Ask HN: Picking your next project - How do you test ideas? | Hacker News Reader

Ask HN: Picking your next project - How do you test ideas?

8 pointshelen84200010 comments

Following on from the post yesterday "Always a starter, never a finisher" - http://news.ycombinator.com/item?id=3753680

I looked at my projects more closely. While some projects got left because of lack of cash or expertise, I'd say most got left behind because I found that they were fundamentally flawed to start with.

How I can focus on the better ideas by validating/eliminating early on?

The thing I've come to realise with this whole starter/finisher issue is that just because the infrastructure to start something is there, doesn't always mean I should. It's ok to have an idea and NOT implement it. As it results in more time, focus and energy for the ones that do make it.

What tests do you run on your ideas before you start?

What results do you have to get before you decide it's a good idea?

View on HN

I am prone to have zillions of ideas. Some are good, some are terrible. They all seem amazing the first time, but after a while they lose the shine and I leave them incomplete.

So, the first test for me is to write down everything I come up with. Then leave it and revise it again after a week/month. If it still sounds amazing, then I pay some more attention to it.

I think a good test is to see if you keep thinking back to the idea without having to be reminded. If it appears in conversation or the problem you are fixing is mentioned frequently - it's probably an idea that will have impact!

Yep, that is another good sign.

For the record, some ideas I thought a couple of years ago keep coming back from time to time. Sometimes it's just not the right moment or you're not on the right mindset, even if the idea is good. What I mean is, what matters after all is the execution, not the idea itself, but to do it properly there needs to be a planet alignment of sorts (the right person, in the right moment at the right time).

Create a landing page, put it up on forums like hackernews, reddit, quora and see what kind of conversion rate you get for the concept alone. Talk to people.

I've done that a few times and it got good results. Taking it to the next stage of finding a regular audience is hard though!

I really like Unbounce for building landing pages. Unfortunately Google refuse to run Adwords campaigns for them though, which is a shame.

How do they refuse to let you run adwords for them? What is their reason?

They suspended the Adwords account and said "We've determined that your site doesn't comply with our Landing page and Site Policies. Any ads promoting this site have been disapproved"

I could understand if it was just a lead gen site collecting e-mail addresses, but I was selling a gift product, with product photos, plenty of info and secure payment processing too. So who knows! I've completed the form to get them to re-review the site but I doubt they'll do anything.

I wrote about my process recently:

http://news.ycombinator.com/item?id=3721015

That post is very useful. Talking to customers is VERY scary so we put it off until the very last thing only to find out they don't want what we made!

I'd be interested in reading the follow-up to the article on ways of finding customers!

thanks - coming next week.

Reply on news.ycombinator.com