"How long is a piece of string? It's impossible to answer that question. But, if I <i>had</i> to guess... in my experience, the string is 19 inches long."<p>This is the kind of thing a lot of consultants seem to post, including the anecdotes of clients going with the cheap guys and the project failing. At the end of the day, the OP's answer is no more convincing then the hypothetical "outsource to Pakistan for 1/10th the cost." I think it's a consequence of someone trying to "sell" to an inappropriate market. If you expect to pay $120 for a mattress, sitting you down and lecturing you on why you need a $1,200 mattress is unlikely to convince you. If you expect to pay $10 for a meal, a $100 restaurant will not be getting your business. This post is the chef hanging out with other chefs and restaurant people to have a chuckle at those lame tourists who expect to pay $10 for a meal. The problem is that you are dealing with people who don't go to restaurants. They're comparing eating out to things they do know about like buying food in a shop.<p>The reality is that pricing development is tricky. Getting an "app" developed if you know nothing about making software is tricky, even if you are willing to pay <i>"well"</i>. A lot of projects fail. A lot of projects are not realistic to begin with. Cost, development time, quality, the definition of MVP, exactly how "bespoke" a project is, etc. all vary hugely. Quality is not always correlated to price. You can pay $60k for your MVP and have that die too. Developers will blame the clients and the clients will blame the developers.
His quote does seem a bit high. In my experience (Rails and iOS developer) if your back end is mostly RESTful resources and a simple data structure, and your front end is standard iOS components like simple tableviews and a NavigationController, you can get an MVP done in the $5k-$10k range in a week or two.<p>HOWEVER: everybody has a different opinion of what is meant by "Minimum", "Viable", and "Product". $5k gets you only bog-standard UI components and a simple data model. Animations? Fancy graphics? Optimized performance? Search? Custom UI? Graceful error handling? Localization/Internationalization/Translation? Integrating with Facebook and dealing with their constant poorly-documented changes to their API? These tend to be little bullet points in the spec, but each on their own can take as much work as the MVP does.<p>With modern tools it's pretty easy to build a basic version of an app quite quickly. But it turns out that most people don't actually want a basic version. Often they have to see the basic version first to realize that, though.<p>So the question ends up being: how important is schedule/cost to you compared with details/performance?<p>To be honest, though, most people who come to me wanting a simple iOS app are better off with a mobile-optimized webapp instead. Much quicker to build, already cross-platform, and no deployment delays while waiting for App Store approval. Mobile apps might not be as sexy as a native app, but saving lots of money is also pretty sexy.
Two developers working for 3 months feels more like 1.0 than MVP territory to me.<p>$60k <i>is</i> a lot of investment simply to test a concept, and would be a luxury out of the price range of most small businesses, let alone individual entrepreneurial subject matter experts who want to build a product - even really committed ones with money to spend.<p>Assuming you actually want to win the business, why not pitch a much smaller project to help them tease out some mockups and build 1 or 2 of the main application flows to MVP level?<p>This wins you the smaller engagement now with the likelihood of the bigger piece of work once the customer has been away and shown the concept to customers.<p>This is why I don't really like these $2k and $5k MVP packages that people are pitching on HN lately. Your iterations should start with mockups rather than an MVP.
<i>"Every group I worked with in the past six months spent several orders of magnitude more than that [$60,000] on their iOS products"</i><p>$6,000,000. $60,000,000. But only if you act now. This is a limited time offer. Operators are standing by.<p><pre><code> Offer not valid in all areas. Certain
restrictions apply. Shipping and handl-
ing not included. Void where prohibited.
Results may not be typical.</code></pre>
"Unfortunately, the developer used Phone Gap and when they needed to access more complicated native API’s in an upgrade, the developer couldn’t get past that barrier with the PhoneGap SDK"<p>The guy was not a good PhoneGap developer, you can write plugins to access any native API you need.
I live in Uruguay, and I can definitely believe it can be made for a lot less than U$ 60.000 (for comparison, that's 3 years of my salary).<p>I actually have a quote from a local company called XSeed ( <a href="http://www.xseed.com.uy/" rel="nofollow">http://www.xseed.com.uy/</a> ) for an app for both Android and iOs, and it comes out for a third of the amount.<p>Yes, you can be burned if you don't know the locals, like the guys with the Egyptians, or the Pakistanis. That's why many companies here in Uruguay get the jobs - the ones that are responsible cost twice or three times as much as the cheap Egyptian, Indian, etc.. companies, but they deliver, while being vastly cheaper than US or European counterparts. The really good ones can charge very close to US prices (go globalization :) ).
This reminds me a lot of the discussions I was having 12 or 13 years ago about developing a website. "You want me to pay how much for a website?!?!? My cousin can do that for 1/10th of that you're quoting."<p>Today, the difference between the plain-old website market and the mobile market is that few people on the demand side of the mobile market have any frame of reference or experience. How many people have bought a mobile app, compared to the number of people who have built a reasonably sized website? My guess the difference is a few orders of magnitude, maybe more.<p>Beyond that, I think it is also the sign of an immature supply side. If you wanted a blog 12/13 years ago, you had to pay someone to make it. Now you get one for free with blogger or wordpress. If you want a decent ecomm solution, you don't have to role your own.<p>Eventually both sides will figure it out: the supply side will add niche/task specific tools [1], and the demand side will gain the experience sorting through the options. [2]<p>[1] We see this this with a few of the dumb "hey, add your RSS feed" options for mobile apps.<p>[2] Also, equally likely in my mind: the raw horsepower of mobile hardware will make skinny clients more of an option, disrupting the app economy in general.
Seems most comments are nitpicking some of the amounts and references he uses to make some of his points. This is what I read into it as a developer who has been in similar types of meetings:<p>1. Charge what you are worth not what the person who has the App idea wants to pay you which in some cases is next to nothing.
2. Pay no attention to people saying, "but my friend says can do it in 3 days". Tell the truth and if they can't accept the truth move on. There is no point in explaining to someone that the most sustainable way to lose weight and keep it off is years of eating well and exercise.
What is an app?<p>The problem with this discussion is that you could be talking about anything from a simple program that displays some content (e.g. for marketing purposes), to some basic social interaction, right up to a full-scale word processor or 3D modelling program.<p>The amount of time and effort required to build a product can differ by orders of magnitude depending on what that product is. So it's senseless to talk about "$X is an reasonable/unreasonable price to build an app" without any context about what it is you want built.
This reminds me of a conversation I had back when I was freelancing.<p>It was along the lines of:<p>Client - "Aren't you afraid that Pakistani/Indian developers would undercut you and everyone would go there for their apps?"<p>Me - "If everyone were going to outsource there, I'd have more work fixing that stuff up, not less"
Software dev pricing is weird since it is so easy to <i>grossly</i> undercut yourself or overprice yourself by an entire order of magnitude.<p>I can certainly sympathise with people who have sent reasonable estimates and have received somewhat borderline insulting responses like "Wow, I could buy a used car for that!" or "That's nearly what my plumber charges!".<p>However in a certain sense I can't help but feel the industry has brought this upon itself. I remember ~2000 when websites were a new thing.<p>There were contractors around who would happily go in and quote very large sums of money for really very small amounts of (often very poor quality) work. Since the market had not really had a chance to self-adjust it's prices yet people would often just accept these costs without understanding what they were buying.<p>I remember a popular tactic being for designers to literally rip the HTML for a competitors website, switch the logos around and come back with a $10,000 bill.<p>So naturally the smart teenagers with a lot of free time to learn saw an opportunity to rebuild these websites better for the price of a few pizzas. This taught the market that costs could be <i>significantly</i> lower and also that the correlation between price and quality was very loose.<p>These kids also figured out the value of sharing code and ideas with each other, leading to a sort of open source renaissance.<p>Now that those teenagers have grown up and have to feed their families during a global recession those in developing nations are basically pulling the same trick on them that they had pulled back in the day.<p>You also have the added problem that there a sort of "iceberg" going on here. For example one can build an impressively featured website in around an hour simply by uploading one of the popular CMS and shopping cart systems to a shared webhost and copy pasting some content.<p>Of course, once you do such a thing you will get the familiar "this is great <i>buy</i> wouldn't it be ever greater if this drop down menu just sort of did <i>this</i>? I mean this entire thing took you an hour , so that change should take like 10 seconds right?" , "Well actually, I will need to develop a custom module to do X and then interface with Y so it's probably 4-8 hours depending.."<p>At that point it might be hard to convince a customer that you are not somehow trying to pull a fast one..
This article (and the comments here) remind me of something I read in <i>Managing the Professional Service Firm</i> by David Maister: whether you're hiring a lawyer, a car mechanic, or anything in between, you usually have little ability to asset their technical merits, so you have to choose based on other factors. That could be price, but it can also be things like reputation, responsiveness, patience & clarity in answering your questions, etc. Realize it's the same way when someone is thinking of hiring you. The article & these comments mostly focus on technical questions, and indeed if the tech is bad enough you have project failure. But if you're talking with technically un-savvy clients, you should help them realize you have more to offer than just a lower risk of botched tech, certainly more than just "better code." Particularly as a sole contractor, you can offer them a more personal relationship with wise guidance that they aren't going to get by hiring their cousin or the cheapest programmer they can find on elance.com.
I'm most surprised the MVP effort can be so consistently estimated. It might be that the pool of "apps" is extremely shallow, so they're just reimplemention #1231 of "XYZ" therefore estimation is simple based on extensive past experience reimplenting #1230, #1229, etc. "fart app #2935315" takes just as long to MVP as doggcatcher or evernote?<p>Also a major problem I've experienced is trying to force what I consider the technical requirements for a MVP past the non-technical people barrier, I didn't see that in the discussion, maybe it was assumed. For example, in my opinion, at least some minimal backup strategy is part of the MVP, whereas the non-technical types just reply "well, if THE hard drive crashes, we'll just sue the hosting provider, my buddy is a lawyer so he'll work cheap" or "if it blows up, you'll just work 36 hours straight on salary to fix it, right, so don't waste money/time preventing it from blowing up or adding debugging code or monitoring code, just work on features for the PR checkboxes".
These type of clients smell like deja vu to me and the post is a good light education for these kind of clients.<p>For people who get shocked at 60K for a full stack solution: go with the devs from [insert any outsourcing country here] or with the friend and keep a journal of how things went. I'd be damn curious to read it... (even though I already know the outcome).
For most iOS applications you don't necessarily need a designer and two developers. Trim the fat.<p>One developer is more than enough for most iOS applications that doesn't cross into games and physics.<p>You also don't need an "Mobile UX, UI, iOS designer expert" either. Just get a good developer that can do both, instead of mixing science with snake oil.<p>In the end, one guy to be the product owner, develop and design will be leaner, and far more better than involving more people.<p>Less is more.
It's so true. I've been a freelance iOS developer for a year now and half of my contracts have been fixing amateur looking code of large consulting firms. Is this endemic of iOS work or all software development? My theory is that consulting firms have trouble hiring senior engineers because working for a consulting firm isn't "sexy".
60k might be a bit much... we can often get a mostly functional app off the ground for 30k. It just depends on specs.<p>Web MVPs (landing page with a signup form) can be done easily from anywhere between $200-1000. Yes, I realize there's stuff like unbounce, but even with those, they still take a couple hours to set up and properly split test. If you've never done this before, you can easily burn through that very quickly, and even if you have done this before, and you're valuing your time as any good founder should, then you're burning through that value allotment very quickly.<p>As for the foreign quotes, I've taken over plenty of jobs where it was started by a foreign firm for a tenth of the cost of what US firms charge... quality is usually terrible, and the reason I took over is because they couldn't get it finished. It's occasionally been so bad that we had to scrap their entire project within 6 months.
I've had a lot of these types of meetings. Everyone has an app idea. I'm always ears though because ideas have come to me that are very interesting. The noise to signal ratio though is pretty terrible.<p>It's always interesting to have them reference apps and designs that they think their app should end up looking like. Very quickly a general breakdown of things in regards to the app they are looking to build can follow. Even taking the approach of outsourcing and painting a picture of how that works.<p>A lot of times I've gone the route with at least doing the mockups of the pages with a little on how the UI will work. It's something that I know I'm selling that will be of value to them even if they choose not to have us continue on with the development and to just really boil down the idea and even more importantly what kinda of money and time it's going to take to develop the idea.
When we needed native Android and iOS versions of our app, we contracted it out to a couple Indian developers. We figured that the development was pretty straightforward - it was essentially a port of our mobile webpage that had some performance issues, so we had all the server-side endpoints and a working example to copy from.<p>Was their work output buggy? Sure. But they got it done at a reasonable speed and cheaply, and they fixed the bugs as we noticed them (and we were paying them to...). For us, being cash-strapped and needing apps just to say we have them, it was pretty OK. I think it worked for us because we didn't expect a perfect, polished app, and the requirements were well specified from the get-go.
In software there is a correlation between spending money and getting good quality work done. However - it's only a correlation.<p>Yes - some people get their nephew to build a modern, well coded, SEO-friendly and accessible, ajax-driven, e-commerce site for $800 but they probably just got lucky. A lot of nephews get out of their depth and never deliver.<p>And yes - some people spend a couple of million on the project with the same specs and get given a piece of crap in return. But on the whole - like good whiskey - the price tag is a good indicator of quality and a good way to reduce risk.<p>So - like many areas of enquiry - anecdotes aren't incredibly useful. If anyone has got any tips on finding the magic nephew, however, I'm all ears.
Amen brother.<p>To me, the key is right here:
"My hunch is that the people who quoted much lower estimates and timeframes were shooting for the development work."<p>Firms/individuals/groups that quote dirt cheap prices are probably either a) desperate for the work (which makes ya wonder why) and/or b) not caring about the work they do at all (therefore producing a sub par final product).<p>A big mission of my new startup matchist (<a href="http://matchist.com/talent" rel="nofollow">http://matchist.com/talent</a>) is to filter out these clients for freelance developers, so they only deal with those who aren't looking for 30 lbs in 30 days.<p>We're not 100% there yet but we're working damn hard at it.
Excellent metaphor, I like how this post explained simply difficult problem.<p>I can expand on it that I also ran into, like most of us, on similar comparisons and while I don't think it is wrong to seek help overseas, especially in places like India/Pakistan, I think that it is completely wrong to have clueless person seek help and then when they get burned, come back running to me and tell me about those bad, bad, people over there.<p>So again, excellent metaphor.
From the Project management triangle:<p><i>Like any human undertaking, projects need to be performed and delivered under certain constraints. Traditionally, these constraints have been listed as "scope," "time," and "cost".</i>[1]<p>It's amazing how often I need to reiterate this to clients and potential clients.<p>[1]<a href="http://en.wikipedia.org/wiki/Project_management_triangle" rel="nofollow">http://en.wikipedia.org/wiki/Project_management_triangle</a>
Thinking about the title, I just kept thinking of Christian Bale losing 60 pounds in 4 months. If that took him 4 months, I don't know how you'd lose 30 pounds in 30 days (well in a health way). <a href="http://en.wikipedia.org/wiki/The_Machinist#Production" rel="nofollow">http://en.wikipedia.org/wiki/The_Machinist#Production</a>
We deal with educating clients on the cost of apps and MVPS all the time at matchist. We get clients who have been burned and understand the value of not outsourcing overseas, yet I think there is still sticker shock at the costs associated with building exactly what you want. What do you think is the solution to this?
This article really got my goat on so many levels.<p>Nice response here: <a href="http://news.ycombinator.com/item?id=5129675" rel="nofollow">http://news.ycombinator.com/item?id=5129675</a><p>Full disclosure: I am an investor in the response posted to the OP.
Funny, I have a completely opposite experience :)<p>It really depends on people, not about location. Talent can be EVERYWHERE, as well as not so talented ones :)
I make $48/hr at my fulltime job. I take an odd freelance job here and there and charge the same rate to my clients. Sure, I could charge more but I don't have a huge portfolio. So, he charges $125/hr and has 3 examples of clients whose previous devs failed. Must mean he's correct.
This is ridiculous, the Simpson-Bowles commission didn't come up with a recommendation:<p>"The Commission shall vote on the approval of a final report containing a set of recommendations to achieve the objectives set forth in the Charter no later than December 1, 2010. The issuance of a final report of the Commission shall require the approval of not less than 14 of the 18 members of the Commission."<p>They never had that vote.<p><a href="http://www.cepr.net/index.php/blogs/beat-the-press/the-bowles-simpson-commission-did-not-issue-a-report" rel="nofollow">http://www.cepr.net/index.php/blogs/beat-the-press/the-bowle...</a>