Did you know that 52% of the Fortune 500 companies have disappeared in the past 15 years? Why? Life expectancy for the enterprise is getting shorter. In 1955 the average life expectancy was 75 years. In 2015 is 15 years. Why these changes? Because the world has evolved and transformation happens at a much faster speed than 50 years ago or even 20 years ago. And if enterprises don’t adapt, they simply vanish.
In order to survive enterprises are adopting a faster pace, changing not only their culture but also their processes and business models. It is paramount to respond fast to market dynamics. That is why terms like Agile development, continuous integration, continuous deployment, and DevOps are becoming standard business terminology in the enterprise world. Software today is not an enabler of the corporate strategy; rather it drives the strategy, keeping the company competitive and making new revenue streams possible. Companies are forced to speed up their application development and are thus moving to fully automated, DevOps flows.
But there is one problem. Legacy software solutions that have traditionally been used to support software development processes cannot adequately adapt to the new paradigm. They are often cumbersome, complex products that do not appeal to the modern developer or DevOps people. They actually tend to hinder the transition to DevOps rather than facilitate adoption of new models and acceleration of release cycles.
And here are four reasons why:
1. Lack of efficiency.
Legacy solutions are very time consuming and actually contribute to slowing down the cycle from code change to production. This is true for both initial onboarding and regular usage. For onboarding, legacy solutions often require professional services and specialized consultants to make them work. Besides adding more costs, this approach increases time-to-value tremendously and also reduces flexibility should the tool be applied to new applications and new groups within the company. Think of the typical phases of IT purchasing in a traditional IT environment: purchase, installation, customization and finally usage. Why would you do that in a world where cutting edge SaaS solutions are readily accessible via online signup for a few hundred dollars per month?
For regular usage, the efficiency gains with best-in-class cloud services are even more dramatic. Think of the time taken, for example, in creating a test script: the coding, the validation, the provisioning of resources, the installation in the test solution. We have seen companies who regularly spend 3 days to create and run a new test with already trained testers, whereas it took them 15-20 minutes to accomplish the same task with best in class cloud services like Nouvola.
2. The suite is dead.
Legacy software solutions rely on the concept of the ‘integrated suite’, a fully integrated set of tools that cover what was called the Application LifeCycle Management (ALM) and include everything from language compilers to source version control to testing management solutions. The pre-built integration was a major selling point at a time when integration between different vendor solutions was basically impossible. The suite also provided significant benefits as interoperability, common look and feel and consistent architecture. In the 1990s, these were real benefits. But today the suite is dead. Large, monolithic, heavily customized, proprietary solutions are relegated to the status of legacy in many enterprises, following the same fate as ERP solutions. The suite is being deconstructed and new models have emerged. Most development teams are rapidly adopting a tool stack of many different best in class cloud services from different vendors, which are easily integrated due to the availability of rich APIs.
3. The Total Cost of Ownership (TCO) is too high.
Legacy solutions tend to be very expensive to use and to maintain. It’s not just the cost of the initial licenses that needs to be considered. It’s also the cost of the necessary infrastructure, plus the cost of the dedicated personnel to run them, plus the cost of maintenance. The typical business model for legacy solutions is the ‘perpetual, non-exclusive license’, bundled with maintenance fee typically priced at 20% of the license. It all adds up pretty quickly. Legacy solutions often have more issues; sometimes there are hard to find patches, or there have been several customizations that make it impossible to upgrade without rewriting code or involving highly paid consultants again.
In addition, there is an opportunity cost to consider. The amount of money spent in legacy solutions reduces the budget that could be invested into innovation. An example: A recent report by the U.S. Government Accountability Office (GAO) found that of the total technology budget of more than $78 billion earmarked for the fiscal year 2015, 26 federal agencies spent a total of $60 billion on legacy investments. This is echoed by many enterprises in the private sector who say ¾ of their budget today goes into legacy solutions, and they are looking to move away from that as it prevents them from investing in innovation to gain a competitive advantage.
Let’s take as an example the cost of setting up and maintaining a testing solution, where it is necessary to install and maintain the servers and upgrade when there are changes. And if there are issues, it might take months before a patch is delivered.
For natively cloud-based service, there would be no need to acquire servers and infrastructure, install, and maintain the software. And if there is an issue, as soon as the fix is ready, it is immediately made available with no extra cost to the customer.
4. Capabilities are missing.
Interestingly enough, most legacy solutions are also lacking features and capabilities that are critical for software development today. The cloud, IoT and new innovative technologies, not only make it easier to achieve the same results, but often offer capabilities that were simply not available in the past.
New best-in-class cloud services usually offer great user experience and are super simple to use, as they are modeled on consumer-internet standards. They have better access to more sources of relevant data, usually because they are narrowly focused on solving a specific problem — see suite comment above. They offer integrated analytics, making data easily consumable and readily actionable. They have built-in scalability by making smart usage of cloud resources. And they are API-centric, making integration with other cloud services from different vendors not only a possibility but a straightforward endeavor.
Think for example of the ability of running tests from different geographic locations in the world, external to the enterprise corporate network. In the past it would have been extremely costly, if not impossible to achieve that. Today, it can be accomplished within minutes.
As the enterprise is under tremendous pressure to evolve and adopt sophisticated software operations, it finds itself abandoning legacy software solutions and adopting best in class cloud services. Specifically, the new innovative solutions that are rapidly adopted by many companies across multiple verticals are natively cloud-based, simple and elegant, turnkey and self-service, consumer Internet-like, and API-centric. This is why growth has slow down to a crawl for the top 5 Enterprise IT SW vendors, and we are seeing tremendous growth and increasing valuations for new players in the industry. Although different companies may be at a different point in this transition, there is no doubt that a paradigm shift is underway, and the new paradigm is dauntingly different from what IT enterprise software looked in the past.
Paola Rossaro is co-founder and CTO of Nouvola, a nextgen performance testing solution. You can also find her on Twitter at @prossaro.