Scores of companies are shifting their technology infrastructure to the cloud. Rather than bear the burden of managing their own data centers, they’ve handed the reigns over to experts in the domain of computing and data storage – Amazon Web Services, Microsoft Azure, Google Cloud Platform, and other cloud infrastructure as a service (IaaS) providers. These cloud adopters are enjoying greater agility, cost-savings, and system reliability as a result.
After seeing the benefits of cloud infrastructure, many are considering cloud solutions for their integration processes.
Traditionally, IT professionals use a blend of legacy solutions, point-to-point tools, and custom hand-coding to connect data, applications, and APIs. But modern cloud integration platforms have sprung up, claiming superiority over these traditional methods.
As the cloud integration market – formally dubbed the integration platform as a service (iPaaS) market – grows more crowded, IT professionals need criteria for evaluating these systems.
Here are four quick tips, in the form of questions, for finding the right cloud integration platform for your business.
1. What do industry analysts say?
It’s not feasible to vet every iPaaS on the planet. A proven way to whittle down your list of possible vendors is to focus on solutions that Gartner, Forrester, and other industry analysts recognize as leaders.
You may have heard the adage, “no one’s going to fire you for purchasing software from the Leaders quadrant.” While I wouldn’t stake your job security on this assertion, there is some truth to it. Integration projects can be big investments. If you pick a solution that is not highly regarded in the marketplace, and your project fails, then it will invite the kind of attention you don’t want from executives at your organization.
In some cases, lesser-known solutions can be a fine match for your specific integration project. But you still run a higher risk of that vendor potentially going out of business or struggling to scale with you.
Thus, a safer approach is to spend most of your energy on evaluating vendors in the Leaders quadrants.
2. How much of my disparate data can the platform integrate today and in the future?
Most organizations have reams of data living in cloud applications, on-premises databases, and other locations. Moreover, that data spans a wide range of formats and needs to move at varying speeds. Finding a cloud integration platform that supports your skyrocketing volume, variety, and velocity of data will pay off big in the long run.
For context, some platforms struggle to perform hybrid integrations (i.e., connecting on-premises and cloud endpoints). Meanwhile, other platforms excel at conventional ETL (extract, transform, load) operations, wherein large volumes of data are hauled into a data warehouse via batch processing. But they falter when it comes to integrating modern cloud systems like Workday, Salesforce, and Snowflake.
Ideally, you’d find a solution that handles a broad range of use cases. When gauging the breadth of a platform’s capabilities, ask questions like – does it have sophisticated pre-built connectors for data sources that are critical to my business (e.g., Zuora, Netsuite, and Oracle)? Does it only do batch processing, or can it effectively do low-latency processing too? To what extent does it automate the process of transforming data? And so on.
The takeaway here is – the more use cases an integration platform supports, the less custom coding you have to do.
3. How long will it take to build integrations on average?
Perhaps you’re exploring cloud integration platforms because you want to improve data security, integrity, and governance. But if it takes you months to build a single integration, you’re left with a different set of problems. When highly trained developers, engineers, and data architects are detained by middleware, it means they can’t sink their teeth into strategic projects.
Moreover, integration is a means to an end, is it not? You don’t move, manipulate, and merge data just for the sake of doing it. Rather, you integrate data so you can perform analytics, gain a 360-degree view of customers, streamline critical processes, and so on.
As such, it’s important to balance governance with “speed of development” when evaluating integration solutions. For each platform, probe into how long it will take you on average to build integrations. Also, explore whether you can reuse certain workflows and pipelines.
As with any software you purchase, you want the integration platform you choose to accelerate the time-to-value as much as possible. The speed of development plays a big role in this.
4. What is the total cost of ownership?
If someone is selling a car that appears in good condition, but they’re only charging $200 for it, most people would think – what’s wrong with the car, and how much could those problems cost me later?
You’re obviously not going to find integration software going for $200, but you might encounter pricing that is oddly low compared to other vendors. In such cases, it’s wise to think through not only the upfront cost but also the costs you might incur later as a result of using one platform over another – i.e., the total cost of ownership (TCO). Frankly, it’s a best practice to estimate the TCO whether the initial price is low or high.
These are the sorts of questions to think through when gauging the TCO.
How many full-time employees will be required to build integrations? What’s the cost of training, and how long will it take? Will I need to hire hardcore developers to manage the platform, or can I entrust it to those with a lighter technical background?
What’s the cost of an integration breaking, particularly if it affects customers? Moreover, what’s the cost of pulling engineers away from strategic projects to put out an integration fire?
How well will the platform adapt as we acquire larger volumes of diverse data? Will it scale with us as we grow, or will we need to replace it in five years? In such a case, what are the costs of finding another solution and then re-training our engineers?
It’s tough to both predict problems that may arise and then quantify the subsequent costs. But going through the exercise of estimating the TCO of a cloud integration platform can help you make a prudent decision that bears fruit in both the short- and long-term.
There are obviously other variables to consider when evaluating a cloud integration platform besides those featured in this article. Some factors will be more vital than others based on the distinct requirements of your organization. But paying attention to analyst rankings, product breadth, speed of integration development, and TCO in the early stages will boost your chances of choosing a platform that yields a massive return on investment.
Speaking of ROI, SnapLogic is hosted a webcast, “ROI Bake Off: Integration Platform as a Service (iPaaS) vs. Custom Hand-Coding,”. I, along with research firm Amalgam Insights and a speaker from Groupon, teach you how to quantify the benefits of modern enterprise integration software compared to that of hand-coding integrations. Watch now!