A community discussing the business of software, from the smallest shareware operation to Microsoft. A part of Joel on Software.
This community works best when people use their real names. Please register for a free account.
"Make Better Software" is a 6 movie course designed to help you as you grow from a micro-ISV to a large software company.
Doug Nebeker ("Doug")
I wrote a blog post about my first product failure, and it was quite a commitment in terms of time that ended up failing.
The blog post can be found here: http://bit.ly/xv3YlJ
I'm wondering if anybody here has had similar experiences, but more importantly, if they corrected their mistakes and ended up with some success in their new projects.
Etienne Yvan Morin
Friday, February 24, 2012
I spent about 3 years on my product. It was released last November, had about 5000 downloads to date, and at least 3000 people actively trialed it (based on automatic version checking).
But last week, I pulled the plug on my website. I've had zero sales, zero contact from anyone (including people I specifically emailed about it); zero anything. It's like it was never even released. Tracking stats with StatCounter and Google Analytics showed visits from posts and blogs where I advertised it, so people definitely knew about it. Downloads from Softpedia, BrotherSoft, etc were okay, but nothing to write home about.
I did all the Right Things (tm). My website explained, in short phrases, what it did, why it benefits the user, why it's better than the competition, and that it came with a 30-day money-back guarantee.
I offered it at a high price (US $49), then a low price over Christmas (US $9), then offered 2 free upgrades to trial users if they sent me feedback about it.
I offered two forms of payment (PayPal and FastSpring), and had my ABN on the website, and my WHOIS wasn't masked. Alas, I didn't have my name or phone number, and being an online product I didn't publish my home address; but I don't seriously think the lack of these is what caused no sales.
My product has 2 direct competitors, with their products not being updated for over 3 years each. So, my conclusion: as with their products, I guess nobody is interested in what I have to offer. Simple as that. What this experience has taught me is that you can tick all the boxes, do all the right things, offer the right price; but it doesn't guarantee sales at all. It's just "one of those things".
My father (I'm 41) said it's a stupid product that he'd never use. But the other night, he emailed me one of the competing product's names, and said "Do you know about this?". I replied with, "Yeah, that's one of the competition. See Dad, I told you there's a market for it!". He didn't reply.
Some people here have said that 3 months is too soon to pull the plug. Perhaps, but I doubt it. If over 3000 trial users haven't paid after 3 months, then my gut tells me they're just not interested, and leaving the website up and costing me money every month just isn't worth it. I'm not made of money to waste like that, just hoping for a sale.
Again, in summary: sometimes you just don't have a wanted product, and you have to accept that, pull the plug, and move on. Didn't someone once say that you should "write one app to throw away" anyway? This was my app. :)
I did learn a lot from the experience, though, about marketing in general, and how to get things right. Here's hoping my next product will do better.
I have a few older products that never sold or that only sold one or two copies. Just for posterity I keep web pages about the programs around, but not linked to from anywhere, noting they are not available and don't even run on modern systems. Every once in a while, like every two years, I get an email about one of them. It's always the same thing. Some guy developing a similar product who would like me to gift him the source code. Usually there's a preamble about what a shame it is for such a fine program to become abandonware, and he is willing to do me a big favor taking it off my hands. I generally ignore these inquiries nowadays.
As I see it it's not abandonware when it never sold any copies.
There's a new product I have now that I released a couple years ago and there just wasn't much interest, it's a bit ahead of its time. I was about to shut it down, but then about six months ago a guy bought a copy. I thought "Drat, now I have to support this thing."
@Etienne: A couple of comments which I hope will be useful for you when you start your next product.
1) 3 years (even 2 if you count only your desktop version) are too long to release the first version. I started my market research in November 2007, wrote the first line of code in December, released Beta in February, did Final release in March, got first sale in April. Code like crazy, don't sleep at nights if you have to but create a minimum viable product in a couple of months.
2) Don't choose technology that you are not familiar with. It is OK to learn some new technology ALONG the development to add features to the product which require that new technology, but do not build your entire product using something new for you. Among other things this will also make it difficult to asses time-to-market (see point 1).
3) When you have something working release it. Why did you not release the desktop version when it was ready? You spent 2 years on it and did not even release it?
I wish you luck on your next product.
"An automated time-tracker and timesheet application ."
There are lots of applications out there that do this, maybe not as bad as Twitter clients but not far short.
"One of our friends tested the application, then quickly pointed out a number of issues."
One of the things I learnt early on is not to take any notice of what my relatives and friends said I should do. If I had I would still be stuck working in some shity job for someone else. Use the market to test your ideas. A lot of the advice you get on this forum will also be just as bad as you got from your friend.
Not my first product, but I have had products that never really took off. See #11 in Andy's blog post:
You might take a read through all of the entries. #12 is in the same category as the one you wrote.
Saturday, February 25, 2012
@Ducknald: I agree with not listening to family and friends. I persevered despite my Dad's bad-mouthing, and while nothing came of it so far, I still actually believe in the app enough to give it another shot in a few months when I'm not feeling so jaded by it, as I am right now. I want it to sell, and I will.
@Scott: My response to anyone who wants the source because it'd "be a shame to become abandonware" would be: okay, make me an offer.
As for the "find 10 people who promise to buy it" tactic, it didn't work for me... even when 2o people promised. Turns out they were just humoring me. So this concept is baloney.
"Some people here have said that 3 months is too soon to pull the plug."
I can say that. I've got a first sale after 3.5 months :) But it was almost 10 years ago, and now times are different...
But it doesn't cost mush to run a web site. Your traffic is low, so you can host it for about $100/yr.
Saturday, February 25, 2012
A comment on the blog itself pointed out that I shouldn't give up just yet, stating that "millions of people use RescueTime" and that it doesn't allow to modify timesheets. I generally agree with him that its a huge market and I might go back and fix the areas and relaunch.
The thing is, after three years of different incarnations, it was time for something different. And as Scott pointed out, if we had users, we'd have to support the application. This wasn't something we were willing to do, knowing that our data storage wouldn't scale using SQL Azure with the amount of data we'd be gathering.
To fix this issues would require many more months. So instead, I've been learning some new technologies and gaining a different perspective on how online businesses should work. If I go back and fix the project, I feel it will be a much higher quality product, as it will scale and I will take the time to fix usability issues from a fresh perspective.
@B2B, I agree, 2 years for the first launch was too long. To do it again, we'd have a very different minimum viable product - automatically track your own time and see your own timesheet. Period. None of these manager reports and submitting timesheets for approval business....we didn't even have the domain knowledge to develop such features - that's where user feedback comes into play. This project, even if it becomes only throw away, has taught us some good lessons. I also agree that learning new technologies for a new project is a bad idea...it will only slow you down. Unfortunately, jQuery, client-side templating, and REST API's existed, but weren't nearly as popular as today, so doing it in HTML for the type of usability we wanted didn't seem right at the time. I have a completely different perspective now....
CleverTimesheets. We pulled the plug, took down our server instances. You won't find it anymore.
I think that the mistake we programmers make in business of software is that we are too "feature" oriented and we are excited about all the new technologies and always are trying to do something new and innovative. This is great from a programmers standpoint, but if you are trying to make a business you must forget this mindset. I say this because you (the OP) in a statement above said you are again pursuing some new technologies etc. Just forget that. Start thinking "business". You must find a way how you'll sell your next product. Don't find excuses for your failure in your bad technical decisions.
I say this just because I am also doing this again and again and learning the hard way.
Saturday, February 25, 2012
"My response to anyone who wants the source because it'd "be a shame to become abandonware" would be: okay, make me an offer."
It's easy to say that, gosh free money.
But think about it for 10 seconds and you realize you'll run into the factors I did when I had those conversations.
1. How many dozens of hours are involved in IP negotiations?
2. How much does your lawyer charge to draw up indemnity and copyright assignment contracts?
3. How many hours to figure out which parts of the old code are now part of libraries you're currently using?
4. How much is code worth that can't compile or run on modern operating systems, that relies on frameworks that don't exist anymore, than never sold any copies to begin with?
Here are the answers.
1. Between 20 and infinity.
2. $450 an hour or $600 an hour depending on whether a senior partner is involved.
4. Less than zero.
You have my sympathy. My own first product is definitely either in a Seth Godin 'dip' or perhaps it's a 'cliff'... only time will tell.
I've had sales, but despite modest running costs it has yet to turn a profit. Early days, and most of the lack of interest I believe is due to the competition being established and far better performing in Google. I am working on this...
Saturday, February 25, 2012
> A comment on the blog itself pointed out that I shouldn't give up just yet, stating that "millions of people use RescueTime" and that it doesn't allow to modify timesheets. I generally agree with him that its a huge market and I might go back and fix the areas and relaunch.
I used to use RescueTime, and from the video you posted your application looks better, both functionally and aesthetically.
The confusing part is the targeting. Managers and employees mentioned in the video are not often use the detailed timesheets, in my experience. Independent contractors, quite opposite, do. May be it should be targeted to independent contractors looking for wait to become more efficient in using their revenue-generating time?
The enterprise version of the software should probably be added on version 2, with a greatly higher price.
BTW: a category of people who:
a) use very detailed timesheets, in 15 minutes steps
b) charge a lot for their time and so can pay for an accurate timesheet
c) use largely simple tools that are easy to classify based on the window title
d) mostly Windows users
are lawyers. Lawyer answers your email? 15 minutes charge, some $100 for him.
Refocus the application to lawyers, add a few features that help them, and you should be OK.
"it doesn't cost mush to run a web site" -- It does when you have zero money to spare and are living week to week. Perhaps the people here haven't ever experienced that? Everything I earn in my day job goes straight to debts, my wife and my kids. I literally have zero cash most months. To get cash for the hosting I previously had, I had to ring the electric company and ask for a payment plan to pay off the bill. Can't keep doing that forever.
@Scott: When I said to ask them to "make me an offer", it doesn't mean you have to negotiate afterwards. Either accept what they offer, or say "no thanks, don't contact me again." It's a quick way to get them off your back, since you'd rather ignore such inquiries these days.
I ate ramen, grew vegetables in a garden, hunted, begged, and did dumpster diving for about 5 years to be able to survive while growing my company. It didn't bring in a living income for almost a decade. I knew that was going to be the case and that's why I saved up enough to live off of and buy a property in a backwoods area so that I could go as long as possible as I needed to have the time to get things going. I've worked over 4000 hours a year for the last 15 years now, and often more than 5000. I am also 10 times more productive at the minimum than the typical guy, so that's more like 50,000 man hours a year. My main product now has several million lines of code that I have written.
You shouldn't be even attempting a startup in your financial condition. Get a programming job and save. Move if there aren't decent jobs in your area. Your poor wife.
Scott: how were you able to stay motivated for years working 80 hours a week? Was it 80 hrs a week coding or doing other activities like research, marketing, etc? When you started, how well was the project defined? Was it one product or several?
I quit work and spent 8 months on a project that didn't reach beta. After 6 months I wasn't making any more progress; I was blocked and new it. I bit off more than I could chew and knew it. That's when I started the job search; total time off was 8 months. I had cash to go another two years but just couldn't justify it any longer. I was only productive for the first 3-4 months or so. I don't claim it as a failed product though as it was never released.
Wow, that's a really interesting question. I've never asked myself that. I'll have to speculate, so it might not be accurate since I'll be hashing up old memories. I might have a different answer later.
I'd say there were a couple things that were extremely encouraging and motivating.
The biggest was making sales and getting positive feedback from customers. At the beginning there might be a sale every 3 months. I'd often start conversations with customers and get a lot of information from them and work together on solving their problems.
The other was eating my own dog food. I used the software to mess around with little videos and effects. It's lots of fun and you can kind of experiment and be artistic with the code and try different things. Kind of like playing a video game, intrinsically rewarding and infinitely vast. There was always something new and exciting to do.
When working for the man before this I often found myself having to work similar hours. I hated it and it exhausted me. But doing it for yourself is a different issue. You're not being exploited.
As far as activities, it was quite varied, included lots of stuff, including designing order and invoice systems, reverse engineering things, reading books on sales and marketing, web design, all the business stuff. Things generally done in spurts of one thing. Context shifts are hard for me.
In your case I think if you were just doing one thing for 6 months straight, I can see why you burned out.
I think it takes longer than 6 man months to build anything substantial.
I know there is a lot of talk by tech writers about having some sort of service that one can have on line and generating fees in a few weeks, and those products do exist, but most products, and I think the most interesting products, take a lot longer to get to the point they are useful to others.
You might consider revisiting after doing something else for a while. Or at least extract some of the work as libraries for a new project. Or just be experience. No matter how badly one of these efforts fails, one owns the code at the end. Over time one builds a substantial code library that they own 100%. Eventually that can be a big jumpstart to the next effort.
I'm curious.. why did it take almost a decade to make a living with your product?
How did you sell it? AdWords? Cold-calls? Or just waiting for customers to show up?
I just shipped my own little product this week, and I'm planning to cold-call a shitload of companies. So far, the outlook is reasonably good.
I know my product is very competitive. It was designed with the help of someone in my target audience, so it's got some relevant and useful features. Straight out the gate, it's just objectively better than almost all of my competitors'.
But still, it's scary not knowing how long it will take to earn a living with this. I'll be _very_ unhappy if it takes more than 6 months, because I feel like my life will be on hold for as long as it takes.
It's true you don't need much money to start selling lots of types of software but some small amount is a massive help. For example if you need to shut down the servers because you can't afford them that's unfortunate, further work could have proved the market and your investment in time might have seen a return.
Also it's difficult to prove some markets without doing something like a slow growing adwords campaign. I wouldn't give up on most products without trying that. Depending on the product price point etc. of course.
Perhaps you could get some consulting to fund your endeavours?
Sunday, February 26, 2012
Meh, could be worse, you could be me. I released 2 'products' between 5 and 3 years ago, spend thousands of hours (but only a couple of hundred of USD) and never sold a single copy. Looking back, I'm embarrassed that I ever thought these 'products' could have a real market.
But that's only the start. The main reason for these failures was that I didn't do any market research, and yet I fell into the exact same trap with my latest product. I've been working on it for coming on 3 years now, but instead of having a shipping version (or even a beta) I got side-tracked by technology so much that I'm further from having a working version than I was about a year or so ago. Ffs, I even convinced myself that it would be a good idea to develop a custom programming language for my product, and so I did, including a basic standard library, and I re-implemented my product (the parts that were working) in my own programming language (well I'm in the process of doing that, actually).
I guess I've admitted to myself that I'm a programmer more than a business guy, and that I like technology more than the business aspects, and that all the things I've been doing are more hobby programming than really building a business. Maybe I'll have a real product at some point, who knows. I have a good day job to pay the bills and then some though, so I don't really feel regret over these things - I could've spend that time watching Survivor or The Voice and I would've been worse off than I am now, I guess.
"Why did it take 10 years"
Bonita, my experience parallels what Joel wrote in "good software takes 10 years".
I had income before 10 years, but around 10 years is when I made as much that year as I would have made if I stayed on the corporate salaryman path.
Patrick's Bingo stats show something similar even though his program is modest. It took him many years before he could support himself from income.
It seems a lot of people think they will spend a few months writing an app and then make a good salary starting shortly after release and that will remain consistent and pay a living wage for the rest of their life off that brief effort. Good luck but it seems unrealistic to me. That's how corporatet jobs work. You get a solid consistent salary, even the first three months when you are training and nonproductive. As a result, when you are highly productive and bringing in millions, you still earn the same salary and the company takes the rest. Trade security and consistent income for control and ownership. That's the trade off.
Entrepreneurship is not for everyone. I think most the people blogging about their startups don't have what it takes for the life.
"It seems a lot of people think they will spend a few months writing an app and then make a good salary starting shortly after release and that will remain consistent and pay a living wage for the rest of their life off that brief effort. Good luck but it seems unrealistic to me.".
Scott, I guess my case is rare but that is what happened with my product. I started making a good salary from my product in about 6 months after the first release and quit doing contracts a year later. Today, after almost 4 years since that first release I have 3 employees, 60% of our company revenue is profit and we have cash in bank that will cover 12 months of our expenses if we stop making any sales right now. I call this success (billionaires are free to disagree) and it took me less than 10 years.
Now, I live in a place with IT salaries about 50% less than in the USA. If I had to adjust my figures to see how it would apply to someone doing the same from the USA the timescale for each milestone I had would shift by about 6 months but still it is nowhere close to "good software takes 10 years". Then again if I were in the USA and were native English speaker a lot of things would be much easier to do so that is a moot point anyway.
So, it is possible, though I have to admit it is probably rare.
As is always the case, there is no luck involved. A failed end product is because of failures in other areas probably earlier on.
Successful, money making software doesn't take 10 years to build. The problem is usually either that the product is not necessary because no one wants it, or, if they do want it, they don't know that they want it or they don't know where to find you.
If you code for a hobby too, like most of us do, then you'll need to re-prioritize once you realize that you can't pay the bills. Don't code a thing until you know you have customers lined up in advance. As a result of my software entrepreneurship, I learned about a bunch of internet marketing stuff, copywriting, etc. Many internet marketers will put up a one page website where they make a brief pitch and collect email addresses. They don't develop the website unless there is sufficient interest. This saves time.
"Before you build a better mousetrap, make sure you have some mice."
— Yogi Berra
B2B: you're mis-interpreting what "good software takes 10 years" means, at least what Joel meant by it in his article.
It doesn't mean: it takes 10 years to make a profit. That's obviously not true and there are countless of cases where profits were made with software in much shorter time.
What Joel meant is that
a) to reach functional maturity (i.e. implement most of the features that customers ask for) takes a lot of time
b) to reach the spectacular, outsized profits (the kind of profits that Word or Oracle database or Photoshop) *usually* it takes many, many years. It doesn't mean you don't make good money earlier or that there aren't exceptions, especially in our heavily networked world that can spread word of mouth much faster and make instant hits (hello, Angry Birds), but it's true more often than it's not
The article also applies only to certain kinds of software: software that is potentially deep in functionality but still useful even if it only has a fraction of potential features. It doesn't apply to simple apps like, say, Twitter clients, or games, that have completely different schedule of making money (they are developed for many years without user feedback, there's only v1 and it has to have all the fetureas a user will ever get and they make all the profit in the first few weeks of sales (I'm talking about traditional desktop/console games, that obviously doesn't apply to games like World of Warcraft).
The point is: what Scott wrote is true more often than it's not true. Rarely in the software world you get an instant, hugely profitable hit. Usually you have to slog for many years, start with v1 that has few features and grow that into v10, with much more features and more revenue potential.
Even your story is an example for how true that is. It took you 4.5 years to get good profits and if you're lucky, you might get to truly mind-blowing profits in the next 5 years, proving Joel's 10 year rule.
Sunday, February 26, 2012
Not sure why people are so confused about this. Really nothing more to say. If you think I was saying my process was waterfall with first release after 10 years, you just keep thinking that while you release a cascade of failed abandoned products you were never willing to commit to long enough to make them into an industry leader.
For those truly confused, around the 10 year point (might have been 8 depending on how I count) is when I first made $150,000 salary after expenses and employee salaries. That's what I was making when I quit to work on my own thing full time several years earlier. It was also right on the exponential growth curve that 10 years is on that article of Joel's I posted and it's been several years ago that I hit that point and the curve grew as you expected, was pulling in 7 figures a couple years after that, and am now at 8 figures, have employees, a company, was able to afford to escape the police state the former US became, and now can do what I want, design what I want, and accomplish what I want.
Joel has also obviously followed this same curve. CityDesk failed and was abandoned. FogBugz was a big hit. He now has built his own offices taking up an entire floor in a nice building in the good part of New York City. All through this time they have been making improvements, and now their product is a de facto standard and he is considered an industry leader.
>As is always the case, there is no luck involved.
I think there is always luck involved - to the extent that there is always information you cant know and lots of factors outside your control. For example, Microsoft or Apple might launch a direct competitor and wipe you out overnight.
Monday, February 27, 2012
Startup: 'As for the "find 10 people who promise to buy it" tactic, it didn't work for me... even when 2o people promised. Turns out they were just humoring me. So this concept is baloney.'
Nope, but it is not a guarantee either. If you can not find ten people who promise to buy it, it is not likely to fly. If you do, then maybe. I suppose the exception would be where nine or fewer sales would be enough, but then, that is pretty much custom software, not a product.
Cheer up. I believe you are hurry to make a conclusion that it is a failure.
“a very hard market to penetrate”
Yes. And it needs time. Some products get customers pretty fast, others need some time and improvement and it takes 2-3 years before people starting to use it. Continue to make it better.
“wrong choices in technologies”
Customers do not care about technologies. Developers do. Some technologies maybe more handy for your goals some does not. Customers care about benefits features provide.
“never proving that our product was something people wanted”
Do you really know that? If it is, add stuff they are interested in. If you already have some base you can easily add some more useful stuff.
If you are pushing a lot of money into marketing, probably you are right, one year after release is a failure. If your marketing tools are facebook, twitter and your blog - probably you may continue.
Who may be interested in your product? Contact with some big player and give it him for free.
Anyway you can’t expect much from v1.0; Check this article http://www.joelonsoftware.com/articles/fog0000000017.html
I agree with it.
If you are feeling exhausted, try to get investor and get more money for marketing.
Tuesday, February 28, 2012
I agree, customers don't care about technology so long as it works... Our choice in technology was guaranteeing that it wouldn't scale, hence not work. After three years, realizing that we'd have lots of work ahead of us to gut the storage and re-implement it. It would be a nightmare to support customers and then completely redo the backend. That's probably the real reason that we didn't stomach the project any longer.
After reading some of these encouraging comments, I motivated to start it up again in the future - just not now.
Etienne Yvan Morin
Tuesday, February 28, 2012
I'm not clear on why SQL Server won't scale on Azure or why that's a concern. For one, is it true? And two, do you care? Should you care for the first couple of years when you'll be at the bottom of the growth curve? This judgement should be revisited I believe.
Otherwise, if you look at the internet scale apps like Reddit and twitter, they're not using the MS stack. I'm not familiar with Azure so I don't know the ins and outs. BUT I do have general unease with regard to Microsoft dictating what I need to build an internet scale application. The cool kids are using EC2 and NoSQL; that's the proof by existence.
SQL Azure had a 50GB limit at the time. They've upped the limit since then, but we thought it would still be too small. Choice in technology, hosting, providers, etc.. matters. We chose wrong, even worst, our backend implementation would have been a nightmare to re-do due to a SQL-heavy logic.
Etienne Yvan Morin
Wednesday, February 29, 2012
It seems deep down you were still trying to learn new technologies. And you did that well. Most of tech people are just that. Tech people who are trying to build their resumes. Switching the mindset to building a business is very difficult once you have made the transition to being a tech person. As an outsider looking at what you DID, without your explanations, was learning new tech using a sample project. This is done in technical books . You started with wpf, then made it web based, then jquery, etc..etc.. You could have written a book using this app as a sample. I have been in the same boat as you, believe me. You need to either change your mindset for real, which is very hard, or have your business partner that can manage. Bad tech choice is a weak argument I think. Answer this question for yourself. If you could add up all the time spent on the app, how much of it was spent selling, marketing(face to face ,online or on the phone) as opposed to developing or learning. was it 50-50 or more like 10-90 ?
We do what we are good at. Research shows people make the choices they know how to make. Comfort zone is the name of the game. You are comfortable learning and developing, and that is what you did.
This topic is archived. No further replies will be accepted.Other recent topics