“We’re 100 people, but we have work for 130 people. We just don’t have those extra 30 bodies,” Michaels says. He adds that salaries for experienced iPhone developers “just keep going up. Our year-over-year salaries are up almost 20 percent.”

iOS development is a demand market right now. There simply aren’t enough qualified developers to meet the enormous business demand. Due to the high technical nature of iPhone and iPad development work, this is a situation unlikely to improve in the near future.

The high market demand is pushing unqualified and marginally-qualified developers into the market, who are unable to deliver on business objectives. This is why horror stories of bad hires and bad contractors on the iOS market keep getting published. TapTapTap, a market leader in iOS publishing, writes about their bad contractor experience:

We continued to sink a lot more money into the project, partly because we wanted to add some cool features that we came up with, but mainly because Daniel couldn’t deliver anything even close to being worthy of shipping. And even though we weren’t supposed to pay until completion of the project, we did so on good faith, mainly because of Daniel’s never ending sob-stories. literally months would go by without an ounce of work being done on the app. When he did actually come around to doing some work on it, subsequent builds improved slightly, but we never got the the point where someone would ask us about plasma and we’d be proud to show it to them… instead it was usually quite the opposite and more of an embarrassment for us.

MacWorld has this to say about overseas outsourcing:

“This is not a skill that goes well with outsourcing because the typical shops in India and the Ukraine are focused on wider-breadth technologies such as Windows, and there aren’t a lot of Mac developers there,” Michaels explained. “Most of the Mac developers have always been around Cupertino and San Francisco, where Apple is located, and there are some in Seattle, Portland and Vancouver.”

When you are looking for an iPhone or iPad developer for your project, make sure that you do the technical due diligence to hire a candidate with a track record of timely delivery of iOS projects that operate well, are written with best practices in software engineering, and meet technical and functional requirements. Your project is too important to leave to chance.

According to the Standish Report, 71% of software projects are challenged or fail completely, and the average budget overrun for those challenged or failed projects is 56% over the original budget.  Software development is a risky business by any measure.  Overruns in software far exceed every other industry, and for those not “in the industry” it is difficult to understand how a project “estimated” at $25k turns into $39k in the blink of an eye, but it is unfortunately a market reality.

The dirty secret about T&M terms is that there’s no guarantee that the final cost resemble the initial estimate.  It’s no wonder that many contractors favor Time & Materials contracts, where when (not if) the budget is overrun, the client eats the cost.

Even above and beyond the Standish Report’s numbers for generic software development, competition in the iOS contract development space is at an all-time high, and there’s increasing downward pressure on iOS contractors to get contracts.  Over the same period, the actual cost to develop iOS projects has remained relatively stable.  This creates an impossible situation.

Since there’s no requirement that a T&M estimate match up with reality–it doesn’t.  Contractors keep producing more and more unrealistic “estimates” in order to beat out the equally unrealistic estimates of competing firms.  The end result is a sham competition to produce the lowest number, instead of the correct one.  We monitor the market constantly, and we regularly see “estimates” going out for 10-20% of the final project cost.  Either scope is aggressively cut, or budget is aggressively raised (often both), and the final product is nothing like the negotiated functionality or cost.  Sometimes the project is simply abandoned.  Either way, the client is left holding the bag.

My grandfather used to say “If you pay a man by the hour, he’ll work a lot of hours; if you pay him by the brick, he’ll lay a lot of bricks.”  Fixed Bid projects mean that you pay by the project, so you’ll actually get that project without any bait-and-switch.  It also means that the estimate the developer produces will be a real one, not a sham number to get you to sign.

In an environment where budget overruns are incredible, the most valuable thing you can have is insurance.  When you sign a Fixed Bid contract, you’re not only getting a project delivered–you’re buying insurance against a bad estimate.  The contractor is legally required to produce the agreed-upon project at the agreed-upon price, regardless of the accuracy or realism of the original estimate.  The developer eats the cost of overruns.

Contractors are reluctant to offer Fixed Bid terms for three reasons.  Firstly, because it’s difficult to accurately estimate the final cost of a software project, and they don’t really have confidence in their estimates.  Secondly, because taking the appropriate steps to ensure good protection against risk can be expensive, and can significantly raise the price of the estimate.  Thirdly, because the estimate is often competing against other T&M (made up) numbers, and quoting the premium for Fixed Bid terms often means the contractor loses the bidding war before it starts.  As a result, lots of deals are signed on T&M terms, and the App Store is littered with the carcasses of T&M projects gone awry, not to mention the mile-high pile of projects that never even made it that far.

Unlike every other player in the market, the vast majority of our business is Fixed Bid, and we have very few T&M contracts only in unusual circumstances.  We have tremendous confidence in the accuracy of our estimates and we’re not afraid to stick to them.  At the same time, our quotes tend to be a lot higher than the wishful optimism of T&M “estimates”.  But that’s the premium you pay to know with certainty how much your project is going to cost instead of being surprised part way through development.

When you’re comparing quotes, make sure that you’re comparing them on equal footing.  T&M estimates aren’t worth the paper that they’re printed on.  Ask the developer what percentage of its projects were delivered within the original budget.  After dozens of Fixed Bid projects, our track record is still 100%.

It should cost much less!

Sometimes it does.  Certain applications may simply not need certain components.  Coke or McDonalds may not need a PR campaign to launch an app–their existing brand recognition may be enough.  Highly technical apps designed for engineers may not require much in the way of design work–accurate technical bits may be enough.  Apps that rely heavily on off-the-shelf components may not require much QA.  And so on.

You may elect to learn some of the disciplines and do them yourself.  We’ve had professional graphic designers, marketers, and even other developers as clients, and so in these situations it’s not uncommon for a development firm to back off and have the client take more responsibility for the project–reducing the cost.  But if you’re not a professional, taking on responsibility can be a costly move when measured in terms of lost sales.

Another tactic is to aggressively cut scope to reduce costs.  It may be possible to cut features to get a Client/Server Application into the Standalone category, or to significantly reduce the complexity of the server tier by cutting features.  By reducing an application to a mere prototype, it may be possible to produce something tangible that can be used to acquire funds to develop it further.

There is also a certain amount of variation in developer costs.  For instance, we have no sales team to pay, and since we rely primarily on the strength of our work for business, our marketing budget is close to zero.  So our costs tend to be lower than shops with an army of sales staff and billboards on both sides of the street.  On the other hand, our estimates tend to be more accurate than the competition, and since projects are universally underestimated, sometimes realistic numbers can be higher than what others quote.  These effects can cancel each other out.

But far too often, instead of responsible scope cuts and other cost-cutting measures, corners are cut instead.  Software may be outsourced and developed poorly, QA may be forgotten entirely, design may be sent to the back of the bus, and so on.  In the rush to ship some product, it’s easy to lose track of what the value of that product is.

Suppose you are trying to build a strip mall in a populated area.  You find a plot of land that is next to a high-traffic road, do some research to determine that a strip mall could be reasonably successful, and start talking to architects and contractors.  They tell you that a typical strip mall is constructed for in the neighborhood of $5M.  Unfortunately, you only have $20,000 for the project.  Undeterred, you start trying to cobble together a team of temporarily-unemployed electricians on Craigslist, some freelance architects in India, and since the budget for materials is tight, you decide to construct it from corrugated cardboard.  It does not take a genius to work out that this will not be a successful project.

In fact, there are many successful avenues open to you–with $20,000 you could build a furnished treehouse or shed, you can give your home a new room or a remodel–these are realistic construction projects within your budget.  These projects have the advantage that you are not simply throwing your $20,000 away.

The same common sense does not seem to extend to iOS projects.  Undeterred by the true cost of building that strip mall, many people are trying to piece their apps together out of corrugated cardboard.  But unlike the construction industry, where there may be building codes, inspections, city ordinances, standards of practice, and professional organizations to protect people from scams and swindlers, there is unfortunately no equivalent for software development.  It’s the closest thing there is to a completely deregulated market.  And it’s the perfect condition for a lot of companies to come out of the woodwork to say what people want to hear–that with this industrial strength corrugated cardboard, you too can own a strip mall!  It’s a sales pitch older than Amway.

With good advice from the right developer, you can figure out what is possible within your budget and execute a sensible plan instead of an insensible one.

Wrapping Things Up

I hope this series has shed some light into the time, effort, and cost it takes to launch a successful iPhone or iPad app.  I’ve barely scratched the surface of what goes into project costs, even in a massive three-part blog post!

There are many factors to consider when thinking about any software development project, and every situation is unique.  If you’re interested in talking to a real developer about a realistic way forward for your project, get in touch.

Development

Any iPhone or iPad app is going to involve developers, people who write the software code.  Depending on the firm, these people may be called software developers, programmers, or engineers.

For many apps, the developers are the people who make or break your software.  Good developers can mean applications that run quickly, are powerful, and generally behave as expected.  Bad developers mean buggy, crash-ridden software that’s behind schedule.

An entry-level salary for a software developer in the United States is $50-60k.  Developers with several years of experience are often paid $80-90k, depending on location and workload.  Very simple applications (at the low end of the “Standalone App” category) may require a few weeks of developer time, more complex applications may  require six months and several full-time developers.  Considering an average time of two months and two developers, you’re looking at a cost of $30k in developer time.

It’s important to meet the actual people who will be responsible for developing your application.  Many firms are much larger than we are in sales staff VPs, managers, etc., but have about the same number of developers!  Others may simply be a friendly sales force for a low-quality overseas development shop.  Don’t let anyone pull a bait and switch! It’s not enough just to evaluate the sales guys for compatibility with your project–they’re paid to be compatible.  Make sure that you talk to the people who will actually be responsible for implementing the project, and ensure that they share your vision.

What about outsourcing?  Can’t I cut costs there?

I’ve written about outsourcing in the past.  Like everything in life, you get what you pay for.  There are wonderful overseas development shops that charge rates competitive with US developers.  If you are paying below market, you are asking for trouble.  Far too much of our work is fixing botched outsourcing jobs that could have been done right the first time at much less cost than it will be patch it up, both in terms of development time and effort as well as lost customer trust as a result of a buggy application.

QA

A good QA department can catch bugs and problems in-house before they escape in a test build, or worse, to your customers.  Every project needs a solid second pair of eyes to spot programming problems and squash bugs.  Good QA engineers bill $50-75 an hour, and you probably want one hour of QA for every three hours of development to spot bugs.

Design

In addition to development work, you need designers, people who lay out the user interface, produce artwork, and do other graphical design work.  It’s possible to have a good developer that drifts into some light UX design, or a good designer that drifts into light frontend development, and so you may find slightly different divisions of work across different types of firms, depending on the skillset of the individual people.  Smaller application developers tend to work with outside designers or have a few contractors “on-call”, whereas larger firms can have entire graphics and UX design departments in-house.  Larger corporate clients may already have an in-house design department that they’d like to use, so the background of the development firm and the makeup of their clients can have an impact on the design skill.

Again, it’s important to communicate and share your vision to the design team.  It’s also important for the technical and design team to have good rapport and to get along well.  Animations or dynamic visual effects may have complex technical and design requirements that require both skillsets to move in lockstep, and the last thing you want is for artwork to be produced or a technical solution to be chosen that is incompatible with the other!  It’s usually important to ensure the artists or UX designers have a good track record of designing for mobile software (as opposed to graphic design, web design, etc.)

Senior UX designers with some development experience are typically worth about $80k/year.  Graphic designers with more limited experience generally earn about $55k/year.  With an average figure of a designer with medium experience for a month, you are looking at about $7k for typical design and artwork production.

Management

Someone needs to coordinate and oversee project development.  At small firms like ours, senior developers may take this role and oversee and coordinate development, design, and the other parties in the dance.  At larger firms, project management may be a dedicated department with a sizeable staff.  Software development managers typically make about $120,000/yr, but usually oversee multiple projects at once, so the actual cost of that person for a single project may be closer to $5k-10k for shorter projects.

Marketing

A successful app launch often requires an extensive marketing campaign.  In today’s market, it is difficult to stand out from the crowd.  Retaining a PR firm can run anywhere from $5,000-$25,000/month or more.  Some PR freelancers bill in the neighborhood of $100/hr for basic press release and whitepaper writing if you want more of an “a la carte” public relations approach, but the economics of the app store often dictate doing an all-or-nothing marketing push over an incremental marketing model.

Tools & Technology

In addition to personnel, it takes specialized tools to create mobile applications.  Software has to be tested on a regular basis on every conceivable device, and so many firms stock up on dozens of test devices in multiple configurations.  Professional developers wear out computers far faster than consumers do, and so it’s not uncommon to budget $10k in new computing equipment per developer per year.

Then you need to factor in software licensing and costs.  The advent of open-source software has reduced these somewhat, but custom tools are still a requirement.  Professional-class bug trackers typically run about $5,000/year for a site license.  Source control hosting, testing software, database editing tools, Photoshop, and other tools of the trade have to be licensed.

Then you must factor in custom frameworks or components to be licensed for use in applications.  Because we’re a very focused company with almost all of our work in iOS, we’ve invested in building an incredible volume of pre-built libraries that are more readily adapted for iOS use (see our articles on CrashBuddy and LogBuddy), and so we tend to have a reduced licensing burden than other firms, but this comes at the cost of our developers’ time to write and maintain those libraries.

Then you have your typical overhead: office space and furniture might be $1000/month/person in a midsized city.  IT, servers, backups, infrastructure, hosting, e-mail, and internet access might run another $250/month/person.  Benefits, taxes, and payroll probably add about 15% to the cost of every contract.

The Final Bill

Consider a fairly simple application that is on the books for two months.  The cost structure might look like this:

• Two developer salaries, two months: $25,000
• One designer salary, one month: $7,000
• Tech lead / project manager : $5,000
• QA: $6,000
• Office space, computers, equipment (amortized): $8,000
• Software, licensing (amortized): $1,500
• Benefits, taxes, payroll: $8000

The final cost to develop our simple application (leaving no profit for the firm) in this case is $60,500.
It’s important to emphasize that these are typical figures for an application of arbitrary size.  Some apps have more or less development costs, more or less design costs, etc.  As discussed in the first part of this article, there are many different types of applications that may require more or less responsibilities from each of the necessary disciplines.

Also factoring into the price, different firms tend to emphasize different disciplines.  We are a developer-driven company, where software developers take more of an active role.  There are talented design-run or management-run firms, in which those skill-sets tend to be more emphasized and tend to extend further into territory traditionally controlled by other disciplines.

In addition, however, there are a few operating costs.  Considering a 12-month application life (many applications run shorter or longer), you may be looking at:

• Hosting costs: $500/month for 12 months
• Support contract: $500/month for 12 months
• PR/marketing: $5000/month for 3 months

e.g. an additional $27,000 in the first year.
So how does our $87,500 number for a simple application stack up?  Chockenberry, the developer of the award-winning Twitterific application estimated their total cost at $250,000.  Twitterific is a design-heavy app in the “Pre-existing API” category, which is much more complex than our simple application.  Another data point is the Barack Obama application, which reportedly cost $50-150k, also in basic agreement with our number.

Was this a lot more than you were expecting?  Continue on to Part III!

Myth #1: Outsourcing Is Cheap

Consider this scenario: a potential client will come to us and ask us for a quote, which we gladly provide.  The client discovers that an overseas firm is quoting half our price, and so they select the overseas firm.  Six months later, they ring us up and beg us to do the project for our original number, because the firm they selected has already billed them triple and they have a horrible, buggy, half-fininshed product to show for it.

Unfortunately this scenario is very common, and we “rescue” several dozen projects a year from overseas vendors.  In many cases, the quality of the product is so terrible that we have to throw it in the garbage and start over from scratch. Many unscrupulous vendors will tell you anything to get the contract signed. If it sounds too good to be true–it probably is.  Don’t listen to anyone who says that your app can be built for a few grand and a prayer–unless you have a few grand to throw away.

A good rule of thumb is that a good developer will dig deep into your project requirements, will ask for a specification or will insist on developing one, and will ask questions and make suggestions that demonstrate a very deep understanding of your project’s design and goals.  Anyone who quotes you an exact number without first receiving and asking questions about your detailed specification is someone who has no idea what you want and is making up a number to get you to sign the contract.  Don’t fall for that trap.

Myth #2: Outsourcing Is Easy

When you select a professional, local iPhone developer or contractor, you get someone who understands your business problem and is someone who can get things done.  You get someone who keeps you up to date on your projects’ progress, who doesn’t need to be babysat or micromanaged, and who you can rely on to be an excellent advisor and partner in designing and developing your project.

When you outsource a project to a bottom-of-the-barrel software vendor, you get a different kind of service.  Language and cultural barriers conspire to prevent you from communicating about problems and issues effectively.  Work may be performed based on a bad specification (or even no specification!).  Issues and errors constantly have to be corrected, and the client often develops a severe mistrust for the vendor.  Cheap developers skimp on code quality, failing to prevent bugs, and even critical components like verifying that the software basically works for its purpose.  Working with a cheap developer means constantly listing and re-listing, e-mailing and re-emailing about every single problem, managing and micromanaging each individual task, and continually worrying about whether the vendor will ever get it right.   Do you want to stay up late at night wondering if your project is six months or six years behind schedule, or do you want a vendor that you can count on to get it right?

Myth #3: All Programmers Are About The Same

Paul Graham, an angel investor who makes his money picking software companies wrote that “A great programmer might be ten or a hundred times as productive as an ordinary one”.  Not to mention how many times an average programmer might be better than a bad one.  The variation in skill between programmers is enormous.  A good programmer can get your project done ten times faster and five times better than a bad one–if the bad one can even get the project done at all.

Myth #4: My Project Is Really Simple

There are no simple iPhone or iPad projects.  Each project requires a unique expertise and attention to detail.  The iPhone and iPad are some of the only platforms remaining that require the rare and esoteric skill of manual memory management, which often makes up well over 50% of the complexity of any iOS project.  Even many good developers do not have experience in this area and as a result iOS development is one of the most difficult types of software development today.

Not to mention the special challenges of performance on a low-power mobile device, battery life, navigating Apple review, limited memory, and testing and diagnosing bugs across many different devices.  Creating an app that is fast and easy to use requires a lot of experience.  Make sure you pick a vendor who is prepared to meet the unique challenges that your app will present.

Myth #5: This Cheap Vendor Has Many Success Stories

You only see the success stories.  IBM claims that over 80% of software projects fail because they were “over budget, late, missing function, or a combination”.  At DrewCrawfordApps, we have a proven track record of delivering custom software right, on-time and on-budget.

Large outsourcing clearinghouses have incredibly high turnover.  Smaller shops are bought and sold, and your project may pass through any number of developers’ hands.  How do you know the people working on your project are the same people who built the company’s portfolio?  You don’t.  How do you get in touch with the developers who actually work on your app?  If you’re lucky, you can e-mail a person who can e-mail a person who can talk to the developers actually responsible for doing the work.

At DrewCrawfordApps, I personally oversee all our projects.  Our clients all have my cell phone number and can always talk to myself or another senior engineer directly who can solve their problem or give them a status update.  We have no slimy sales people, no telephone menus, nobody to screen our calls, and nothing to prevent you from getting any concerns or anxieties you may have dealt with immediately.  Since I personally handle all our client aquisitions and proposals, I can answer any questions you have and because I’m a developer, not a sales guy, I’m not afraid to tell you if for some reason we’re not the right fit for your project.

Choosing the right development partner is the most important decision in ensuring your project’s success.  So before you outsource that project, think about the hidden costs.  Doesn’t your app deserve a professional developer?

Contact us today to discuss your projects’ unique needs.  But even if we’re not the right fit for your project, we’d encourage you to choose a local professional developer, or even an overseas developer of comparable quality and cost.  But we’ve rescued enough botched projects to know that if you solicit quotes from a cheap shop you’re asking for trouble.