A platform is essentially an Erector set with very poor instructions. Sure -- go ahead and build a motorized crane. If you a. know what a crane is and should do and b. know what parts need to be assembled in what order to do it. Otherwise, you’re just trying stuff.  

Your first few cranes won’t work very well. People will likely use something (anything) else they can until you have built a better crane. You have on your hands a Science Project.

This means that IT departments spend years designing requirements, user interfaces, information architectures, etcetera, for everything from content management to HR to marketing tools. Which is fine, if you are a content management, HR or marketing company. But if you happen to sell services, or other digital or physical products, it is unlikely that you want to invest your IT resources in reinventing disciplines that are not core to your business.

There's an alternative to the science project: turnkey solutions. Turnkey software or technology is that which you can buy as a product and it just works. Increasingly, turnkey SaaS offerings solve enterprise problems out of the box. Billing, email, intranets, marketing automation, HR management, even desktop software. Depending on your organization, your expertise and your strategy, this could be just about all of your software or it could be all of your technology. Historically, not much enterprise software has been sold this way. Ironically, Microsoft Office and Outlook are probably the most pervasive and effective of these.

If you believe that your company’s marketing, HR or whatever expertise is so advanced as to give you an advantage -- then a science experiment could be a good investment. Consider Zappos customer service platform (note that I have no idea what technology they use, but I assume it is a. pretty sweet and b. something they invest in). It isn’t what they sell, but it is how they profit.

But if you are like most companies that sell sneakers or software or whatever, IT science projects are far too common and far too costly. They frustrate nearly everyone involved. So why oh why oh why are they so common?

Why IT Buys Science Projects

1. Requirements blindness

In general someone from IT will be tasked with identifying requirements. That person will generally have a strong bias toward technical requirements. It is rare for the person to have either deep enough insight or collaboration with the business to create a set of requirements that solve a business problem in a way that helps users be successful. Yet they never miss security, API, obscure customization, or other line item tech requirements. Often an organization will come up with a design, deem it perfect, and lose their ability to see the forest for the trees. In other words, they become so focused on the perfection of their own solution, that they have a hard time considering meaningful alternatives.

2. The desire to remain “flexible”

The future is increasingly difficult to predict, and IT departments are trying to protect the best interests of their companies by ensuring that there is maximum flexibility. But this is a double-edged sword. Many years ago, a friend remarked on another friend’s infant “look how incredibly flexible he is.” Answered another “well, yes, but he also can’t walk.”

Look for the right kind of flexibility – not we can build anything you need as long as it looks kinda like SharePoint, but how do we integrate more sources and destinations of information going forward? How rapidly can upgrades be applied, or – and this is often a radically overlooked issue – how much am I losing by taking 12-18 months to deploy in the first place?

3. An outdated enterprise software model

For the better part of 25 years, enterprise software has been very expensive and very complicated. Both vendors and IT departments are organized around these assumptions. Fortunately more software is available these days that makes this model unnecessary, by running out of the box, or with only minor integration efforts.

The Troubles Science Experiments Cause

1. Bad outcomes

Bad UI, poor information architecture and navigation, and/or a mismatch with business needs, leads to software that is not used, and not used as intended.

2. Desperately long deployment times

From requirements analysis to planning and resourcing to waiting for free resources, development, debugging and deployment, can be over a year -- in a good environment. Then you might need an upgrade or maintenance, and it starts again. Endlessly.

3. Upgrade purgatory

Once you’ve customized and platformized the heck out of something, any changes -- whether they are feature requests, UI modernization or upgrades, become increasingly difficult. You will be out of date. Plan on it.

4. Less business expertise

A great piece of technology embodies expertise. It doesn’t just throw features at you, it is thoughtfully designed to arrange a smart collection of features in such a way that it works for you, building your expertise rather than demanding it. Put another way: I do not want marketing automation software or collaboration software that says -- Hey, here’s 1500 features, pick what you want and configure it into the perfect solution for exactly your business.

I want someone to say, Hey – we’ve put a ton of research and experience and effort into understanding this problem. We have 25 features and workflows that will solve the vast majority of your problem far better than what you could have hacked together, and it's well tested. We also have some simple ways to extend and integrate so that you don’t need to worry about the core of the problem and can focus on the best way to adjust it to your particular business problem.

Perhaps for a medial clinic there needs to be a way to connect to x-ray results or medical IDs. A trucking company might want to reference weather maps and routing options. Every business has different requirements -- but that doesn’t mean they should start from scratch.

The Beauty of a Canned Solution

More and more business software is available as cloud-based and turnkey solutions.

Yes, you give up something in flexibility, but you gain in design quality, ease of deployment and maintenance, upgradability. The convenience and quality more than compensates for flexibility -- for MOST problems for MOST companies.

Strategic technology may or may not be available this way: it depends on your specific circumstances. A quality turnkey product gives you a specific solution that draws on deep experience to give you exactly what you need to get started and grow. No need to reinvent the wheel.

Solution or Science Project?

In those cases where a certain technology, process or approach is a core differentiator for your organization, an enterprise science project is a very good investment. For those things (like an intranet or CMS or HR or CRM) which you need, but don’t need to reinvent, your best investment is likely in the highest quality cloud solution you can find.

So -- which do you need? Turnkey or Science project?

2014-11-Sept-Lavoy-Image1.jpg
The best choice -- turnkey or science project

If you have time and the required expertise and you believe the project is a strategic agent of growth and differentiation, you want a custom science project with which you can reinvent your industry.

If your project is urgent and time sensitive -- regardless of your level of expertise, regardless of the strategic nature of the product -- you will most likely do better with a stock solution than you will from initiating a science project now. You can always build on or replace the stock solution with a nascent science project down the road.

Science projects can yield great results. They are the only things that bring us light bulbs, airplanes, engines and medicine. They are among the cornerstones of innovation. But you can be a magnificently innovative artist without mixing your own pigments or a chef who does not have a farm.

Editor's Note: This is the third and final installment in a three part series on our infatuation with technology. Be sure to read the first and second posts in the series.

Title image by Image Editor (Flickr) via a CC BY 2.0 license