Hacker News new | comments | show | ask | jobs | submit login
The $20 Starbucks Test
13 points by ced83fra 7 days ago | hide | past | web | 16 comments | favorite
The "$20 Starbucks Test" is the best test of a startup idea. You don't even need a prototype. You just need to get to your local coffee, walk up to strangers and talk to them about your startup idea pretending it's your brother's. Then collect feedback.

Great... great ?! However good this idea might be, some might have some problem to strike up a conversion with a random stranger.

How would you (have you) overcome this ? What pushed you to do it ? Do you have any idea of a startup-idea-test that does not involve coding ?

Any advice welcome.

One gotcha with the Starbucks approach is that it's very dependent on how much the audience for your startup overlaps the audience for Starbucks. This is going to heavily depend on both your startup and your area - asking random people at my local Midwestern coffee shop about a new AP database-as-a-service is not going to go well, but it might at some spots in SF.

One approach I've seen used to counteract that issue is the old "landing page without a product", where you use targeted ads to put your landing page in front of people who might be interested.

There's a second gotcha, common to both approaches: there's a significant leap between "gets excited about the product / signs up for the mailing list" and "actually pays for the thing".

The story patio11/Mckenzie tells about Appointment Reminder is about half way down the page here (search page for 'chicago'):


If you're launching a startup and you can't approach and talk to someone you're going to have a hard time running a business.

This x1000

If you can't talk to people but you want to start a business you've got 3 options;

1. Work with someone with less spaghetti in their pockets 2. Work out a way of removing to spaghetti from your pockets 3. Don't start a business

How about the $5 DMV test? There's a bunch more people there, and they're all bored stiff!

I have talked to literally hundreds of people off the street, coffee shops, etc. about my startup ideas. In all honesty, most of them were a waste of time.

Now, I take a different approach. I talk to my friend's friend's friends until I find the right customers to talk to. I contact my friends and ask who they know and then I ask who they know and so on. Its pretty easy once you get going. And a warm intro is an easy way to get the time you need.

Other resources are LinkedIn, Twitter, Reddit, Quora, etc. Find someone on there that has an interest in the idea you are working on. After you have a short interview ask who they know. Once you get the referrals flowing its very easy to find enough people to talk to.

A great read on this topic is Talking to Humans. A couple of takeaways: 1. Approach talking to others as a way to question your assumptions. 2. Ask for stories. People do a better job telling a specific story about their life, rather than answering hypothetical questions. 3. Look for clues, workarounds, passions, and pains.

- https://www.amazon.com/Talking-Humans-Success-understanding-...

The "$20 Starbucks Test" is the best test of a startup idea.

Is that best in a quantifiable sense, or best as in "the one I like most"?

It is the best in that you will have the maximum amount of feedback/suggestions/critics. It helps you discuss with your potential users and helps you refine your ideas. It's one aspect people critically need at the beginning. It's one aspect people do not have with the anything that uses ads on internet (landscape+ads to see if some facets of your idea are successful, or prototype+ads).

It is the best in that you will have the maximum amount of feedback/suggestions/critics.

I can't imagine that's true. In a couple of hours I reckon you could talk to maybe 20 people in a busy Starbucks. Spending a couple of hours writing a great Show HN or Producthunt post could get you more feedback from a wider audience without costing you $20.

It helps you discuss with your potential users and helps you refine your ideas. It's one aspect people critically need at the beginning.

Strangers in a coffee shop aren't "potential users", even if they think they are (nor are HN readers for that matter). Any randomly selected group is effectively useless for gathering data - there's too much noise and not enough signal.

If you want good feedback you need to approach people you know are actual potential customers. You can't rely on luck. Go to places where they gather - user groups, meet ups, websites, etc.

I just started reading The Mom Test. I highly recommend it. Your questions, even pretending to be about your brothers idea, are still introducing a bias. You will get bad data. The book shows what type of questions you should be asking to avoid these issues.

Aaron Patzer used a similar approach for Mint, but asked people what they thought of what eventually became "bank level security" on his website. I think that's a better approach for the random stranger feedback.

If you want to know if there's demand for your startup, you could simply build a website and pitch it to people, however, you won't truly know if there's demand until you build something that provides value and ask people for money. (Just like DoorDash & Flexport did) Even then you'll face distribution problems but that's another story :)

$20 could buy you a lot of honest opinions on Mechanical Turk. I've been considering this route lately. I believe it even lets you solicit work from people based on specific demographic data.

That's a waste of time and $20. You validate your startup idea by trying to sell it and seeing if someone actually sends you money.

Sometimes people will give you positive feedback and blow smoke up your ass saying it's a great idea, but the real test is if you asked them for $XX right now for the thing, would they give you the money?

Based on your description, I don't see where the "$20" comes in. Are you buying people coffee for their opinions?

Sorry to be negative, but spending more money on a scientific survey would give you believable data to work with. This sounds like a waste of both time and money.

This is what https://www.usertesting.com/ is for.

Unless your startup is about coffee i highly doubt it. Create a landing page and spend the $20 on SEO

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact