5 ms·

Good post, Rolf. Keep up the good work. So, this past year, I've tried 3 different business ideas. Project 1: This was a 4 Hour Work Week idea. I built a c

Good post, Rolf. Keep up the good work. So, this past year, I've tried 3 different business ideas. Project 1: This was a 4 Hour Work Week idea. I built a contact database for every hospital in the US, I put up for sale on a yahoo store and I pointed Adsense at it. I had done Adsense testing check for demand. It failed. I thought the from my testing I should have been getting 2 to 3 sales at $150 per sale each month while spending $100 a month on Adsense. I didn't get a single sale in 3 months. So, I lost a grand for the whole project. No big deal. I learned a lot. Project 2: Social cute site. It was a reddit clone for cute pics (cats, dogs, etc...). I shut it down after 3 months, because I couldn't take looking at cute kittens that much. But, I took that code base and what I learned from it to start http://newsley.com http://newsley.com . What I've learned in each one has been invaluable. The argument of whether or not we should try to build a startup to succeed or fail is rather moot, I think. No one builds a startup to fail. But, after having failed at 2 business ideas, I still think it was very worth while. Ultimately whether my success rate ends up being %25 or %29 or even %5. It doesn't really matter to me, because I'm going to keep working at this until I do succeed, and I'm enjoying the process.

ntoshev 17y ago
Do you have a theory why initial adsense test showed demand but it didn't materialize? What did you learn from this?
- iamelgringo 17y ago
  Yeah. I did and original Adsense test in October/November. It was a simple 3 page web app. The first was a sales letter, selling a database of hospital contact information. The second was a price sheet ($150 for an address list of all the hospitals in the US and territories) which clicked through to the third page, which was an email signup form, saying the person was interested in buying it. I spent $100 on Adsense advertising over the course of a month, which got me around two hundred clicks on the front page, 50 or so clicks on the price page, and 6 email addresses. So, after doing that, I scraped the initial hospital contact list from several different sources, and I paid $500 to have a team of people in India verify each address individually. I then threw the address list up on Yahoo stores for a few months in February through May. I think that a big part of the reason that I didn't make any sales was because of the economy. I don't think that people were buying much of anything this past winter. Why else didn't it work? I'm not sure. But, after 3 months, I decided to stop paying for the upkeep and take everything down.
  - ntoshev 17y ago
    Another hypothesis is that people interested in something would click through in hope to learn more util you ask them to pay to continue. It would mean adsense demand testing inherently doesn't work (at least for some types of products / testing setups). Someone could setups the following experiment to test it: after you build the product, split the traffic with most going to the product buy page, but with some percentage going to the original testing page that only collects email addresses. When a person gives you an email, immediately auto-reply with a link to the buy page. In this way you can compare the effectiveness of demand testing at the cost of some sales (because you make some of your potential customers jump through more hoops). But if I went a couple of months without a sale I would try it just to see if my demand testing worked at all.

ntoshev 17y ago

Do you have a theory why initial adsense test showed demand but it didn't materialize? What did you learn from this?

iamelgringo 17y ago
Yeah. I did and original Adsense test in October/November. It was a simple 3 page web app. The first was a sales letter, selling a database of hospital contact information. The second was a price sheet ($150 for an address list of all the hospitals in the US and territories) which clicked through to the third page, which was an email signup form, saying the person was interested in buying it. I spent $100 on Adsense advertising over the course of a month, which got me around two hundred clicks on the front page, 50 or so clicks on the price page, and 6 email addresses. So, after doing that, I scraped the initial hospital contact list from several different sources, and I paid $500 to have a team of people in India verify each address individually. I then threw the address list up on Yahoo stores for a few months in February through May. I think that a big part of the reason that I didn't make any sales was because of the economy. I don't think that people were buying much of anything this past winter. Why else didn't it work? I'm not sure. But, after 3 months, I decided to stop paying for the upkeep and take everything down.
- ntoshev 17y ago
  Another hypothesis is that people interested in something would click through in hope to learn more util you ask them to pay to continue. It would mean adsense demand testing inherently doesn't work (at least for some types of products / testing setups). Someone could setups the following experiment to test it: after you build the product, split the traffic with most going to the product buy page, but with some percentage going to the original testing page that only collects email addresses. When a person gives you an email, immediately auto-reply with a link to the buy page. In this way you can compare the effectiveness of demand testing at the cost of some sales (because you make some of your potential customers jump through more hoops). But if I went a couple of months without a sale I would try it just to see if my demand testing worked at all.