4 WAYS TO FAIL SOFTWARE DEVELOPMENT PROJECT

Kasya Shahovskaya
5 min readDec 28, 2015

I. Why projects fail?

The state of Oregon has recently canceled work on its massive online health exchange program and switched of to the federal government system but not before paying $134M to Oracle for its years of work on this failed project.

Study by Stanford University found that nearly 2 out of every 3 information technology projects in USA fall short or fail in time.

You may think that big companies are much more effective in development processes because they are more experienced and have big budgets for hiring best teams. But big companies fail too.

A $170M project for McDonald’s was completely abandoned in 2013, while a $600M a luggage handling system for the Denver Airport that never worked was canceled in 2014. It seems like an epic fail, but wait for the next example:

A project for the London Stock Exchange that was originally budgeted at £6M was canceled after 10 years and £800M without even touching the ground of its operation.

Look like organising a better way of managing IT projects is not a trivial exercise. There are many external factors which affect any project like market change, suppliers, competition, even political and country risks. But today lets talk about internal factors, such as inter-company issues, policies and workflow in organisation which affects development process.

II. Which internal factors affect software development productivity?

1. Top management does not fully know which product they want to receive. And instead of going to the market research and find out what CLIENTS want, top management base their request on their own thoughts. Than after some time “BIG IDEA” in managements’ mind receives some changes or new features. IT team in this case have to make big changes in the product or start it again and again. IT people are usually very logic and straightforward, so they frequently lost in frustration trying to understand what management want from them. That tremendously affect time of the development and its budget.

2. IT team rules development process. Yes, its still happening in many companies when IT people decide on business side of the product, choose features, approve designs and development plan. In this case company frequently is ending up with the project which it can’t sell or use. It’s often the engineers and designers that are among the most celebrated. They are the people who literally build amazing products and services that we enjoy. But it’s not magic that helps guide them to take an idea from start to finish — that heavy responsibility rests on the shoulders of product managers (PM). And many companies don’t have PMs in the team at all.

3. Top management controls every step. Market changes every moment, if project lead have to approve any changes he want to test with the top management, than company will never succeed in development process. There is no need in the every day meetings and detailed weekly reports.

CEO and top management should be involved into the discussion only at strategically significant moments to obtain feedback and support. Ironically, by exercising his power to give orders, the CEO actually reduces his real power, saps his energy and his organisations’, and slows down progress.

Company should hire professionals, give them power and budget and leave them alone without telling what they should do.

4. Bureaucracy. Especially important for big companies. The longer decision making process is the harder to create new product for them. That’s why its so hard for a big firm to innovate.

III. How to do it right?

1. Hire a Product Manager for analysis of the business side of the product. The will be the bridge between product manager and It team. For the small companies PM may also serve a role of enterprise architect (EA) managing the organisation’s strategy, processes, information, and information technology assets. For the big companies its better to have EA as a separate role.

The role of the enterprise architect is to take this knowledge and ensure that the business and IT are in alignment. The enterprise architect links the business mission, strategy, and processes of an organisation to its IT strategy, and documents this using multiple architectural models or views that show how the current and future needs of an organisation will be met in an efficient, sustainable, agile, and adaptable manner. They aren’t the direct supervisors of the engineer or designer and can’t fire anyone for not following through, and focused on the success of the product’s mission.

2. Make sure you have an IT lead, who can serve a role of Software Architect and manage software development process.

3. Make sure you have “lean” philosophy of your development process. Most of the projects should grow from a prototypes via testing to the beta version and than to alfa version. Doing software project in one leap is a waste of time and money. Testing results with ended user is curtail for project development. Ideally, project needs to grow through a one-two weeks duration slaps with eternal testing. We in our company have nice principle “test before invest”, which means that we are not investing large amount of money before new projects’ business model will show it’s sustainability (or at least first sales will happen).

Big companies should consider their projects as a startups and follow the way of startup development — with constant changes, market testing and daily evaluation of results.

The shift I am proposing is from traditional, top-down, control-oriented management, to self-organizing, customer-driven, so-called Agile Innovation for the whole company and new product management. This is, of course, nothing less than a revolution. And the principles of it are described in a great book “Agile Innovation”. I strongly recommend it. Facts on failed companies in this article are coming from this book too.

PS. Buy any book people ask for

Any company should have a good library. Thats the way people learn. Most great developers I know read a lot. There is no way to force people read books, but at least it should be extremely easy to go and grab a good book to read.

Thank you for reading!

Author: Kasya Shahovskaya

@: k.shahovskaya-at-gmail.com linkedin.com/in/kasya

--

--

Kasya Shahovskaya

Digital anthropology researcher, thinker, strategist, MBA, MSc Computer Science. Lived in 7 countries, now in 🇩🇪 www.inspiremetofly.com