Ask HN: How do you validate your startup idea when you've no distribution? | Hacker News Reader

Ask HN: How do you validate your startup idea when you've no distribution?

89 pointsdotvignesh70 comments

Hello HN,

Wondering how people with almost no distribution channel manage to validate their ideas!

I've built and published many apps throughout school without validating the idea, but they received 20K+ installs in a short span.

After reading tons of books and articles, I realized people validate their ideas before spending too much time on them.

So I decided to build a landing page with a waitlist for my product and posted it on Twitter and Instagram.

Though a lot of people I spoke to sounded "excited", I got very few signups. (They're not my target audience, so I didn't ask them if they'd pay for such a product)

Now I'm wondering if my idea is bad or if my distribution is awful.

What am I doing wrong?

(P.S I'm working on a creator SaaS product)

View on HN

I've been around the horn a few times on this. Anything short of a commitment to pay for the product (up to and including prepayment) is just noise. It's not so much that people lie about what they say they would do, but that it's too easy for you to lie to yourself about the signal you're actually getting.

Here's a good example:

> Though a lot of people I spoke to sounded "excited", I got very few signups. (They're not my target audience, so I didn't ask them if they'd pay for such a product)

Those people weren't in your target audience, and you didn't ask them to pay, but I presume you take it as a positive sign that this particular audience was "excited." But it's not. It's not anything since they're not even potential customers. That's how we subtly deceive ourselves.

Talk directly to people in your target audience (contact on social, then move convo to email, then move convo to telephone) and try to get them to put some skin in the game. One popular tactic is to offer a free feature request for a refundable prepayment.

Yup, one friend who had a pre-internet sw biz with many different smallish products (think utilities) told me an interesting story.

He started out describing a proposed product and asking if people would be interested and buy it. "YES", no question they liked the idea and would pay money for it. So he went and built it. No takers, lots of "maybe later"s.

After a few rounds of that losing game, he started asking not what they wanted, but about specific pain points - 'What is most inconvenient about this?', 'does this task consume too much of your time?', 'what is the biggest time-waster?', etc.. When he listened and built products addressing the complaints, he'd make sales on the first call, even tho no one ever actually told him they were interested.

I'd suggest for OP to design a question list around things solved by the proposed product, and related work. Do not tell them that you are working on a product. Talk to prospects and see if those are true pain points - and try to get them to say they're just fine - avoid politeness & confirmation bias (i.e., you really don't want to fall for a few ppl being nice, agreeing that it's a problem, then building something they don't care about}.

So, if you try to convince them that there is no problem, and they insist that there are no workarounds, they've searched and it's a PITA that costs them money, then you are on to something.

Edit: typos

Will give this a shot, thanks!

Good luck!

To be honest if anyone outside of my work showed me their product idea I'd at least pretend to be 'excited' about it just to be polite.

In retrospect I’ve semi-unknowingly done this, and I regret it.

The parent comment is right. There are two useful outcomes - they are willing to take their credit card out, or they are willing to give you direct feedback. Everything else is noise.

What "excited" really looks like: https://www.youtube.com/watch?v=OnB1TgxgwEA

Agreed! I've been trying to reach out to content creators, but the problem is they mostly ignore since they get so many such requests. That's why I wanted to know how you get over it when your distribution is bad. Thanks for the suggestions!

I'm a founder of a SaaS company and can provide some insight that I am surprised is not really written clearly in this thread yet!

There's three types of innovation.. 1. What you think people want 2. What they tell you they want 3. What they actually need

Most startups die because their product strategy is #1, the chances your view of the world is going to be adopted by people with many different circumstances is very low. The few think, lets go talk to people, but then die due to misdirection.. customers aren't PM's, they dont know how to solve problems but they know what their problems are.

so you must figure out what do people actually need..

You have two ways (well technically 3) to validate your idea 1. Quantitative 2. Qualitative 3. Resegment an existing market (questionable if you can skip traditional validation but your chances are better here than 1 or 2 above)

If you have funding, you can do quantitative. Build stuff and get it in front of your audience with paid acquisition/marketing.

#2 is the best if you want to be a scientist about things. Learn Jobs to be Done, start conducting interviews, find the problem folks have.. figure out if its a priority and a pattern.. there's lots of good JTBD literature out there, read it all.

#3 - segment an existing category. Take form building for example, you can build a form builder with limited features for a niche audience.. higher ed? (Qualtrics anyone?)

Hopefully this just scratched the tip of the iceberg for you and gives you some direction!

Just to clarify, qualitative research is less costly per dollar but does cost time. Check out the book "Nail it then scale it".

There's also #4 which is "What I definitely actually need". Tools that you build for other people have a market size of [0, everyone). Tools you build to scratch an itch and actually use have a market size of [1, everyone). That may not seem like a big difference but the scale is exponential, there's as much space between [0, 1) as there is between [1, Infinity).

The next step, and one where people frequently fail, is to introspect and to ask why is it I find PMF with my tool, to what degree are those reasons personal vs universal and what are the minimal tweaks I need to make to expand the adoption of the tool by the next exponential factor? A trivial example would be like, a tool that you build for yourself doesn't need to have authentication but a public tool does. Various things that you hardcoded in because you built a tool that fit you like a glove instead need configurators and other things to fit other people as well.

The most common form of failure with this approach is that people who build their own tools are very high variance from the population so you're likely to build something too niche or too obscure to reach sustainability but that's in many ways a better problem to have than a tool with zero adoption.

As a dog food lover (product dev, not Purina), I found this very insightful. I may frame it on my wall.

Landing pages and waitlists are becoming too common. It's very common to come across a great product on Product Hunt, only to find that it's just "validation" and doesn't even exist. Waitlists are a signal of vaporware and many just avoid them.

True, maybe I was allured by the Dropbox and Robinhood MVP story to attempt this. Is launching a functional MVP a better choice?

A MVP is just something that disproves the weakest assumption on your list.

If you're trying to prove a value proposition, I find that a survey works better than a landing page. PowerPoint decks work too. You can do something similar without being deceptive. Or you can just go to a subreddit of your target market and see what's popular or what people are complaining about.

Makes sense, will give it a shot. Thanks for the suggestion!

What worked years ago May no longer be as effective now. Just like banner ads were Wow in the early days, but now users tune them out

MVPs aren't a bad idea but you have to make sure you're doing then right. Have a target customer segment, have a clear objective, build the smallest version of your end goal that actually provides value but won't break the bank when you realize how much of it was actually wrong. Then listen to the complaints and feedback, cut out the parts that don't solve for a customer need and deliver another one. Rinse and repeat a few times and then put real effort into the things that you've identified as providing customer value. Don't feel bad about getting wrong, as long as you weren't excessively wasteful along the way.

Spend the money on the validation!

Recipe to spend $1000s marketing dollars vs $10,000s of development hours to validate ->

(1) Think hard about exactly who your target customer is and how your product would be received.

(2) Use google ad planner or FB's audience tool to see how accessible that audience is online, along with projected CPC costs

(3) If you are super budget constrained scour the internet for free adwords / fb advertising credits eg - https://www.diydigitalstrategy.com/150-google-ads-coupon/ - many online marketing classes will come with a $100 - 150 coupon

(4) Build a landing page framework WITH pricing on the landing page (or a pricing range eg $XX a seat with team discounts). Both FB & Goog are anti "waiting lists" so you need to think about how to get around this. Consultative sales, eg a "Somebody from our sales team will be in touch" doesn't break their TOS as long as you do a good job following up with the customers

(5) Spend the money on the validation and get the added bonus of talking to some very early preliminary customers that are NOT in your orbit

You can do it for $10s by using old school sales. Cold outreach still works if you're solving a problem.

This is tough in B2C though.

any ideas for B2C apps?

Ah! It was something I wanted to try, will take this as a framework and try it out. Thanks!

Landing pages with a waitlist can be useful tools.

Have you read The Mom Test? Here's a summary: https://www.slideshare.net/xamde/summary-of-the-mom-test

I recommend reading the book.

I haven't read this book yet, just went through the summary in the link. Seems to be useful though, thanks for sharing!

It's a quick read - you can bang it out in two hours, and it's totally worth doing!

Came here to recommend The Mom Test

Product market fit isn’t just building a cool product. It’s when that product meets an audience that values and pays for it.

Distribution IS part of a startup idea. It’s an essential ingredient. Don’t think about idea and distribution as separate things. If you can’t find your target audience to validate your idea with, you have an incomplete startup idea, by definition!

If you are unsure how to find your target audience today, building your product will not fix this problem. Speaking to real users should be a higher priority than building or planning features.

For advice on speaking to users I only recommend one short book - “The Mom Test”. In that book you will learn why “excitement” is a very misleading signal. There is a good reason why YC partners constantly recommend it.

Whoa! So many people in this thread have recommended this book. I'll read it right away. Thanks!

I once tried the following approach and it saved me a ton of development time. In reality, it validated that there was not a good pattern for a problem to be solved in the space I was looking at.

Figure out who your target audience is first. You might have to talk to a lot of people. Get good at listening and asking open ended questions ( see book Mom Test )

Once you have a rough idea who your audience is, try to find about 30 people in that audience to talk to face to face. You can use a script like in the book Running Lean. You want to keep good notes on each conversation. Build a table/matrix of all things you uncover across these 30 or so conversations. If you can see a clear pattern in the problems a majority of the people have, that could be the idea you build a product for.

This strikes me as recipe for ending up in a startup you hate.

Why not build something you would pay for? It would give you more drive to deeply care about making the product happen.

That could be true if you have no affinity for the audience you're targeting.

If you choose an audience you are part of then you are less likely to end up hating the end product.

Building to scratch your own itch carries the risk of you inadvertently building for an n of 1.

I'm targeting content creators. Trying to solve the media deplatforming issue. The problem reaching out to my audience regarding this is that majority don't realize all their audiences are the ones they've rented from the social media platform they're on. So I end up telling them the issue and how my solution is what they need for long-term sustainability. I'm not sure if that's the right way to validate my idea.

I'm not a fan of validating ideas. The problem is that modern people often like things which are not good for them (e.g. processed foods, drugs, social media, bureaucracy, etc...). The product might make them less efficient or harm them, but people still like it for some reason.

Customers are often irrational, especially with the current heavily manipulated monetary system. I only want to work on products that are beneficial for users (in a meaningful, long term way), even if people might not understand the benefits today (and it might not sell at all).

Bootstrapping (unlike VC funding) allows you to wait out the insanity instead of forcing you to profit from it. If you have a good product, people will come to their senses sooner or later - You will get steady, linear growth but it will be solid growth.

Most of the time, exponential growth is fickle. Users got overexcited, they made an impulsive decision; the next phase is disillusionment.

I built a $100M SaaS. Do you want this more than anything? Are you willing to do do whatever it takes and go 1-2 years without a salary?

If you answered yes then you just need to leap and iterate on the user feedback loop.

My first mvp compared to what I'm selling today is completely different. I pivoted 20 times at least.

So the name of the game is to just start building and not be overly concerned with making mistakes. Find what doesn't work and move toward what does. Mimic nature by constant evolution.

> If you answered yes then you just need to leap and iterate on the user feedback loop.

The question was how to get the users and their feedback in the first place.

If it's for creators, you have a built in sales list. Go to all of your potential customers and send an IG, FB, Twitter, etc. and ask to talk to them. You probably wont get responses from people with 2M followers but thats not your target audience, they already have a business built around them.

Once you do this, the creators will spread the product for you if it's good. Being a creator is all about collaboration so they are used to this in terms of the products they use and sharing those products with their peers (and communities)

In terms of distribution, this is a relatively easy one to solve. The classic: Figure out who your customer is and talk to them, you have it easy in that your customers are TRYING to be visible.

source: I'm an athlete with > 100k IG followers

How often do people reach out to you like this?

(1) Validation via MVP for SaaS is very different to, say, a complex hardware product. (2) What people say they will do and what they will actually do are two different things. (3) Take in to account the potential for market shift (new competitors, technology change) during your delivery period and have risk mitigation strategies (eg. pivot/sale/exit). (4) Scratch a common itch. (5) Validate by proxy - see similar ventures gain traction in allied markets that demonstrate clients exist and investors have interest.

Is it always a good sign when investors invest or show interest in a product? Asking this because I got shortlisted for an interview in a fellowship program. (They invest in my startup if I get selected, they shortlisted around 20 applicants)

VCs are usually wrong. Angels are probably wrong even more. However, nobody at all showing interest might not be a great sign, either. Secondly and perhaps more importantly: don't confuse accelerator/incubator programs with investors per-se. The former group tend to be very focused on a specific period such as feasibility or pre-seed and in a poor case are actually consciously making their money out of systemically reselling the x% of the companies they bought in to for tea-and-biscuits at a higher valuation with high frequency and rapid turnaround. They can be very clearly misaligned with adding long-term value, and just want to push for a fast flip. That's why some will pressure for MVP/POC, stupid timeframes and in-person presentations even when it makes little sense technically or commercially for the venture itself to do so. It can be real rabbit/hat smoke-and-mirrors stuff. That's private equity in general.

Thanks for the heads-up!

As other people have mentioned, landing pages + waitlists aren't the best nowadays. I do think a better way to get traction is to go and find 1-2 customers who will work closely with you to build a product. You mention this is for creators, so if possible, leverage their follower base to drive viral growth.

This sounds like something that'll work for me, thanks!

Good luck!

Why does everybody here think defrauding customers with a bait-and-switch (up to even taking deposits for non-existing product) is a good way to start a business?

That's the modern way to do business, rent out hotel rooms without a license, start a cab business without a license. Pay less than minimum wage. Lie to your customers. The fact that you're undermining the fabric of society is just a negative externality, it doesn't go on the balance sheet.

"Now I'm wondering if my idea is bad or if my distribution is awful."

This is a false dichotomy, I think. Trying to grow is an essential part of how you evolve the idea and discover what the "growable idea" really is. This is one of the core things I took away from this PG classic: http://www.paulgraham.com/growth.html

Maybe I should just launch first and iterate like PG says!

I wonder about this as someone working a small indie software business rather than a startup.

Most of the replies here seem applicable only to products offering new capabilities, as opposed to products intending to be better implementations in existing categories.

The problem I see is I can say, "My product is like X except it's more usable/efficient, less buggy, and looks nicer," —but they'd have to take each of those claims purely on faith, so it's not very useful.

Thinking about it more, maybe it's impossible by definition to pre-validate this kind of product: your offering is strictly in the implementation (not some abstract new feature concept), so the validation has to take place with respect to concrete details of the product's realized form.

Maybe the best you can do is take this into account when devising strategy around the ordering of features to develop: e.g. if your claims are on usability or performance, focus on that first.

(Then again, seems dubious since most all of the example qualities I mentioned are sort of emergent from the entire product being in place, can't really develop them in isolation...)

Any thoughts?

You don't want to spend too much time, but you have to spend some amount of time and effort to validate it. I think it scales with the ambition of your idea, but something like 3 months of full time work on well-executed, but non-groundbreaking service sounds pretty reasonable.

The problem with landing pages, waitlists, etc is that they'll give you very little - until people are using your product everything they say is suspect. They need to be using your product, be vested in it, care about it, have opinions about it, etc. Landing pages will only get you a "huh, that's interesting, I might try it out"

To validate it, find several dozen people that will use and love it and give you real feedback. This is a highly manual process and it should be.

Thanks for the suggestion!

I've only ever worked on something that solved a personal problem. I just go ahead and build something rudimentary/MVP that solves my problem. That MVP could still include manual bits, be sitting on a free subdomain (subdomains from Google App Engine), not have a proper logo (just use text for the branding), etc.

Once I get rudimentary version, I can then put it out there in public to validate it. Maybe do a Show HN or do a 'low budget' Ad on Facebook/Twitter. This is what I did for my current project, https://nocommandline.com (I did a Show HN).

I feel that software needs a 'trial' version to be able to gauge whether people will actually use it or not.

I used to do exactly that! Too many startup validation stories messed around with the way I do things, not sure if it's right or not.

Here is what i am doing to validate ideas..

I launch them on appsumo marketplace which helps me get in front of active buyers

i have done this with blurweb.app(https://blurweb.app) click2contact.app(https://click2contact.app) sendsimple.app(https://www.sendsimple.app)

and will be doing the same for the others which you can see here https://twitter.com/indianappguy/status/1429157263353126912

I built a SaaS product, but it's too early to say if it's a valid idea yet. What I did do was find 5 companies with the same problem, and gave them free, very early access with regular calls for feedback. Interestingly, several were very precise in their requirements, and others were open to my approach and ideas. It's important not to try and implement every suggestion you're given if it doesn't feel 'right', otherwise you risk creating a product that's inflexible for everyone except the customer who asked for it. Try and find a sample of people who will give you the time of day.

Sounds good. Thanks for the suggestion!

Write out your idea. You can start by validating your idea by making sure your answering the right questions. Then you can reach out to people, you don't necessarily need a landing page but a demo probably helps.

I typed out my idea on a Notion page, did market research, existing competitors, and all that. It looked quite promising to me. I included a small demo on my landing page as well. Maybe I should try reaching out to people with a better functional prototype. Thanks for the suggestion!

https://twitter.com/justinkan/status/1059989657218248704?lan...

Whoa! That's where my question arose from, lol

You may have to network your way into finding customers. Who do you know who knows creators? Is there a way you can get in front of creators so that you can test your assumptions in front of them?

Trying to get that done now. Thanks for the suggestion!

What's the idea?

I think in general if you're honest with yourself, you have a decent chance of evaluating your idea's desirability. Go with the YC mantra of "Make something people want". If you are scratching your own itch, that could be a good way to start.

"Creators" are a fickle bunch so not always the best audience.

I'm building a tool that lets creators build a website with no code, engage with their audience, monetize content, and own their data. By "Own their data" I mean creators can export all their data and self-host it on their servers. Trying to act as a bridge between Web2 and Web3 technologies by getting non-tech savvy creators on board now and gradually move all their data (like articles) on a blockchain.

Hey Vignesh! My cofounder and I ran into this problem at our last startup. Mind shooting me an email (email is in profile)? Would love to share some of what [somewhat] worked for us.

For full disclosure: I’m trying to build a better way to validate ideas, and I think I too can learn a tremendous amount from talking to you.

Mail on the way!

There's no substitute for doing the legwork. You gotta go find people where they are and talk to them.

Have you read Pat Flynn’s “Will It Fly”?

Will give it a read, thanks for the suggestion!

google or fb ads

Reply on news.ycombinator.com