Lean Cuisine

Lessons in running Lean Startup practices from the comfort of home


I live in a house in South Austin along with 3 friends/co-workers. At any given time, we have no less than a dozen different half-baked ventures in various stages of “launch.” It’s a glorious mess of entrepreneurial spirit, desperation, and hubris. I wouldn’t call it an “incubator,” per se — more like a collective (but without all the hippie commune connotations).

We generally try to follow a kind of Lean Startup/Agile Development process wherein we identify interesting opportunities and execute product experiments in two week sprints. Usually we sit around the kitchen table on Monday mornings, brainstorm until our heads hurt, then commit to 1-3 “projects” we want to test over the next two weeks. At the end of those two weeks we make a decision whether each project is worth continuing, pivoting, or trashing.

Today we share our first tale of success, which I hope is succeeded in time by many more. This story is about an app we built for this year’s SXSW called FOMO//ATX. My friend Ed and I thought of the idea just 10 days before SXSW this year, launched it right as the festivities kicked off, and were pleasantly surprised by the response it garnered. It was an incredibly fruitful experiment in Lean Startup practices that I think is worth sharing.

Here are a few things we learned:

This headline is about as tired as Gaga’s act.

Lesson 1: Listen to the Hipster Masses

“SXSW just ain’t what it used to be…”

How many times have you heard some variation of this cynically nostalgic trope over the last few weeks? A couple hundred? More?

Us too.

And to be fair, it’s true. I’ve been attending SXSW for 5 years now, and even in that relatively short time period it’s changed an absurd amount. It’s so much bigger, so much more commercial, and so much harder to navigate. Still, if you know where to look it can be so unbelievably fun.

When you unpack that cynically nostalgic motto, I think what you actually get is a pain point that can and should be solved. In this case, that pain manifested itself in people’s inability to navigate the madness that has become SXSW. It’s not that there aren’t great events going on, it’s that most people get distracted by all the over-branded noise.

Lesson 2: Combine/Refine/Re-Appropriate Past Successes

A few years ago, Twitter was absolutely indispensable during SXSW. Pre-planning for SXSW was always kind of a joke when so much of the best stuff is unofficial and spur of the moment, but a quick #SXSW twitter search yielded hundreds of festival goers sharing insights and tips. These days, that same search yields millions of brands and robots retweeting platitudes and fake rumors. So we asked ourselves a question:

What if we made a tool that was as useful as Twitter was 4 years ago?

What would that mean? How do we solve that uniquely SXSW pain of standing in line, wondering if there’s a better party somewhere that I can actually get into? How do we address that that crippling “fear of missing out” — FOMO — that seemingly fuels SXSW.

That was the insight we ran with: If we could enable spontaneity and cut through the bullshit of SXSW, we’d win.

So on Friday we had the idea, and by Sunday we had drawn up a few wireframes of what it might look like. Our original concept was to curate the best content from the last 30 minutes of Twitter activity. We’d pick out the best things and show them in a scale of how popular they were at that moment. This immediately presented a few technical issues: namely, we knew we’d have to filter out a bunch of crap (we’re looking at you #chevysxsw). We also realized that we might need to present more stable and topical data.

Lesson 3: Look for Broken Processes as Indicators of Opportunity

This led us into the great world of South-By spreadsheets. For the past few years, there has been a growing market of unofficial spreadsheets that are passed around by people in the know. These spreadsheets list every single event going on during the festival and some basic data about the event.

In years past, we’ve used them extensively to plan out our calendars, but in the moment, they can be unbelievably cumbersome. So much SXSW time and attention was wasted on this broken process of checking Google Drive for your spreadsheet, then Twitter trends for late-breaking news, then a massive group chat for personal recommendations. Our aim became broader and purer than replicating the value of Twitter. It became a mission to fix this painful workflow.

For both time and complexity’s sake, we decided that the app really only needed two features — one for “What’s Next” and another for “What’s Now.

Next” showed events that we thought were interesting in the next few hours, and “Now” was a modified heat map of what was bouncing around Twitter. All of it came with an appropriate amount of snark and irreverence. We were the no-bullshit guide to SXSW. Did an event require a badge or hand out tacos? Free Newcastle and Ice Cream? Long lines that you needed some gimmicky app to bypass? We had you covered.

Lesson 4: Know When to Automate; Know When to Curate

We did the vast majority of engineering, design, and copywriting in 48 hours because we knew that we had to wait 7-10 days for Apple’s App Store review process. The pressing time constraint meant we had to make important architectural decisions early on. If we wanted to be able to change a feature on-the-fly, it came at the cost of increased bandwidth usage, which on Austin’s ravaged cell networks during SXSW means almost certain death. We baked in most of the design, layout, and icons within the app itself for submission and only lazy-loaded text content.

We powered “What’s Next” by hand. We made our own highly-curated Google spreadsheet and basically treated it as a CMS. Each day at noon, I holed up at a coffee shop and picked what was going to be good the next day. All in all, I documented and commented on 150+ different SXSW parties.

For the “What’s Now,” we tried to be more automated, and we more or less succeeded. We’ll probably do a technical write-up at some point, but essentially we pinged Twitter at frequent intervals and grabbed any tweet that mentioned SXSW. Then, we identified popular collocations (common groupings of phrases) in tweets from the past 30 minutes and ran it through an algorithm that cut out common words and repetitive content.

The saving grace was making all of the app management accessible from our phones. Because we were out in the mix of things, anything that wasn’t automated could be updated in real time by hand. If a party ran out of beer or a headliner cancelled, we’d push new content and make a note of the change.

Lesson 5: Be a Part of the Conversation

As we mentioned, Twitter tends to turn into a useless mess during SXSW. However, finding an audience was critical to the success of the app, and Twitter is still a viable method of doing that organically and cheaply. So, we found a way to engage with Twitter/Reddit/Facebook in a compelling fashion and drove an extensive amount of traffic to our app.

#SXnoobs was a hashtag where SXSW veterans were giving advice to people who had never been before.

How? By being a human. We carried over the snark from the app itself to Twitter. Leading up to the start of South-By, we started interacting with the community.

Once things got started, we made fun of events as they were happening

We got lots of love from the existing Guide- To-SXSW community and loved hearing from strangers.

We had thousands of downloads of the app, which were pretty evenly spread out over Interactive and Music weekends. Our repeat usage numbers were phenomenal (the average user opened the app 44 times throughout the week), and seemingly every line we stood in included one or two folks refreshing their apps, using it precisely during that use case we were solving for.

The Moral: Providing value (at SXSW at least) isn’t as hard as brands make it look

How many brands strive to have a “presence” at SXSW and end up getting lost in the shuffle? Hundreds? Thousands?

To my eyes, for every Doritos or HBO — companies who spend millions to leave no doubt that conference goers are aware of their brand — there are 20-30 mid-level brands spending thousands on parties that only a few hundred people ever even hear about, never mind attend. They rent an overpriced bar for a night, pay someone from the Jersey Shore to DJ, and give away useless knick-knacks to the first lucky, lanyarded few who stumble inside.

Rather than fight for attention shilling the same swag as everyone else, another, infinitely more appealing option is to get scrappy, think creatively, and actually provide value to attendees. This is a lesson that I think is applicable not just during SXSW, but to big-tent marketing at large. Building tools and providing services — be they apps, shuttle services, whatever — that offer inherent value other than as a marketing gimmick is a viable solution to brand erosion.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.