Three trends in business app development you need to be aware of

Three trends in business app development you need to be aware of

Summary: Disruption! Transformation! The future is upon us, application architects.

This isn't really what business applications look like, but let's go with it. (Illustration: Andrey Prokhorov/iStockPhoto)

Listen up, enterprise architects or anyone responsible for application delivery in their company: there are trends afoot. 

Yes, trends.

Ten of them, in fact.

In all seriousness, market research firm Forrester published a new report on Monday outlining its predictions for how business application deployment will change in the near future. It's worth a read if this is a major part of your job.

Three of our favorite trends (there are 10 total):

Cloud deployment models are changing the economics of applications. "Traditional on-premises applications are tapped out," the report's authors write. "Software upgrades have become so costly and difficult that most customers defer them for years, sometimes even for a decade." And it only gets worse with customization. In the future? Software-as-a-service (or "SaaS") models, of course, which offer a flexibility, stability and scalability (how's that for -ilities?) that the on-prem model does not. Adoption will vary by industry—financial institutions will remain hesistant, for example, given security concerns—but it won't be long before most enterprises have as-a-service in the mix. (If not already.)

User experience is improving. "Early business apps captured data using bland, character-based screens with rows of input fields in systems such as teller workstations in early core banking systems or in HR or retail apps," the report's authors write. Colors, drop-down lists, icons, and other basic features didn't help. "Millennials [will] reject experiences that fall short of the high expectations set by consumer apps, so vendors are delivering new user experiences with rich graphical features that embed analytics, decision support, offers, and direct customer interactions." In other words, with the basics down, there's more focus today on business outcome. Though the authors warn: "Rich, interactive user experiences make for great demos—but they must deliver tangible business value to be truly competitive."

Componentization supporting smart functionality. Efficiency dictates that a common system is ideal from a development and maintenance point of view, yet lines of business know that they need customization to themselves work most efficiently. The answer is componentization, which "will enable app delivery teams to blend the best elements of custom-built apps and components and off-the shelf-components for functions such as loan origination and know-your-customer apps." That even extends to smaller firms, which tend to use prepackaged business apps. "They will break today's pattern," the authors write. "This will enable the economies of scale needed for integrated, flexible, easy-to-use, and quick-to-deploy business apps." And they'll presumably be more competitive, too.

That's not all, of course—for the word on social collaboration and elastic computing and big data analytics, you'll have to read Forrester's full report.

Topics: Software Development, Web development

Andrew Nusca

About Andrew Nusca

Andrew Nusca is a former writer-editor for ZDNet and contributor to CNET. During his tenure, he was the editor of SmartPlanet, ZDNet's sister site about innovation.

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • Yet another example

    This is yet more garbage reporting about a garbage study. Most apps in the business world today are old and traditionally based on the old IBM mainframe. Enterprise apps are COBOL-based for the most part. Company's are always touting the replacement of these antiques but never really have the will power, cash or resources to make such monumental shifts. The trend in most business environments today is to re-face these old apps with a browser-based shell stick an app server or Internet server in fromnt of the old geezer app and run in this mode, all the business rules and logic are still based in these old COBOL mainframe systems. To say "Traditional on-premises applications are tapped out" is laughabale because this is what these analysts have been saying for years and although I agree that I would love to see a lot of these business applcations replaced with cookie-cutter off-the-shelf software packages - this really isn't real world or realistic at all. Most companies today will live with the "face lift" and leave all the old logic alone, some will face-lift and replace their old legacy systems in portions, they do this with a mixture of on-premise customization and package-based solutions (where it makes sense). SaaS is simply code worded for package-based software running in another location - you know the SAP's Peoplesoft's, Oracle's, etc. running in the cloud (As I have been able to witness fortunately from afar and sometimes unfortunately up close) package-based solutions will only work if your business entity/unit you are building the system for changes ALL their processes to align with the package and its capabilities. Now some businesses actually do this; they change their processes to align with the package they purchased - the off-the-shelf (aka SaaS) model works and is in tact based on this scenario. However, there are huge amounts of buisness entities that need huge amounts of customization - which means you essentially will end up re-writing the off-the-shelf package, this does not follow the SaaS model at all and you are presented with constant problems, slow to market solutiuons, poor quality, slow response and in short - an utter chaotic mess.

    If businesses want solutions that really fit their business and they want quality and speed to market, as well as adaptability, what I have witnessed is either applications are still built and constructed on-premise (they may eventually run some where else), but they are still custom built or you will have a hybrid (highly customized) off-the-shelf software. Pure SaaS or a pure off-the-shelf simply doesn't work, nor ever has. Hence the "on premise apps being tapped out" statement above is simply bogus and laughable - there is no such thing, there is nothing to back this up anywhere, and as long as businesses want to be unique and compete, the cookie cutter standardization that is being discussed here as SaaS will never work in it pure form.
    • We don't have a line of Cobol in any of our internal systems

      And even if one did, there are cloud environments that can do Cobol (like Azure.)

      Andrew certainly speaks for me. I have some very old LOB apps - in the sense that they have existed a long time - that have been kept current by regular platform review and overhauls at 5 year intervals.

      I expect the next major cycle will involve the cloud - in either a tepid or jump right in way.
      • Disagree

        You may be in the minority. There are large amounts of businesses who run legacy COBOL apps and as I already mentioned, they can certainly physically move these to a cloud center, this doesn't equate at all to SaaS. Many people have these old legacy apps. For many many years people have stated "We are moving off the mainframe", but do date they never have and most likely never will - there is simply too much invested in these old relics. These are the people that may indeed move their computing resources to a cloud center, but will never replace that legacy app with some panacea package-based SaaS solution. Not going to happen. Sorry. YOU may not have a line of COBOL code in YOUR environment, but there are literally thousands of major Fortune ### companies that do. "Moving to the cloud" doesn't mean replacing all your software assets, it could just be moving all those "Traditional on-premises applications" that are "tapped out" to another center, but they are still the same app with the same logic, the same data and the same architecture.
    • I'm quite happy you left this comment

      ...which is detailed and thorough and exactly the kind of discourse we want to see on ZDNet. Kudos.

      With that said, I don't appreciate the "garbage" comment. If you don't like Forrester's point of view, please disagree -- with vigor! But please don't shoot the messenger. (There's a reason we write, "...the authors write." Because it's not our point of view.)
    • You're speaking of some, not most.

      You shouldn't use the words 'most', as in most apps, or most companies. I've consulted at three that have unplugged their mainframes recently. I've also worked in a couple others that had programs in place to move to new modern enterprise packages. So enterprises are not using those old platforms because they want to, it's because other priorities come up. But you can only put off for so long the inevitable replacement.

      And SaaS is not running package based software in the cloud. At least not real SaaS which involves running software designed for cloud scale and multi-tennancy. There are old line software companies (e.g. Oracle) that are putting the 'cloud' marketing on old platforms, but those are the exception and mostly during this transition phase.

      An organization (or IT within) can ignore the cloud if they choose, but in 5 years time their world is going to be rocked. Because, in about 5 years and peaking around 10 years companies will no longer have dedicated data centers. The reasons to are getting fewer, while the economics will punish those that do.

      This is history repeating itself, as in the old days on the PC introduction. Expect the same types of fall out.