Hacker News new | past | comments | ask | show | jobs | submit login
Ask HN: How do you validate demand for your side-project?
147 points by boris_v on Dec 26, 2021 | hide | past | favorite | 78 comments
There is an awesome thread here: https://news.ycombinator.com/item?id=29673707

But reading between the lines it's obvious that the major problem in most cases was lack of demand or inability to market the product.

So I'd be interested hear what and how you do early demand validation for your side project or idea, if any?

Share your story.




* Build the minimum needed features to solve the problem for your own use case

* Make it open source

* Give it away for free

It's not the golden rule that will work for everything. But that's what I did with Lunar, my app for controlling monitors (https://lunar.fyi) and it helped enormously with adoption and with seeing if a large enough user base could be created.

A lot of users offered to donate, others told me that I was stupid for not asking money for the app, others constantly asked for features on Github which helped me understand how the product could solve the problem better.

In the end, I prepared a large update with most of the features that were requested and made those features paid while keeping the old version free and open source.

99% of the users applauded this decision and were happy to pay (the other 1% sweared angrily through emails, but what can you do..)


I guess the 1% maybe weren't around for it have forgotten the shareware distribution method.


Great idea! Will adopt this framework for my side projects


Good idea here


Set yourself a discriminating target.

A friend’s kid asked me for some advice about his startup. It is a SAAS in the elder care space; he’d developed it for his grandmother and had a few friends using it as well.

He asked this same question (except it wasn’t a side project). I suggested, “If you can get 50 people whom you don’t already know to use this product for more than a month then you might have a winner.”

The three keys here are:

1 - more than a month meant they actually found it useful. For his SAAS it would imply they’d stick around after a starter period month.

2 - people he didn’t know: his friends and their parents might take extra effort to use it. He had to find people with actual need

3 - 50 is arbitrary but not terrible; a dozen or so doesn’t tell you a lot while 250 probably doesn’t tell you a lot more than 50 or 75

In the past six months he’s tweaked his product based on feedback from this effort and has gotten into the high 20s.


If you have to completely build the product before you validate that there is demand for it, what is the point of validating the demand


Note I mentioned that he had a fire friends of his using it. This was for the MVP of his free offering.


Maybe to justify or not investing in advertisements for it.


If I build something ready to use I'm gonna advertise it anyway just so that I can sleep at night.


I'm a big believer that you can make a lot of money starting a business and only ever selling to "friendlies" (your known network)

That said, business value goes up exponentially when it earns the First Dollar, a sale to someone with no existing connection to tue company or anyone in it. Very rarely is First Dollar sale so unique or edge-casey that there aren't dozens, hundreds, or even thousands of prospects that have the same problem and would buy for the same reasons.

TLDr - if you want to make some money start by selling to friendlies. If you want to validate a business that may be able to grow by leaps and bound, focus on the first dollar.


You can get stuck in the early adopter ghetto. Read Crossing the Chasm.


Still, that’s better than having no business at all. Crossing the chasm is a good problem to have


Sell one unit to someone. In fact, do it before you start coding. Coworkers, business contacts. "If I built X would you pay for it? Yes? Cool, what's a fair price?"

Get one of those first, then open your terminal. Aside from showing there is a non-zero market for your wares, it also proves that you are willing to do the hard part which is the selling, not the building.

If you won't do the hard part then you are wasting your time. So do the hard part once first.


According to "The Mom Test" that's actually something you should avoid doing. If you ask people they might answer "yes" because

- they're being nice to you

- they think in that very moment they would but in a real situation they wouldn't

In my experience, reducing "value risk", which OPs question is essentially about, has to happened incrementally. So build something small and then try to sell it to someone. Work with them to improve the product and then try to sell it more people and so on.

The veed.io blog has some nice articles with great insight into this topic: https://www.veed.io/blog/


I mean if they actually cut you a check for it and you are talking to people outside your close circle of friends, then this could be some part of the post mom test process.

Someone literally handing you money for your thing kinda invalidates a lot of other processes, theories, etc.


> If you ask people they might answer "yes" because > - they're being nice to you > - they think in that very moment they would but in a real situation they wouldn't

You forgot that one: - they would actually buy it

> In my experience, reducing "value risk", which OPs question is essentially about, has to happened incrementally.

That does seems to be the goal of the question... you reduce the risk. You can't be sure it's perfect, like every single step of validation, but you do reduce the risk. That's true because it's still possible they answer "no".

I think it's also extremely important because of the point of OP:

> it also proves that you are willing to do the hard part which is the selling

This is the biggest issue with a big majority of people on Hacker News. We are good a building, not selling. Which is also a big reason why it's pretty useful to find another founder that would fill theses lacks of experiences.


> This is the biggest issue with a big majority of people on Hacker News. We are good a building, not selling.

That’s actually an excellent point. Selling is an underrated form of art and essential for any successful venture.


My experience as a professional sales consultant for SaaS startups is that it is not the this cohort is "not good" at selling, it's that it has very little idea of what to actually do. Being a bad salesperson is a different ball of yarn to unspool than not being a salesperson at all.


What's your advice for typical a builder with no sales experience? Any courses or books you can recommend?


I didn't say to ask if someone would buy it. I said to actually sell it to someone. As in, collect the money or get a contract signed.


I did this with an idea I had. I thought of making a collaborative online video editor, like Google docs for video. First thing I did was talk to filmmakers. The idea was unanimously panned. I was immediately told it sounded like a terrible idea. Turns out filmmakers don’t want collaboration in post production, they want control. My tool would have taken that away from them. This saved me hundreds of hours of development time and let me work on other things instead.


And yet frame.io seems to be doing very well, it is popping up more and more places.


That is a good point, I haven’t been following film industry closely at all for the last 5 years but

> Adobe’s $1bn acquisition of Frame.io took many in the industry by surprise [1]

I am surprised. Perhaps this is a case of me not talking to enough people, or maybe people really don’t know what they want until you show it to them.

[1] https://www.ibc.org/trends/what-frameios-1bn-sale-says-about...


Frame.io is more of a collaborative review tool and collaborative asset & pipeline tool. It seems potentially quite different from your idea of a collaborative editing tool.

I also pondered starting building a company and tools for filmmaking and bailed out, and I’m glad I did. My background is CG & VFX, so not editing exactly, but talking to people in VFX helped my business partner and I see that VFX software is thin margins, super competitive, and not a good long term strategy in the US.

Anyway, the success of frame.io could still mean your takeaway was correct that editors don’t want many cooks in their personal kitchen, but also that your view of collaborative editing could have been widened into include a broader range of collaborative activity. In some sense, pro filmmaking is almost always highly collaborative already, not with multiple editors, but definitely with multiple people.

Another possibility is that frame.io doesn’t resonate with editors, but does resonate with many other people making films, ads, social media videos, movies, etc. It might not be the editors that are buying it. It also might not be experienced filmmakers buying it either; one problem with interviewing the experts is that they rarely understand the markets for non-experts…


Collaboration means different things in different fields. Did the filmmakers think - secure, easy to share, organized, and fast film editing software all provided in a browser with instant access to expensive render servers? Or did they think publicly editable film with easy access to a pool of “collaborators” to help with editing?

Many users of google docs don’t use private docs or the enterprise version.


veed.io is also doing well. Check out how they started on IndieHackers [0]. IMO their key insight was to go towards Canva type users: people who aren't professional video editors but those that just need to edit some video, make social media ads, etc. In other words, casual or prosumer users, not professional users.

[0]

https://www.indiehackers.com/product/veed

https://www.indiehackers.com/podcast/195-sabba-keynejad


Or you talked to different people. My impression is that frame.io is interesting for marketing agencies and freelancers collaborating (on an ad for instance) and not for people who make movies.


> And yet frame.io seems to be doing very well, it is popping up more and more places.

The key is knowing who your target customer is and solve for them [0]. Then, you are in a good place where you either fire ones who aren't [1] or change your solution (rather, start with a burning, hair-on-fire problem your target customer has, not solutions) [2].

Typically in a b2b setting, one starts out by addressing all needs (must-have) of their target customer as opposed to their wants (good-to-have). That is, other customers whose wants are NOT your target customers needs, you do not address them in the early stages of the product development [3]; and price / position [4] the product accordingly [5].

When changing the solution, figure out if there's another much bigger adjacent problem, that you weren't paying attention to [6], staring you down [7].

[0] https://hbr.org/2016/09/know-your-customers-jobs-to-be-done

[1] https://www.michaelseibel.com/blog/users-you-don-t-want

[2] https://www.ycombinator.com/library/8g-how-to-get-startup-id...

[3] https://apenwarr.ca/log/20211024

[4] https://hackernoon.com/obviously-awesome-a-product-positioni...

[5] https://en.wikipedia.org/wiki/Marketing_mix#Lauterborn's_4_C...

[6] http://paulgraham.com/schlep.html

[7] https://tomblomfield.com/post/33506878578/making-something-p...


That's not a sale. That's a question with absolutely zero investment or action asked of the participant.

Take an actual payment. Even if it's for vaporware (that you can eventually deliver on and that's properly messaged).


It's implied when they said sell it that you would take the actual payment.


Some things I've found helpful (I run a niche search engine and curation platform):

1) Find niche subreddits where people that are your target audience hang out, observe communities, help people out, participate in the discussion, if someone has a problem that your product can solve, then help them out in the comments and see what their feedback is

2) Learn a bit of SEO - you don't need to be an expert to fix some low-hanging fruits and get your product in front of people who might be using search engines to find a solution for their problem. Downside: SEO can take a few months to a year to kick-in, so if you have a very short time-window, it might not work immediately

3) Create a landing page and a sign-up list - see how many people sign up

4) Find bloggers and community members that write about your topic, cold email them telling them about your product. Be prepared for the response rate to be pretty low, but it can work.

5) LinkedIn - ok, ok I know I might get downvoted for this, but imho LinkedIn still has some of the best organic reach of any social platform, so you might as well use it to post about your product (esp. if it is a B2B one)

FYI - for consumer tech products, willingness to use something doesn't equate to willingness to pay for something, which you have to account for when doing research.


Agreed on #1. Once youre a legitimate member of a subreddit, post a beta version of your site. If it actually provides value and is useful (and you are not shilling), people should love it.

Here is an example: https://old.reddit.com/r/Denver/comments/qshpov/i_started_ma...

522 votes and 51 comments on a relatively small subreddit put it in near the top for a day or two. Idea validated.


Did any of that traffic become a paying customer?

My experience has been that reddit is great for validation. Traffic and kudos. Compliments and constructive critique. But almost no traffic from reddit resulted in people willing to sign up and pay money.

Now, admittedly, I sold my profitable side project a couple years back. Maybe the trends have changed - so I'm curious whether reddit users are now actually potential customers, or is it still just good for traffic?


> Once youre a legitimate member of a subreddit

This part is important. I’ve been caught out on Reddit sharing things I’ve built, expecting a “Show HN” type of response but shot down for promoting a product. Unlike here, most people on Reddit aren’t builders and mods tend to have a heavy hand for anything promotional (even if it’s free and you think it’s useful).


I think it depends on the audience. In my case, I built my side project to try and solve a real problem that my accountant had. It worked, and several other businesses — mostly accountancy firms but certainly not exclusively — have paid for premium subscriptions. Some businesses have been paying me every month for years.

I think my audience is the type that prefers it if you pick up the phone and have a conversation with them, or email them, or write them a letter. So that’s what I did. I had lots of conversations with lots of people. I tried to build relationships to make sales, and the validation part only came when the customer actually paid money.

I tried the build almost nothing and throw up a fake credit card form like some people suggest, but it didn’t work. I know it works for some people, but it didn’t work for me.


For a side project? I build it if I want to use it. There is a totally different goal for a business, but side projects are hobbies. They bring value to my own life regardless of what other people think of them. It is a great bonus if other people also get value of it and are willing to pay something, but there is nothing wrong with writing a project that simply helps me in my own life.


You're right. I meant a side project that you build for profit. I totally didn't imply it's bad to build something for your own use without making profits of it :) How would you formulate this question?


In that case, I wouldn't call it a side project at all. Whether it is your full-time job, or something you are building as a part-time business, the validation of the demand doesn't change -- Make a prototype and ask people if they'd pay for it. If they say yes, move on to an MVP. If not, ask for feedback and either try an updated prototype or try something else.


I listed the widget I made on Tindie and then promptly forgot about it for a few months. But when I got some initial purchase orders, I started paying attention again. When I got a repeat order from a customer, it had my complete attention. I called that customer, we talked about exactly what problem it was solving for them, and I learned a lot. However, it still took me another two to three years before I stopped everything else I was working on to focus on the widget full time. But now that widget has been my main focus (and source of income) for the past seven years.

Edit: One thing that helps is if your widget is a good-enough replacement or alternative for something that people are already buying for 10x the price of your widget. Even if your widget kinda sucks (it will! it does!), people will be so happy that they're saving 10x of their $$$ that they won't mind your widget's flaws... and they'll enthusiastically tell you which flaws matter and which ones don't, so you know how to improve it in the future.

tl;dr: 1) Pick a widget category where you can make a thing 10x cheaper than the competition. 2) Get a customer. 3) Talk to them. 4) Take notes.


Inspiring story. What is the barrier of entry for selling gadgets on Tindie? Do you need to get FCC certifications for example?


Pretty easy to create an account and make a new listing. My widgets are used as test equipment. There is a specific FCC certification exemption for testing tools. 47 CFR § 15.103 (c)


I think the main thing is that you have to actually talk to the people you want to sell to to find out what their problem is and why it's still a problem despite their next best alternative. Then you need to talk to them about the potential solution you have in mind to make sure you're not off the reservation entirely.

But the real moment is getting a specific person to actually give you money, hopefully before you put in too much work on the experiment. Promises or "that sounds cool" won't do it, you have to prove it to yourself by getting a person to give you real money for your thing, that's the only way to really know.


This is a really important aspect. B2B apps start out as semi consultant firms relatively often as it’s somewhat easier to simply promise you’ll solve a problem for a company for a certain amount of money than convince them that a new app will solve their problems.


I had a recent side project, a pdf generator of sorts that I made to solve my own problem.

I decided to try selling it, even though I was 100% confident that no one would buy it. During the first couple of days I’d also overpriced it, because I didn’t want anyone to buy it before I’d had the chance to fix a couple of bugs.

To my surprise (and frankly, shock) people were buying it. I was sure that requests for refunds would start rolling in immediately due to the aforementioned bugs, but they didn’t. I eventually lowered the price and even made it free, because the money was immaterial to me and I’d already learned my lesson from this project. (I also felt guilty for not setting up the infrastructure to ship updated versions to buyers, so I took solace in the fact they could get them for free on their own.)

I might go back to it, but the product is seasonal and doesn’t create recurring revenue so I’ll more likely be focusing on other things.

I’ve never had a project that hasn’t generated at least some excitement in the relevant community, so I might just have a knack for these things. Too bad I only just realized I could’ve been making some money from it!


Would you be willing to link to it?


I would, but unfortunately I've already posted about it here from a profile I wish to keep anonymous. (This one is practically my name.)


I feel like validation is less of an end-state and more of a process of increasing probability.

Like you might start with one data point ("hey, it's really annoying when I try to do X" or there are a couple posts in online forums from people describing problem X).

Then you get more data (see if more people have that problem, talk to the users on forum posts about their specific needs, find out where potential users hang out and how many people might have this problem).

Then maybe you build something simple to address the needs and see if people are interested / use it / pay for it (landing page or minimum valuable product).

Then you get feedback from the initial users about what they're really needing and iterate lots of times to get it right while figuring out where and how to market/distribute.

Throughout this process there's an increasing sense of confidence that you're on the right track (or not).

But also sometimes you build something just because and it turns out that lots of people like it (Minecraft). But the risk of wasting a lot of time and effort is a lot higher without continuous feedback.


The single necessary and sufficient condition for a business, says MIT’s Bill Aulet, is a paying customer.


First you have to make an important distinction, one that seems oddly difficult for many here on HN - you have to decide if you are earnestly committing time to this project *to make money*. If that's the case, now your time has value and you have to actually figure out if an audience exists and furthermore if that audience (which may or may not exist) is actually willing to pay for your solution. I chose the word solution because solutions implicitly derive value, a new note-taking markup or photo sharing app does not.

The second step is to field your solution and see if people are happy with the problem you've solved. Marketing is also a concept far too often underestimated or overlooked on HN. For instance, I wrote some software based on complaints from a very specific diy agriculture subreddit. Once I reached a bare minimum MVP I posted it on reddit, people were excited because a cool tech thing solved a problem they had. They shared with their non-internet friends and orders / emails started steadily flowing in. Unfortunately, the time/revenue ratio wasn't good enough and I basically broke even on the project.

That said, think about the problem deeply before spending valuable time "validating" what might be a solution nobody actually needs.


Start with a problem and not with a solution. Specially a good problem to solve. How do you know if it’s a good problem to solve?

Im my experience using Buffet’s approach to investing works well. Invest in good companies at a fair price and not in fair companies at a good price.

That means that the problem will usually have a higher cost to solve (time, effort, money).

Now, this is my opinion. Take it with a grain of salt. In the end, a focused approach to problem finding tends to work best.


Read The Mom Test for the only repeatable, reliable process. It is a fairly simple approach and if you execute properly it will be absolutely brutal on your ego.


I’m doing a surfing class booking platform.

I felt a need for something like this as a surfing student: I was booking classes via messaging/phone.

A few years go by and I forget about it, but recently I meet the owners of two surfing schools who tell me about the need for a such a booking system.

Trying to launch soon.

Hope this counts as validation.


Would you be interested in a call to share experience?

I'm currently developing a similar platform for music practice space (https://noisycamp.com).


Sure! Just sent you an email!


I wrote a handbook a little over a year ago about creating startups [0]. One of the most important points is ensuring that your product addresses a market gap. Demand cannot be created, it can only be awakened. In brief, one must: identify a market gap, contemplate a solution from the user's context and needs, create a prototype that demonstrates the core of the product (Minimum Viable Product), and do sales or pre-sales of that to complete the validation.

[0] https://www.amazon.com/Rocketshipping-accelerated-thinking-S...


Marketing/sales has been the easiest part of the work to me, and I'm a "tech guy". You simply tell people about it and they'll throw money at you.

My first business was selling coffee. I put a coffee machine on a metal table, put some pallets around it. Some dude came and asked for an espresso. He said it was too small and tasted nasty. I told him I'd refund that and give him a latte, which was 8 times bigger than an "espresso". He refused and gave me money anyway. He was later a daily customer after the first latte though. More people came and we sold 100 cups on the first day. Did we do marketing? Yes, we told the people staring at us that we'd give them a full refund if they didn't like it. Nobody took the refund.

My first real startup was a recipe app with an ingredients store. People just went to the store and clicked "add to cart". I did not expect them to. We didn't have a dashboard for viewing orders (nor a cart, the purchase button just took their phone number and address). I hacked something that put together a cart on the dashboard, then I'd contact them on WhatsApp and give my bank account details (we didn't have a payment gateway either).

The hard part is often the product. If you have bad lattes, nobody's going to order another. If you don't have recipes for flaxseed flour, not that many people are going to order the flaxseed flour.

I'd say don't listen too hard to entrepreneurship gurus who tell you to "validate before writing a single line of code". Often the code is part of the validation. I tried to run ads for flaxseed flour and nobody bought it. There's a niche for it, and that niche was overadvertised by other sellers. Writing the code exposed us to a marketing flank we never knew existed, and the app was eventually acquired as a diet food marketing channel, because the CAC was lower than Facebook ads.

You can build something of value and give that for free. That's actually a little easier than trying to go the validation route. From there, you can use the feedback to pivot into something of higher value or is more monetizable (e.g. DLCs). Metallica didn't think they'd make money off music, so they started with t-shirts.


Fake door testing. Create a marketing page and see the reaction you get.


This. Have done this for my own and in some cases friends' potential startups: a fake door landing page and a couple hours of posting to discussion groups (don't spam, do it where communities allow) or a small Google ads advertising budget.

There's been a couple cases where that simple amount of initiative plus about a half day of work, the resulting feedback was enough of a push to overcome hesitations my friends' had in starting a company.


You’ll also need some kind of SEO to drive traffic to the page.

And if your project is in a niche area, then it probably not help at all.

I setup a landing page and in the last 6 months I’ve had less that 100 visits and no interactions with any of the call to actions; even with paid advertisements on linked in and google.


I thought about doing that, but it felt like false advertising and I didn't want to (a) piss people off, (b) underwhelm them, or (c) commit to getting the project ready for use by any specific date. Am I overthinking it?


Set an expectation that you plan to launch by some date and collect email id. If they are willing to share their email then that shows they are really interested.


I’d also recommend putting a Kickstarter together for your idea even if it’s unrealistic to launch one.


There's a simple way to ensure that you're making something that people want - just solve an existing problem instead of finding a new one.

Search for your problem on google. If the top results are blogspam with ads for monetization, that's a bad sign. Are there existing products in the niche and are they doing well? Is there a lot of search volume for related keywords? Do you have a better or unique understanding of the niche that others lack?


Do a short exercise about your provisional persona: needs, behaviors, goals, facts. Then ask yourself where you'd find that person and post a smoke test/fake door test landing page in those spaces (for example if your demographic is 50-somethings, women, disposable income, parents, etc. you may look at Facebook and find some groups to distribute your landing page).

Look for microtraction, and then experiment as much as possible.


I'm not sure what their rules are for a software project, but it kind of seems like this is what Kickstarter is far. If you meet your funding goal then you've validated the idea with people already putting $ on the table.


From the B2B side, demand is validated when multiple (>10) organizations are willing to schedule time (30 minutes) to watch your demo. That requires scheduling and team efforts. That’s when you know the pain is real.


Don't overcomplicate things. Ideally you or someone you know is willing to pay for said product. Then you find other people like you or someone you know who are also willing to pay.


Simple for me: how easy is it to find your first actual ( and paying) customers compared to building a website.

Fyi: yes, some experience building websites for others is required.


I’ve heard good things about this book: http://momtestbook.com/


I bought this book after it came up in a previous thread on HN. It really is good. Short, simple, and direct.


Can someone recommend common places to list new side projects?

In my case small SAAS app for business folk.

Surely there’s spots on the web for this?


I would try two things:

1) Where do people that have use for your app hang-out or post about the problem that your app solves? Are there any online forums or subreddits or facebook groups where your target audience is? If so, I would join these communities and when someone has a problem that your product solves, I'd help them in the comments and point to your product. Many groups are (for a good reason) hostile to adposts, so your best bet would be to spend time and engage with the community as opposed to going for direct adposts immediately.

2) Post on LinkedIn (consistently) using keywords / hashtags that target your customer base. Use infographics or tutorials of your product to showcase how it solves a problem. I know LinkedIn can be a pain, but imho it has one of the best organic reaches of any big social media platform and it's free (apart from the time it takes to write and post).


What do you hope to get out of listing it?

If you hope to get customers - You're thinking about it wrong. First, the typical buyer doesn't trawl a "SaaS Apps" list site looking for services they need. Second, even when you list on a site like Product Hunt, the chances are low that any given visitor that sees your site the day it's promoted also needs your service. Instead, think about what search terms someone might use on Google when they have (or are about to have) the problem your SaaS app solves. Try and write the best answer on the web for those searches.

If you hope to get feedback - Again, probably thinking about it wrong. I'd try finding people who A.) You don't know and B.) Are uniquely qualified to give feedback on your service, and get in touch with them, try and offer something of value to them in exchange for feedback.


Just do it, ask for money and see what happens. We just did it and it worked, there was no previous research.


I don’t. Money is not a primary motivator for my side-projects so product-market fit is not a concern.


Build a tool that does early demand validation. There is a certifiable demand.


I mainly do 2 things to validate my project: 1. find relevant threads on reddit or other communities. If people are asking the same question again and again, it's a good signal.

2. find if any products exist to solve the problem. If there are a few of them, try to guesstimate the market size. Mostly likely, there is a market for that if the number is ok. If there are many existing players, there is no need to validate because obviously there are many competitors in the field.


Ask for payment. People will tell you it's a good idea and then balk at paying even a nominal amount.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: