Live data from Hacker News

The $20 Starbucks Test

blog.snootymonkey.com

41–50 of 68 posts

Re: The $20 Starbucks Test

#41
Just ask people's opinion people! Its not that hard.

Find someone in the cafe that looks like they fit your target demographic and isn't too busy. Put a smile on your face, walk up to them and say

"Hi ... my name is xxxx, I was wondering if you could help me figure out some possible problems with this business idea I've got. And no ... I promise I'm not going to try to sell you anything"

Smile again. Rinse, repeat and refine as necessary.

Re: The $20 Starbucks Test

#42
post #8

It sounds like an okay idea other than the lying about worrying about your brother. I would much rather a person be honest than take up my time while keeping up a charade about a crazy brother, if someone did this I would probably just tell them to talk to their brother rather than me.

The point about using the "brother" is to provide the distance required to make the stranger feel like they can give truly honest feedback without hurting your feelings.

The parent post said the method was dishonest. You are arguing that it is useful. The two things are not mutually exclusive.

Re: The $20 Starbucks Test

#44
post #30
post #3

Earlier quoted context omitted.

It seems to me that the point is to avoid missing a glaringly obvious problem with a project that, while you may be blind to, a random stranger may not.

I got the point of it, I just think it's horrible execution. If you have a business idea, and you cannot find a way to solicit honest feedback from an ad-hoc consumer panel, then you should probably check yourself. I don't understand why you need the whole deceptive angle. Go to Starbucks and approach someone and say "hey... I've got this interesting idea for [a way to do X better/faster/cheaper]. I'll give you a $5…

I don't really think this is very clever, either, but unlike you I can't quite put my finger on why. Has either the poster or the originator of the idea really tried to measure the effectiveness of this technique?

Furthermore, why are people in a coffee shop sitting around without coffee, waiting to be bought a cheap drip drink (as you mention)?

I think the dishonesty makes some sense as people are nice and have a natural tendency to sugarcoat or outright lie, but I am entirely uncomfortable with it.

Re: The $20 Starbucks Test

#46
Wow, a lot of sensitive people here (and in saying so I expect my comment to be downvoted into Hades).

Like I mean, come on, he's just demonstrating a way he got new insights into his product. I think it's a great idea.

Take it with a grain of salt.

Re: The $20 Starbucks Test

#47
post #6

I think I'd be uncomfortable with this not because of approaching random strangers but rather approaching them with false pretenses. You're essentially fooling people into listening to your ill-formed business plan by lying to their face about non-existant siblings in non-existant danger. Whatever happened to being honest with people? "Hi. Can I buy you a cup of coffee in exchange for some feedback? Five minutes is a…

Approach it like they do in psychological studies: given their initial consent, deceive them if you like, but give a debriefing afterwards.

Re: The $20 Starbucks Test

#48
This was exactly the technique employed by a young drug effected person asking for money who was standing outside the local supermarket. "My mother is suddenly ill and I need $10 / $20 bucks to get to X." Except the guy doesn't tell you his startup idea.

Re: The $20 Starbucks Test

#49

Better idea: pay people on Mechanical Turk 10¢ each to bash your idea. You'll get 200 people's feedback instead of 5 from the Starbucks. If even only 1 out of every 30 takes the job seriously, you'll get a better bang for your buck.

I have used a mechanical turk tool to do usability testing, this type of qualitative feedback would probably not do very well on there.

People doing mechanical turk have one goal, to complete your task as fast as humanly possible, even when laid out fairly concrete steps you end up with a lot of feedback that isnt necessarily useful, either because they purposefully skipped it or there was communication issues or they just decided it wasnt time efficient.

I know mechanical turk has a feedback mechanism, but there is so much signal to noise that we found it much more useful to go and meet real people.

Re: The $20 Starbucks Test

#50

Ahhh, the joys of working in silicon valley where it can be assumed that every individual in Starbucks is going to comprehend you "brother's" start-up idea. Also, if I tried this method in my local Starbucks, I'd end up interviewing people just looking for a complimentary cup of coffee. Kind of reminds me of the Encyclopedia Britannica "get a free thesaurus" promo where you'd sit through an hour long sales pitch just…

Free thesauruses were fine but as a kid my father made us sit through 3 hours of timeshare crap on a family vacation in Hawaii so he could get 2 free cases of pineapples. We were quite sick of pineapple within a week.

Amazing what we will endure for something free.

Post reply on HN