Taking off the rose colored glasses

On Danielle Morrill’s suggestion I’ve been reading a book called The Happiness Hypothesis. The author, Jonathan Haidt combines ancient philosophy (including my personal favorite, Stoicism) with modern social and psychological research to paint a picture of a happy life that I find really compelling. What struck me however is how applicable a lot of what he says about the mind is to startups.

The thing that stuck out the most is a passage where he describes how bias works. He talks about a study in which pairs of research subjects are given a real legal case to read. One is assigned to play the defendant and one is assigned to be the plaintiff, and they are given real money to negotiate with. He says:

When both players knew which role each was to play from the start, each read the case materials differently, made different guesses about what settlement the judge in the real case had imposed, and argued in a biased way. More than a quarter of all pairs failed to reach an agreement.

However, when players didn’t know which role they were to play until after they had read all the materials, they became much more reasonable and only 6 percent of pairs failed to settle.

What he’s saying here is that when players knew their roles in advance, they genuinely read the information differently and came to different conclusions about what a reasonable outcome from the case was. This reminds me of how people evaluate their own startup ideas. Here’s how it works:

1. They come up with an idea

2. They get very excited about it

3. They avoid doing real research on competitors.

4. If they happen to come across information that indicates that there might be problems with their idea they dismiss it immediately.

5. If they come across ANY positive sign that what they’re working on might be valuable they get incredibly excited about it, and cite it as a reason why their idea will work.

6. They think that their blind insistence that their idea is The Next Big Thing is why their vision will come to reality

Conversations with these people tend to have a very similar script.

Founder: “We allow companies to view their support requests from customers as threads and then post those threads publicly to avoid answering the same questions over and over again.”

Me: “What about Zendesk? Doesn’t that do the same thing?”

Founder: “Yes, but Zendesk only allows you to submit requests as text. We’re going to allow customers to submit videos of themselves describing the problem. I read an article in TechCrunch last week that said video is the future of online communication, and we’re super passionate about videos so that’s why we think we can win.”

There are a few problems with this approach. Continue Reading

25 Oct 2012, 5:32pm | 3 comments

Selling Umbrellas in a Synagogue

A few months ago, Twitter invited me to participate in their advertising program. Basically what they do is allow you to purchase follows via a Promoted Account and purchase clicks, favorites and retweets via Promoted Tweets. I poked around the interface for a little while and ultimately decided not to try it out. To me, there’s something kind of icky about purchasing Twitter follows (even from Twitter itself).

But I got this in an email from them this morning:

Given that they were nice enough to credit me 100 bucks I decided to put aside my misgivings and give it a shot. We’ve been thinking a lot about online advertising recently at my startup, and so I thought it would be interesting in comparison to Adwords and LinkedIn Ads.

I set aside $50 to spend on a “Promoted Account” and $50 to spend on “Promoted Tweets”.

The Promoted Tweets function allows you to pick a few of your tweets and have them show up in other peoples’ stream. I selected a few of my most scintillating 140-character tidbits and let ‘er rip.

Those are pretty cool results! An 8% CTR is rather astoundingly high by online advertising standards.

The next thing that I noticed is that promoted accounts don’t seem to work too well:

They tell you to optimize your bio so people can connect with you more readily. I neglected to do so and forged ahead with all of my current settings. The result was that for $11 I got 6 followers.

My first thought was “this is dumb”. The ROI on something like this has to be very low. For the Promoted Tweets, I did get 130 clicks, but I couldn’t target those clicks at all beyond geographic region. And as far as the followers, I already have 3,000. How valuable could 6 extra really be?

For a total of $61.81 spent it didn’t seem like I was getting very much.

But as I thought about it more, I realized that what made that money seem like such a bad deal, was tied to the reason that advertising on Twitter feels icky to me: when you throw ads onto Twitter, you start treating it in the same way that you would Google. You begin to start measuring things like CTR, conversion rate and ROI.

But the way people interact with Twitter is fundamentally different from the way people interact with Google. And so measuring things like CTR and conversion rate on Twitter is completely missing the point. Continue Reading

18 Oct 2012, 9:00am | 11 comments

How to build a blog readership

Discuss this post on Hacker News

Despite the recent success of this blog over the past year (250,000 uniques, hundreds of new subscribers, republished in Lifehacker and others) the truth is that I’ve been a failed blogger for far more time than I’ve been a successful one.

The other day I got curious and went back to look through sites I had written on before I started my current blog. I found a two: one primarily written in 2005 called Techcast (I used it to promote my tech podcast) and one written in 2007 called Sanity for Dummies. Side note: I was 14 in 2005.

Go take a look at them. They’re horrible. Nobody read them. Frankly, I don’t even know why I bothered to link them up.

But really I think it’s a good example of how something that looks like a very fast road to a big readership really took a long time to build. If you visit my blog today it looks like I’m some kid who threw up a couple of blog posts on the default Posterous template and started raking in the traffic. In reality though, it took a long time for my writing and my experiences to catch up with my ambitions and begin producing returns.

I get asked a lot to give advice on how to blog. Having gone back and read through my old posts (this one is by far the best – seriously I was a lot funnier when I was younger) I thought this might be a good time formalize a few of my thoughts on how beginners can approach improving their blogging skills. There are a lot of patterns to be found in poor blog posts, and I’ve done my best to formalize a list of them and discuss how they can be avoided. So without further ado here are a few tips for aspiring bloggers. Continue Reading

2 Oct 2012, 11:24am | 21 comments


Never miss a new post