a broader problem than trying to make the experience on [the] platform a little bit better.”

We feel like Tettra fits the bill of “thinking broader.” Synchronous and asynchronous communication are complementary and should be integrated together, but separate tools. Time will tell what happens as Slack expands their product offering, but for right now Tettra and Slack are symbiotic and work better together.

You don’t own your destiny

Let’s get away from software platforms for a minute and talk about media platforms. What happens if you build an audience on a platform and that platform shuts down? You lose your audience.

This happened earlier this year when Twitter shut down Vine. People who had spent years building up a following on Vine lost their entire subscriber base.

The same thing can happen if a platform decides to shut down a mission-critical endpoint that your company relies on. After raising $49M and growing to 33M users, BranchOut, a service to see where Facebook friends worked and help you find a job, ended up having to shut down. Facebook changed the way its API worked, cutting off developers from accessing data about friends-of-friends and the way apps could post to people’s Facebook walls.

Our risk assessment

Our second version of Tettra was built 100% on top of the Slack API — we didn’t even have a database when we launched. Doing so allowed us to quickly validate whether people wanted a wiki connected to Slack.

And the move was great for us. We went from concept to live product in 2 weeks and received over 200 sign-ups within the first 2 days of launching.

We also had a plan to move away from this heavy reliance on Slack ASAP. We knew from the start that, if we saw product-market fit, we would rebuild the product to house our customer’s data in-house. For us, the plan to rebuild and store our customers’ data made Slack a safe choice.

The Perks of Building on a Platform

Now let’s talk about the positives. As a geek, Slacks APIs are some of the best I’ve ever worked with for a few reasons.

1. You can tell from the consistency of the APIs that the Slack API team put a lot of thought into the design of it all.
2. The entire company is super responsive with support inquiries for developers. They even opened up an endpoint for us last year so we could launch our MVP on time.
3. Most importantly, the user base is super enthusiastic about third-party apps and we’ve seen great adoption from teams who want Slack to be the backbone of their company.

Being connected to Slack is one of Tettra’s main selling points and helps new users find us. From there, we’re able to sell them on the vision for a simple, internal wiki to share knowledge with the rest of their team and easily explain why Tettra’s the best solution for them.

But beyond just Slack, there’s plenty of good reasons to build on platforms.

Access to a massive amount of users

Most of our early sign ups for Tettra came from our Slack App Directory listing. Even a year later, it’s still a steady source of traffic to our site and new accounts for us.

Distribution channels take a long time to build up and scale. Piggy-backing off the customer base that another company has already built is a good way to jumpstart your early growth while you build your own acquisition sources over time.

Clear value proposition for prospects

Tettra is an internal wiki for teams that use Slack. In fact, the product only works for teams that use Slack and there’s no way to use the service without it.

This value proposition limits the upper bounds of our growth to the number of teams using Slack, but it’s also crystal clear: we are the wiki for Slack teams. If you’re using Slack, you want to use tools that deeply integrate with it and are tailor-made for the Slack experience. We’re able to deliver a great user experience for a small group of customers. We can always expand our offering beyond Slack if we hit the upper bounds of the market.

Another great example of this strategy that I saw recently is Upscribe, a new subscription widget. There are literally hundreds of subscription widgets on the market right now, so you’d think there wouldn’t be room for another one. That’s not the case though, as Upscribe got almost 1,000 votes on Product Hunt and received such an overwhelming response that the load actually crashed their site.

So how did they do it? They went niche. Upscribe is the subscription box for Medium. If you write on Medium, there’s a clear reason to choose Upscribe over the other options out there. By starting specific, Upscribe has a foundation to build out their user base.

Buffer started life as a post scheduling app for Twitter. Once they got traction on Twitter, they expanded their offering to include other services along with additional social media management tools teams. Now they’re doing over $12M in revenue each year.

TroopsHQ, a Slackbot for Salesforce users, recently raised a $7M Series A round, just a little over a year after founding the company. If you use Salesforce and Slack, you probably will use Troops too.

Starting on a single platform gives you a clear value proposition to use in acquiring your first customers.

Increase product engagement & retention

The conventional wisdom is that it takes 21 days to build a habit, with some studies finding it might take even three times as long. Habits take a long time to form.

Getting people using your product regularly is just as challenging as building a new habit. In Hooked, Nir Eyal outlines a strategy to actually get people using your product regularly through a habit formation loop:

By integrating with Slack, Tettra’s able to hack that habit formation loop because we show up where our users hang out every day: in Slack.

Through Slack’s APIs, we can push trigger notifications about recently published Tettra pages right into Slack. There are 2 benefits to this:

1. It’s easier for the person publishing a page to spread the word without having to send an email or manually post a link in Slack.
2. Our Slack integration also helps with product engagement because the rest of the team sees the Slack notifications, which subtly reminds everyone else to share their knowledge with their team by creating their own pages.

This kicks off the loop again and helps build up a documentation habit inside the entire company. Hooking into another platform and integrating with your user’s existing workflows removes steps, making it more likely that people will continually use your product.

How to safely build on a platform

If you’re still worried about building on a platform, you should check out Josh Elman’s talk from the 2015 Weapons of Mass Distribution Conference about How to Build On Other Platform.

His four main points for building on platforms successfully are:

1. Get users to come back directly to you: Bring people in through the platform, then figure out a sustainable way to get people coming back to you directly.
2. Remember the value exchange: If you’re building on a platform that monetizes through ads, don’t build a service that pulls away eyeballs from them. Figure out a way to drive back value to the platform in the form of either users or money.
3. Build features 20–1000 on the roadmap: All companies have limited development resources, but don’t build a feature a platform is likely to build into their core product themselves. Focus on features further down the roadmap that the platform is unlikely to get to anytime soon.
4. Platforms are an accelerant, not sustainable fuel: Take advantage of the moment but don’t rely on a platform. The rules can change at any time.

Platforms are the future of growth

Competition in today’s world is fierce. In 2006, the average number of choices a consumer had when choosing a product was four or five. In 2016, that number has spiked to approximately 15 vendors. Creating an amazing product just isn’t enough anymore. They key to winning is being the best at distribution and showing up where your users already hang out.

People don’t want another login to remember or interface to learn. Connecting to existing platforms like Slack makes it easier for your users to access your product and helps you grow. If you’re early enough to a platform and build on top of it right, hopefully eventually you’ll grow big enough to become a platform yourself.

Enjoy this post? Hit the ❤ button below.

What do you think? Is building on a platform a good idea or too risky? Share your thoughts and stories below.