“I give the customer what he wants. I don’t think it’s my place to offer dietary advice. If they want red meat and boiling tar… then buon appetito,” said Tom Wambsgans, Shiv Roy’s husband and head of one of Waystar Royco’s divisions, in hit TV drama Succession.
This idea of feeding customers what they think they want regardless of the consequences is not necessarily a new one. While Wambsgans hit the nail on the head with Waystar Royco’s approach to customers, how many IT leaders can look at their mix of software-as-a-service (SaaS) apps, platforms and IT services and not feel that, somewhere along the line, they’ve been oversold?
This sums up nicely the almost Wild West approach of buying IT during the pandemic, when SaaS, for obvious reasons, had something of a growth spurt. The problem for many of these organisations is that data, which is needed for the efficient running of SaaS applications, resides in different enterprise systems. Often, the only way to overcome this is to buy a new SaaS product designed specifically to plug data gaps between enterprise systems with no application programming interfaces (APIs) in place. And so it goes on.
According to a NetApp’s 2023 cloud complexity report, released in March, 98% of senior IT leaders have been affected by increasing cloud complexity in some capacity, potentially leading to poor IT performance, loss in revenue and barriers to business growth.
In its recent report, Innovate or Fade, Accenture talked about the current “technology deficit”, which refers to “the disparity in adoption, implementation or effective use of technology (both established and leading edge) to create business value”.
Data disconnect
For Stephen Hateley, head of product and partner marketing at Digital Route, a usage data processing business that assesses SaaS companies, among others, this is typical. “Moving to the cloud is often billed as a silver bullet to beating many problems, however, the popularity of various SaaS products has led to some organisations creating disjointed internal systems and data silos,” he says.
As a result, Hateley adds, data is no longer confined to a single enterprise system, but is spread across different cloud-based platforms as well as on-premise silos. “After years of successful scaling or following a recent merger or acquisition, most businesses possess multiple data stacks and siloed systems, resulting in complex data operations,” he says.
For all the good it has done, there’s a growing sense that the rush to SaaS has, in some quarters, created a monster. Hateley agrees, saying many businesses aren’t able to fully integrate their data silos, and this is leading to missed revenue streams, development opportunities and organisational gains.
“The resulting problem means businesses are unable to communicate internally via multiple systems,” adds Hateley. This issue comes to a head when working in industries traditionally awash with data, such as financial services.
According to Craig Walter, principal solutions manager at payments platform Vitesse, an integration solution is key to overcoming the multitude of big data challenges that businesses are increasingly facing.
“The data ecosystem in the payments industry is a complex one, comprising multiple merchants and third-party players, who work across a variety of formats and structures,” says Walter. “To process that data at scale, we needed an integration solution that would allow us to digest large amounts of data in specific formats, alternating on a per-region and per-customer basis.”
Vitesse teamed up with SnapLogic, a “visionary” in Gartner’s 2022 Magic Quadrant for data integration tools. SnapLogic is an integration-platform-as-a-service (iPaaS) product, which automates much of the data processing work.
“Our biggest challenge was a scarcity in internal resources, as development teams had to be taken away from day-to-day tasks to facilitate integrations for customer onboarding,” says Walter. “With AI [artificial intelligence]-powered integration solutions like SnapLogic, we can onboard new clients at a far lower cost than previously estimated, as well as automate a large number of processes that previously required dedicated manual work.”
He says this has eliminated any dependency on software development teams, as well as removing the blockers that were previously delaying the business in bringing additional customers onto its platform.
It is a similar story for online clothing retailer Hush, another SnapLogic customer. Brad Woodward, head of data at Hush, echoes Walter’s concerns over manual processes and data inputs soaking up resources, but fundamentally it was the need to connect data, to make it sing.
“We realised early on in our digital transformation that integration would be the key to getting the most out of our data,” says Woodward. “While it’s nice to have large volumes of data, when it is inaccessible, it isn’t very useful to the business.”
Caught in a trap
Trapped or hidden data is not uncommon. Even with SaaS, it’s possible to get cloud-based and on-premise silos. This is a recurring challenge given the breadth of SaaS applications and the different approaches to data management and interoperability.
DigitalRoute’s Hateley says that, in his experience, organisations can encounter a lack of accessibility between departments due to permissions or simply systems not communicating with each other.
Inconsistency between datasets and siloed data, perhaps collected and stored in different ways, are making it difficult for organisations to analyse and consolidate data accurately. “The quality of the data in question can also make it taxing on an organisation’s time,” says Hateley. “Some silos may be outdated, incomplete or even duplicated, impacting its integrity and accuracy when it comes to reporting.”
For Hateley, the most obvious answer to integrating SaaS is APIs. “It almost beggars belief that SaaS applications continue to remain closed off to sharing data easily,” he says. “It’s the sort of scenario that gives fuel to an organisation such as the MACH Alliance, a not-for-profit industry body of over 70 members dedicated to “open and best-of-breed enterprise technology ecosystems”.
Composable technology is at the heart of this, something Conor Egan, vice-president of product and engineering at Contentstack, a composable content management system (CMS), clearly advocates. APIs and enterprise SaaS ecosystems are fundamental here.
“In an API-first system, organisations can adapt to changing compliance requirements by adding modular components, which can be easily updated through the cloud to help maintain your security posture,” says Egan, referring to the data governance challenges of a SaaS-heavy business.
This is something that touches most businesses with data across multiple platforms, but Egan believes the biggest challenge concerns the speed in which organisations are adapting to customer needs. A lack of APIs, data integration and interoperability is only going to lead to trouble.
“With the e-commerce boom, enterprises are dealing with more content, channels, data policies, guidelines and integration points,” says Egan. “The biggest proof case for composable technologies is eliminating integration complexity. The ease of integration and subsequent flexibility that APIs offer enterprises provides them with the agility needed to scale their organisation whilst managing data compliance across systems with ease.”
Putting this into context, he says the average marketing organisation has deployed 91 different tools across its stack, raising the spectre of fragmentation, with data in silos and decisions being made on incomplete intelligence.
On this point, analytics cloud platform company Alteryx recently released some findings from research revealing that UK businesses are struggling to make timely, insight-driven decisions due to what it calls “analysis paralysis”. The research found the majority of strategic decisions (55%) within a company take an average of four weeks.
Overcoming this paralysis is fundamental to sharing data, but as Egan points out, there are also security concerns to consider here. With data moving across multiple platforms and via various virtualisation tools, there is a risk of leaving some loose ends. Composable architecture, he adds, helps mitigate security risks by segmenting different types of data into different systems.
“This means that in the event of a breach, bad actors would only have access to the data managed by that piece of the architecture, safeguarding the system as a whole,” says Egan. “Monolithic systems, particularly on-premise systems, could potentially provide a threat, giving bad actors the ‘keys to the kingdom’ from one single entry point.”
Onwards and upwards
All of this is not going to stop the surge in SaaS, and neither should it. As an IDC report into enterprise application growth revealed earlier this year, the market will see a five-year compound annual growth rate of 8% to reach a whopping $385.2bn in 2026.
Discussing the findings, Mickey North Rizza, group vice-president for enterprise software at IDC, says: “The digital world is completely altering the way software is utilised and incorporated into the organisation, from modularity to APIs, to low code/no code, to business process automation, to TaskApps, and even with innovation.”
It’s a good point, but somehow doesn’t ease the idea of complexity. For organisations continuing to plough the SaaS field, this is an issue that won’t go away. Data integration is a continuing pain point that needs addressing. The rise in data virtualisation and connector tools is testament to the growing needs of businesses to overcome SaaS integration challenges creating an increasingly confusing and fragmented space. This one is not going to go away any time soon.