Testing ads on iwillteachyoutoberich for 3 months

October 3 29 Comments latest by MH

In the spirit of my post on experimentation yesterday, I’m running a 3-month test of advertising on iwillteachyoutoberich (www/RSS/newsletter). I’m not sure whether it will work or not, but it’s worth a test.

Here are the metrics that I’m using to measure success/failure:

1. Lack of distraction to me / my writing
2. User feedback
3. Revenue

I’ve run this site for three years without ever running ads, so this is new for me. If you have any feedback, good or bad, please let me know. I take it really seriously and will reply to every single email. In trying this, my only data points are some other PF bloggers’ feedback and a study I ran in May of 2005, which showed that 82% of iwillteach readers wouldn’t mind unobtrusive ads. As Vanessa Williams eloquently put it, “Oh how the years go by.”

Anyway, this is one of the small, rapid tests I’m running. If you’re curious about what short-term experiments you could do, see yesterday’s post on experimentation and another article I wrote, Think in Weeks, Not Years.

(No Ratings Yet)

 Loading ...

29 Comments latest by MH

How a beggar in Grenada uses data to optimize donations

October 2 9 Comments latest by Björn Klose

“…the country demands bold, persistent experimentation”
– Franklin D. Roosevelt, 1932

In 2005, I visited the Caribbean island of Grenada. My sister, who was living there at the time, took me shopping in the local bazaar, where my sister pointed out a young boy who was about 8 years old. He carried a notepad with him.

He was a beggar, but no ordinary one. Every day, he would vary a small element of his presentation — his clothing, his grammar, his request style — and measure exactly which one produced the largest tips from shoppers. This little boy was one of the most sophisticated amateur social psychologists I’ve ever seen.

The key was incremental improvement every day through data. He had hunches that he subjected to testing, not handwaving. This is one key differentiator that, I believe, allows people to be an entrepreneur, by which I mean having a disproportionate impact that’s larger than the average person. Instead of just going on gut, they make lots of small, incremental improvements, measuring each one along the way. Doesn’t this sound familiar?

How data relates to personal finance and entrepreneurship
The obvious way data relates to personal finance is with a budget. Create one and you’ll see surprising results you never expected (e.g., when I did, I discovered I was spending 70% of my money on food).

When it comes to entrepreneurship, data takes an even more important role. The very best entrepreneurs I know have hunches, like everyone else. But where they excel is in testing those and relying on the data, rather than their ideas. How many meetings have you been in where someone said, “That button should be red!” I have sat in those at other companies and wanted to take a komodo dragon, swallow it as much as possible, and let it casually dine on my entrails as the meeting continued so I could be let out of my misery. Lots of people argue and debate minutiae like this for hours. But the simplest way to do it is to test it by putting up an A/B test and seeing which one produces the best results.

This is on my mind right now because we’re in the middle of some detailed experiments at PBwiki which are producing double- and even triple-digit surprises in terms of conversions, signups, and engagement levels. In other words, by tweaking a few things, we can see huge results.

Do we waste 1/3 of our life doing things that don’t produce results?
Anyway, I’ve had this nagging feeling lately that about 1/3 of the things we do don’t produce any results. In other words, even if we didn’t do them, nothing would seriously change. Following that logic, maybe 1/3 of the things we do sort of matter, and 1/3 are critical (e.g., getting enough to eat, to pay rent, and to be physically and emotionally healthy).

I don’t know if I’m right or wrong, but I’ve started tracking my time to see exactly what matters. It’s just like tracking my spending — it’s hard and boring but has already given me something surprising to think about.

So with my 1/3 experiment underway, here are ten examples that highlight the experimental approach.

• Generic champagne tastes better than many expensive champagnes, a blind test revealed. The literature on blind tests of wine and coke are legendary. If you think you can taste the difference, you probably can’t, and I would take that bet with anyone.
• Geektronica wondered if he was saving money using Netflix instead of his local video store.
• Aaron Schwartz, a tech entrepreneur, writes about Seth Roberts, who exhaustively tested his theory for a new diet on himself. Instead of just hoping to lose weight, what if you kept detailed records of what you did, how you felt, and the results?
• Dr. Susan Blackmore writes a gut-wrenchingly honest account of why she’s given up her life’s work of chasing paranormal activity. The short version: She went with the data.
• Rob constructed a test to see how to slow down speeding cars in his neighborhood.
• UPS, one of the kings of data analysis, instructs drivers how to insert their keys into the ignition, put on their seatbelts, and get to their routes in the most efficient way possible.
• Investors do better with fewer choices, something Barry Schwarts also writes about in The Paradox of Choice.
• “According to their time-diary analysis, only 26 percent of Americans in 1994 went to church weekly, although the Gallup poll for the same period reported the figure at 42 percent.”
• “Consumers often say they want to be socially responsible when it comes to buying food, clothing, office supplies, and the like. But consumers’ noble sentiments are not often reflected in their actions at the checkout” ($ required).
• Finance professors think they can beat the market, even though that’s clearly irrational. A friend of mine works at a Very Large Finance Company that you have definitely heard of, and they conducted an internal study that showed their fancy-pants portfolio managers allocated their personal assets no better than the average investor.
• Do Cramer’s recommendations produce results better than chance? No.

If you’re curious to read more, see my links on expertise, research, and data and, if you’re a big nerd, my optimization links.

How does this apply to you?
Over two years ago, I wrote a post called 10 things about yourself that would surprise you. In it, I ask what would happen if you started tracking certain things in your life like how much money you spend and how often you call your friends. What would happen if you tried it today?

(No Ratings Yet)

 Loading ...

9 Comments latest by Björn Klose

Previous Page