Trouble is ... my ideas suck. Either I have a really good idea that requires inventing new technology, or I have a niche idea that I can't fathom how to monetize or market. I'll take your points about being boring to heart and hope something finally springs to mind under those constraints. But as of a month or two ago I already realized that I needed to tone my ideas down, and after all that time with it in the back of my mind ... I still have nothing. Frustrating.
A phone app that makes scanning documents stupid easy by using machine learning to identify when a document is on screen and automatically take a picture, and then use more machine learning to convert the picture into a cleaned up image. The concept to train the machine models is pretty easy; just print out a known document, take a bunch of pictures, and train the model to convert from the pictures to the original document. But it would likely require massive kernels to get enough context to do the necessary 3D transformations. Could probably reduce the strain on the algorithm by using traditional picture->document algorithms and then use the model to clean it up further. Either way, it's new tech that would require a lot of work.
Vocal isolator for extracting vocals from music, using a machine learning model. These tools kind of already exist, but they don't use machine learning, usually require user interaction to tweak knobs, and don't work particularly well. My bet is that a 1D conv network can do great, and generating examples for training is super easy. But, again, new tech, and as easy as machine learning sounds it usually devolves into long nights of debugging, tweaking, and banging your head trying to figure out why changing one hyper-parameter ever so slightly completely changes your results.
A hardware password manager. Like KeePass/LastPass/etc but as an actual hardware device, possibly a wearable/watch. The idea is that a hardware password manager provides more security than a software password manager, by virtue of not being susceptible to malware (though yes passwords can still be stolen 1 by 1 if used on an infected machine), and most importantly works with phones/tablets over bluetooth. Requires developing hardware, productization, etc. Not a big market; nobody cares about security.
I have a quiz website that I built for my wife, who's a teacher. It allows here to give online fill-in-the-blank type quizzes to her classes, with a set amount of time to take the quiz, and a set amount of time that the quiz is available. She's happy with it; I made it because she had something similar with Edmodo but the school dropped Edmodo. I can't imagine such a simple website is worth much though, and it's currently only designed for her and a bit rough around the edges. It'd take quite a lot of UI updates and a bit of backend tweaking to make it usable by the general public.
A couple game ideas, mostly geared towards programmers. Pretty niche, and games take immense effort versus profit.
A service to provide email to Bitcoin addresses. You authenticate yourself using your Bitcoin wallet's signature mechanism, and then can receive email like at e.g. 1p7H5w1LfgLT6tat951e85dFXEDQjNp8L@example.com. I got this mostly working, but rendering emails is super hard, who wants to use yet another mail client, how do I monetize, how do I fight spam, etc, etc. Lots of security and anonymity angles to this, but security is hard, so again lots of work.
NSFW: Machine learning based porn assistant. You thumb up/down images, it learns what you like, and finds more of it. Keeps learning as you thumb up/down its results. Classic machine learning model. Requires new tech, and it probably requires training a model per user (super expensive), or some kind of transfer learning.
Lots more beyond that, all very pie-in-the-sky, new tech related, or super niche.
The point being: most practical machine learning I've seen has involved a lot of effort chasing 75-90% precision, which means that a lot of commonly considered problems can only be aided by machine learning for human-assisted or human-curated tasks, or cases where precision can be sacrificed and benefit still derived.
If you haven't picked one idea and run it all the way through to a working solution, you have the choice of actually picking one and doing it, or not attributing more value to machine learning than it actually deserves.
It seems to me that innovative uses of machine learning are more around finding a user interface or user experience that compensates for inaccuracy than it is the sophistication of the algorithm itself.
I think that for a lot of people who might try small businesses, the "use of surplus energy" is intermingled. Maybe you want to make some cool piece of software (because you just want to); and you want to use it to do something creative (because fulfillment); and you want it to also be a business (because you want more money).
In that case it's much harder to focus on the "boring" part. (I am assuming of course that there exist things that are "boring" in the business sense but still allow for some satisfying creativity and/or fun hacking. For instance CoderPad gives you plenty of room to do cool programming things, if doing those will make you happy. You just might have to book them on the lifestyle/happiness ledger. The bingo-cards thing, maybe not so much.)
Edit: damn, I love your presenting style. Keep it up man!
For instance, the way you recommend a hyper-specific style of elevator pitching I think has basically no impact on success. Success with pitching is (according to my dumbass) way more about identifying who your actual audience is and why they care than about getting this perfect "pitch" correct. The discovery process around that is much harder and ambiguous than defining a pitch on paper.
Also, you have really specific stuff in there like "blast your contact list". I mean, okay, I think that might work for some people in some industries, but I don't think it's great advice for the median new founder. Neither is doing a CAC/LTV for a business when you haven't even figured out potential distribution much use, either.
Anyway, my longwinded point is that this is kind of culty. What makes cults what they are? I think it boils down to their unshakable conviction that their specific rituals are correct and their motivation to convince others of that. I don't think that "follow these specific actions and you will have a business" is advice that you can give in good conscience.
Generally, most people do NO (or very little) research, testing or thinking about their business idea before investing significant resources into it. My aim was to get them thinking, even if my way isn't the best. If I can help them avoid analysis paralysis and just DO something that moves their dream of starting a business forward, without risking too much money and time, then I'm happy. :-)
With over 10k words I didn't have the time (or frankly, energy) to dive too deep into any one topic. For example elevator pitching, which could be a 10,000 word blog post in itself. Happy to provide links for further reading though if you have any specific resources you find useful?
I actually think the culty thing is kind of a compliment; some of the most successful companies are kind of cultish and many great marketers talk about building 'Tribes' (e.g. Seth Godin). Maybe I just absorbed a bit too much of Noah Kagan's charisma when I worked at AppSumo, I dunno. ¯\_(ツ)_/¯
This is a bigger topic again, but in my experience rather than 'optimizing' (investing a lot of time choosing the best possible path) it's much more efficient to 'satisfice' (invest less time and choose a good enough path). Does it really make that much difference which elevator pitch format you use? Or is it more important that you actually do write one? Providing multiple options would only scare most people away because it's too hard to pick.
Hope that's useful background / partial justification for cultiness. ;-)
Of course consulting requires more in terms of people skills - you're basically "interviewing" every time you find a new client. But I think in some sense that's a hump to get over either way - there aren't many startups that don't benefit from founders having soft skills.
Sort of an orthogonal point to the article, of course.
If you cannot even run a single-person consultancy successfully, you probably shouldn't try a startup.
However I have one more for you: the businesses that buy from you as a consultant are very unlikely to buy your product.
Why? Because if they need a freelancer / consultant / agency to do the work for them, they don't understand the work enough to do it themselves (or don't have enough time). Therefore any self-serve solution tends to be met with blank stares and confusion. Likewise, the people who use our product on a self-serve basis are typically not a good fit for our agency.
- indeed you bill much more per hour, so in theory spend less hours.
- your work actually gets implemented, because you bill much more per hour. Generally "project management" is much better because they are buying a product and are incentivised to get a return on their investment.
- you deal with CEOs/managers as a peer, and learn (faster) from successful CEOs as a result.
- you will rapidly learn to sell. How do you differentiate yourself from the other 1,000 consulting outfits out there? How do you justify charging 5x what that Bangalore outsourcer is charging?
- it's a series of small successes that give you a boost psychologically when the product is so far away and the startup side seems not to be progressing.
- you think you're different because you're not taking money from "these risk-averse people" who "don't understand" your product. You'll get there the hard way but own everything at the end! Once this feeling becomes less important to you, you are ready for seed investment.
- by far the biggest is the time suck, and focus suck. You're context switching constantly. My first CTO quit in part because he was consulting as well, and the product got nowhere for months because we invested too much time in our main client. This time around I'm the only one consulting, CTO is full time on the product. It's still exhausting and inefficient, and he's justifiably annoyed when I'm unable to do startup stuff because I'm focusing on clients. Client takes precedence because you cannot screw up with them, you're delivering a professional product at certain standards. The business stuff is very important, more than it looks, and so the product launches later, sales happen later, marketing happens later, etc. which hurts the company because time is also a currency, arguably your most precious. The dynamics are well depicted in the movie Primer, where one of the group is working full time to pay for the others to build their time machine experiment. Watch the group dynamics when he returns home late at night.
- at some point you'll have to cut off clients, when your startup takes off. When this happens, if you were actually good at consulting, it's going to suck for them. If you are professional and ethical, this hurts you as much if not more as it hurts them, you grow to like a good client.
- Clients quit without notice. In my case, one sold to a larger group, another ran out of money. Then you're scrambling to fill the gap and spending time doing consulting sales instead of startup sales, because cash is king.
- should probably point out: what you bill and what you do is not exactly the same. If my client wants me on site for a 1h meeting, I'll probably commute half an hour there, then have lunch with the PM afterwards, then commute back to my office - total 3 hours for 1 billable. On top of which you are now on the Manager's Schedule  and have important opportunity costs from not having blocks of time to yourself.
A nice parallel is when I was a student and many of my colleagues had this idea in their head that they could be "everything" - have a phenomenal career, be a helicopter parent always there for the kids, retain a cool and active social life with Facebook worthy trips. After a few years they all ended up picking one of the three, because there's only so many hours in the day, and sleep deprivation actively harms your ability to think, which is a prerequisite to making progress in at least two of the above. You can't be an MD at Goldman and know what your kid was up to at school today.
Money buys you the time to focus on the startup and that's much more valuable than it looks. Of course this comes up on HN all the time, I just wanted to add a bit of flavour and an extra data point.
I followed the consulting path, but now I would caution against the consulting/freelancing path, unless one was building a consulting-agency. Theoretically everything sounds good, but has many problems in practice.
1. If you are successful at consulting, it means money (and potentially a lot of it). It takes stupidity, insanity and courage to leave the money on the table when work keeps pouring in. For last 6 months, I have halved my work hours to 20-25 per week to work on developing products. That was not an easy decision ( I still work total of 50-60 hrs per week). I had a few ideas that fizzled out quickly and now I can't say what exactly I'll be working on - I still have a few good ideas that I need to try.
But last 6 months part-time consulting did make me miss out on the money and opportunity cost. And the pressure mounts every day because I gave myself 1-1.5 years to try out.
I might be trapped too because if I fail at products it makes sense to start a consulting agency instead of searching for a job.
2. You might think - "I'll earn from consulting and invest in products". Really! That did not work for me. Bad hires, management-stress, and lack of focus was pretty evident. Not saying it won't work for others - but less likely to work if you don't invest focused time in products. I just let many go(in a nice way) and decided to start afresh.
3. I worry constantly and this also pushes me hard to do more work. I haven't made a successful cut into products, I am losing out my years and the opportunity cost is way too high compared to being an employee. My life is much better as a consultant-freelancer in absolute material terms, but this is not what I headed out for.
4. Hard to do justice to consulting project and product both. When I was doing multiple consulting projects, it was easy to switch between those. But now for some reason, its hard not to think about my product when I am also developing someone else's product. On one hand you want your customers to be successful and on the other hand you envy them and crucify self because you made the product work for them but haven't done it for self. That is just a feeling, but it hurts!
There are positive points for consulting - One gets exposed to the world of business, paces up, upgrades self, learns stuff like selling and negotiating. These can be done as an employee too (but slower).
If one starts on the side as an employee, there is only one issue to take - Getting out of the comfort zone and focusing on one thing. Employee is very comfortable in that setting and finds it easier to blame others/circumstances for failures. As far as that attitude can be avoided, starting up while being an employee is still a better thing.
P.S: Freelancing/Consulting is not a bad path - its wonderful, just check priorities.
>We've spent millions of dollars ... so you don't have to.
This two person company that's had, from what I can tell, a hand full of clients, has spent millions of dollars testing their marketing? I find that a little hard to believe.
That said, their Playbook page is pretty cool and informative, even if seemingly most of it is behind a paywall. It'd be nice to have more information about how these ROIs were calculated. If it's a sample size of three it doesn't mean much. Or whether it was done in combination with other things, and how long those results were measured for also matters a lot. As it stands right now it just sort of comes across as marketing bullshit because it's impossible to tell how meaningful the ROIs are.
The case studies are also pretty weak. Your solution is two sentences? Really?
We've actually just hired our 15th person and for the past 2 years we've worked with 70+ startups and spent around $2m. That's not counting one client who spends about $40k a day on FB ads. Before that I worked at Efficient Frontier (now part of Adobe), Travelzoo and ShopStyle. In each place I managed or optimized multi-million dollar budgets; the learnings from which I applied whilst building our playbook, and I apply every time I write a post.
ROI for our playbook is calculated based on the average of the experiments we've run. Looking at the numbers we've averaged about 2 concluded tests per tactic, but those numbers are highly skewed; there are some that have run 10+ times, others that have never 'run' (just things I've tried before but don't have conclusive data to hand, or things I've read about in blog posts). Either way, this data should be taken with a pinch of salt and used directionally; while eventually we dream of being able to build a recommendation engine that auto-prioritizes these tactics, the reality is that we'll need a lot more people using the system (hence the incessant blogging). :-)
Hope that helps explain where we're coming from?
* Have a page for your team to show the size of your business. It adds a lot of legitimacy in an often slimy, fly-by-night industry.
* Advertise that you've worked with 70+ startups. I had assumed this was a fairly new landing page for a two-person project who struggled to get three clients so they could put up a case-studies page. Obviously this isn't the case.
* You spent $2m of your own money that could otherwise be in your personal bank account right now? Or $2m of clients' money? Because there's a pretty big difference, and it's blatantly misleading if you meant the latter. It would be accurate to say "We've managed the spending of millions of dollars" if that's the case.
* Having a percentage ROI attached to a tactic suggests a degree of accuracy of at least 1%. That's obviously not the case. These are, as you described, calculated from statistically insignificant data or complete guesses. Maybe just consider a 1 through 4 ranking system of importance.
* Have meaningful content in your case studies. Be more specific. Give concrete examples.
Of course, maybe these changes aren't what sells a client on your services in this industry. My advice would definitely make for a more honest, less-bullshit website, but it might not make for a more profitable company. Just don't expect a warm reception on social media sites, especially ones like HN that are known for cutting through the bullshit.
That said, you handled the shit I threw at you very well.
* Very fair point, just added to our Asana.
* Client's money. We've spent about $20k of our money over the past six months or so. I would argue we're even more on the hook when it's someone else's money (and we do a lot of design / content marketing in house, that you may normally pay for externally), but I definitely take your point.
* Also fair point, but some data is better than no data IMO (otherwise how do you prioritize this massive list?). Open to suggestions though!
* Yeah we do try to; are you talking blog post now or playbook? We see our blog posts as living, dynamic products in their own right; we plan to update this with more useful information (and edit it down to remove unnecessary info) over time.
* We actually get that feedback all the time. "I don't know what you guys do, but it looks cool" is like the main line we get when we talk to leads. However, the site converts at 19% (seriously) and we haven't been able to beat it yet by being more informative. Working on it!
<tips cap> I'm relishing people actually caring enough to take the time to (fairly) criticize us. Much better than languishing in obscurity. :-)
Like I said, just have a 1 through 4 system. 1 being most important (or vice versa). It's roughly the level of accuracy you have already, and it accurately portrays that these are estimations/opinions, not actually data-driven numbers.
Conflicts of this kind would be a good reason to quit before you get too deeply into research, which goes against the title of the post.
* Do this on your own time. Maintain separation and document as necessary
* Make sure that you did not sign anything as part of your employment agreement that claims ownership over your ideas, thoughts, inventions, data etc. outside of working hours
* Do not use any company resources
* Read your assignment of inventions agreement _very_ carefully.
* If you live outside of California check your employment agreement for non-compete provisions.
If in doubt consult a labor lawyer. Getting qualified legal advice when starting a company is worth the money.
I mean, when I work for other companies I can crack down on problems they identified and have customers for and often I have good solutions for them. But finding these problems is hard.
That's the point of Step 1: Wanna know what problems a client is facing? Ask. If you don't ask, you'll never know what solution to build.
The quitting of the job can come later :)
Point is, the people I work for see me as a tool. They pay me money for my time to solve the problems they identified themselves.
It seems it's only applicable to business models that rely on getting revenue directly from customers (i.e. buying subscriptions or widgets).
Do you have any thoughts about how you might test an idea where your service is free, but you get revenue through advertising to your users? It would still be very valuable to get some sense of how useful users would find your site, how often they'd visit, etc..., but you don't really have a strong test (i.e. will they pay?)
The proving of willingness to pay came from surveys of those and subsequent users that said "yes, i would pay X per month for this"
Also I don't think my network is big enough to hit up a significant number for a new software idea aimed at a specific niche (ex used cars sales).
Those are the pieces I've never been able to put together in all my time with the Lean Startup methodology.
"Local educational startup looking for K-12 school teachers to survey"
I've done it multiple times for startup ideas and user testing. I've always got numerous replies. Most people are more than happy to have an email dialogue and many are also open to meeting in person to test your app.
If I take up more than 30 mins of their time, I usually get them a $5 Starbucks gift card or buy them lunch (I don't tell them ahead of time).
But the point is you can still do that on your own time without making huge sacrifices. Taking the concept of "go out and talk to people" too literally just serves to prevent you from actually finding and reaching your ideal audience.
Need to know where to start? Angel List, LinkedIn, Meetup, and the like. Write some cold emails. You can always build a network, and you should if you want to start a business.
What am I to make of this? Is the fact I got 50 signups a signal the product has value and I should continue, or does the fact users don't return a signal the product is crap I should scrap it? I'm really at a loss here.
might help us to know a bit about what the product is, but is there something specific that users should be doing in this free tier? what % of the 50 are taking this action?
Then you drive them to an landing page where you ask for the sale. But right before the purchase, you specify that the product is under development and will be shipped at an specific date. Thanks your "almost costumer" and collect his email for a future launch.
I think the internet is your best tool here, specially if you have some bucks to spend.
That is not really the hard part. It is a great excuse, and I use it myself to convince me I shouldn't look. But it's not really the hard part.
I heard Flexport did this by getting 300 companies to sign up before Ryan (the founder) wrote any code.
Eliminating risk isn't entirely possible. You could create a website for DoorDash/Ladder or Flexport, get traction, build Version 1 and then get told NO but I bet statistically you're more likely to have success with this method than building V1 first.
As an example, Optimizely got something like a $2-$4k commitment to pay without writing any code.
Again not everything works out this way. Snapchat, for example, would be a terrible example for this methodology but it works for some ideas.
1. I like the blog post. It may sound similar to the methodology in Lean startup, Y Combinator, and a bunch of other stuff but so what. It is clearly written and specific.
2. I have no idea what ladder.io does! I looked at the web page, and it is very slick, but I still can't figure out what you do.
Quite a difference.
You have a "real time tactic database". Is that the same as a listicle on the top 5 areas to spend marketing $?
You offer continuous testing. Is that Google Analytics?
Basically; we started as a growth hacking agency and then built our own platform for keeping track of what we tested (and what worked). If you want a free trial + demo, just sign up on the site and we'll walk you through it.
Thanks for the interest!
After all, for some the point of starting a startup is not only about business and money, as a software developer I want to build an application I love, and see it appreciated by others by making it a product and profitable business.
If you have read Paul Graham's essay, you'd know there's another way, the so called "organic ideas", i.e., to build a product that you love and useful to yourself, then sell it to others who have similar needs. To me this is a better way (albeit riskier), in the end of the day if I fail, I still would have a lot of fun and learnt useful skills, rather than just failed to fill a market gap.
What a truly heinous product. I've known a number of people who have been stalked, I can't begin to imagine the harm a stalker could cause with this kind of app.
But they all want to do something big. The test-drive concept is to go into building your new startup with data-backed reasons for why it's the right thing to do.
It helps you justify your decision to yourself and those around you. It's not just the job, but the people involved. At the end of the day, the entrepreneur is NOT the only one affected by the decision of starting a new business.
Except the part where you get a paycheck.
The opportunity cost of quitting your job is thousands of dollars a week. That's a big decision to make.