This is the third and final post in a series from Nicholas White, the co-founder and CEO of The Daily Dot.
It was one year ago this month that I moved to Austin, Texas, in full dedication to creating The Daily Dot, the hometown newspaper of the world wide web. This week, we officially launched the site.
First of all, I do not consider myself an entrepreneur. I actually kind of resent the term. I don’t “solve problems” wherever I find them, I’m not especially interested in starting something up for its own sake, and I’m not doing this for an “exit.” For me, starting The Daily Dot was propelled by the same three reasons that Vanity Fair editor Graydon Carter gave the New York Times for owning a publication: “One is to create great journalism. Two is to have fun. And three is to make some money.”
The past year has been an interesting and exciting one, filled with challenges, questions, answers and, ultimately, even more questions. It wasn’t simply a matter of navigating the usual entrepreneurial challenges of how to create and launch a successful startup, but more specifically, how to create and launch a successful media startup. That endeavor, we realized, is an entirely different proposition, and there is very little guidance or communal knowledge available on it.
Here are a few of the lessons we’ve learned along the way.
Raising Money is Hard
We started in the logical place: raising money.
So I found myself making a deck for three months last year. I know I made umpteen versions and had, by the end of it all, probably hundreds of slides.
The process was driven by a constant back and forth with angel investors. My co-founders, Josh Jones-Dilworth and Nova Spivack, and I went through a few drafts ourselves and then decided it was ready. We took it out to our “friendlies,” people we felt would either bite or tell us exactly why they wouldn’t. But we were also going for the smart money. Reactions ranged from kindly advice (and a lot of it) to, in one case, actual anger. Really, our deck — which was only emailed to this individual — with its graphs and charts actually moved someone to rage.
In part, the challenge, I think, came down to the fact that the economics of a media startup are VERY different than those of the average tech startup. In fact, part of our thesis was that so many media startups fail because they try to behave like tech startups. Everything, from Backfence to Patch, is about creating a technological solution and keeping human expenses as low as possible. That only works if you’re creating a content farm.
When sites have gone (comparatively) premium — The Huffington Post, Politico, TMZ — they can be successful, but they are expensive. They have a lot of employees. They also make a lot of money. Fast. Politico, for example, made $4 million in its first nine months, TMZ $3 million in its first 11 months. But claim you might hit $1 million in your first year, and it just looks like self-delusion to the experienced angel. And what the hell? Maybe it is.
Most importantly, however, I have realized that we were just selling an idea. Half of everyone thought it was a great idea. But an idea is not a company.
The other half, by the way, thought we were nuts. My favorite reaction actually came in a different context, but it was typical of some angels’ reactions. As I explained that we’d report on what happened in Facebook, Twitter, and World of Warcraft, my listener interrupted me: “Wait, you’re going to report on fiction!”
Yes, yes I am.
A couple people, chief among them, Troy Williams, one of our wonderful advisers, and Dudley White, my wonderfully impatient father, sat us down and told us to stop screwing around trying to raise money from top-tier angels. Just get to market however you can.
So we took their advice, got commitments in a quick friends-and-family round, and began the year looking for an executive editor.
Excellent Foundational Hiring is Essential
Hiring, as anyone who’s ever done it knows, takes a stupidly long time. I am especially prone to underestimating this fact, however, because I’m the kind of person who, for better or worse, decides to leave his family’s company and move to a new state in a matter of days.
And, it turns out, hiring people into a company with no track record, history, revenue, office — nothing — isn’t as easy as hiring usually is. Especially when they have a good job they already like, and a demanding schedule, as did our founding editor Owen Thomas. While we talked with a lot of people, many of whom I still have enormous respect for, we selected Owen as our editor because he had varied experience, from Time Magazine and Suck.com to ValleyWag and VentureBeat, and because he has a true passion for the online community itself.
Owen and I tested the limits of Skype for a while, and eventually we brought him out to Austin for SXSW. I don’t know for sure what eventually convinced him to do something crazy, but I like to think it was the karaoke.
We announced Owen’s hire on April 1, because we’re journalists and thus perverse. It was reported, declared a hoax, then declared true. It was glorious. We figured we’d launch in late June/early July. Folly!
Hiring reporters in an existing company with a good system is something that can be done fairly rapidly, as hiring goes. It ended up taking about twice as long as I expected in this environment. That was at least partly because we realized that without any existing momentum to catch people up and carry them along in the current, we needed to have just the right people to start out with. We had to be extra choosy.
Be nimble with the platform
In the meantime, we’d been building a website as a sort-of proof of concept. We’d also been working with a few freelancers to create sample content. And we even did some experiments to verify our content’s inherent appeal — articles about the online community are about 50 percent more compelling to social media users than the same stories about the offline world.
By early April, we decided we wanted to build the real site in Django — we needed something powerful enough for our big ideas. We had started the design process, but it went a little slow, and I underestimated what we would need. In the past, I’ve always worked with a full-time designer on staff, so we did enough mockups to get happy with the look and feel in advance, and then we would create mockups and design things on the fly as needed during the site-building process.
The question for us became, however, how do we beta-test a media site? First impressions matter, and if readers come to a news site for the first time and find it lacking, then there will be few chances to woo them back. The impulse is to impress out of the gates, and yet we were generating quality stories from our web communities that needed a home before it was built.
Part of the beta-testing strategy from the start has thus become one of the basic tenets of The Daily Dot: While we, of course, will have a central website for all our stories, our true end point of publication and distribution lies within the communities upon which we are reporting. This only makes sense.
So while we built out our official site that we launch this week, we were producing individual stories and posting them in a variety of formats and communities. If we’re covering YouTube, why would we not produce and distribute within YouTube? We posted stories as public Google docs, as Facebook notes, and even Reddit threads, until finally we could post and share them as isolated stories on our own domain. It was more about disseminating the voice of The Daily Dot rather than driving traffic to our site.
This strategy was unorthodox, but we feel effective. It gave us time to get the site properly developed, and our editorial team time to ramp up, while also helping integrate The Daily Dot into the communities we covered.
Yet development of the site took much longer than we had anticipated. We were working with a customized CMS (The Daily Dot is one of three publications — including The Texas Tribune and Bay Citizen — helping develop the new Armstrong CMS), and we soon realized that we didn’t start with enough people and should have hired someone to just plan and run things. Everyone was a coder, but we were rapidly at a position where planning was itself a full-time job.
And if I were to step back, I would say that not enough planning is an endemic problem. I think this has been exacerbated by the physical and functional space between us. On one hand, we decided to be virtual — we cover virtual communities so it would seem hypocritical to make everyone move — and frankly, we could get the best people if we didn’t make them move anywhere. But simply being in the same room helps create a sense of urgency; it makes it easier to ask questions and get into serendipitous conversations, and it puts you face to face with the people relying on you.
Our virtual newsroom has put Yammer and Campfire and Skype to some extreme tests, and we’ve managed to establish a functional newsroom routine between these. Yet I feel this is not so much a function of those tools themselves as it is the dedication of the team we’ve put together, and their commitment to our online and virtual vision.
Lessons Learned, in summary
1. Hire, hire, hire.
This is job No. 1. Get the team in place, and get them in place early. You can never have someone too early — you can always manage budget or other timing issues by having someone keep their current job for the time being and start in a month, or three months.
2. Plan, plan, plan.
In a startup, everyone’s impatient just to jump in. There are so many unknowns, and you need to get things done — or so you think. But really, your grandmother was right: measure twice, cut once. There’ll be plenty of the make-‘em-ups anyway.
3. Establish norms of doing what you say you’re going to do early.
Startups are so formless, so routine-free that it’s easy to let plans and tasks be kind of vague. Start small and conservatively, but everyone has to commit to being utterly reliable.
4. Don’t compromise on people.
People make any business. But early on, there are no norms, no momentum to sweep people along. Everyone is going to feel like they’re dragging a big boulder uphill. There’s going to be lots to improvise, and relatively little room for learning on the job. You need people who are ready to operate on the best information they have, and yet still question everything they know. You need people who can take surprises in stride, who are ready to plan, but also to act.
5. Get to market.
The reader is the ultimate and only judge. Getting the reader to care is the only convincing you need to do.
And in the end, the uber-lesson is this: Minimize the X factor.
All of the above comes down basically to this principle. To be an entrepreneur, you have to be risk-tolerant — even risk-seeking. When it comes to the actual process of starting a business, however, the job is all about trying to eliminate risk. Being out early and iterating often with your product (something we did very well, actually) is about verifying your assumptions. The same principle goes for the internal process. There will be plenty of surprises on the way; your job is to eliminate as much uncertainty as possible.
There’s a rich literature online at this point about tech startups. Much of it has been useful to me, but starting a media company has its own unique trials as well, and we need to be sharing those and creating a similar literature. Creative destruction requires a huge rate of innovation — otherwise it’s just destruction. If we can eliminate, as much as possible, the foolish reasons that media startups fail, and create a communal knowledge of success, we’ll have a better home for a bright future for journalism.
Why I Gave Up the Newspaper to Save Newspapering by Nicholas White
The Necessity of Data Journalism in the New Digital Community by Nicholas White