If you're like most organizations, you've probably made some serious investments in your software landscape over the years. Chances are you have an ERP system, a CRM system, and an assortment of specialized tools for things like project management, HR, or field service. On paper, it looks like a reasonably complete picture. In practice, though, a lot of that value never gets fully realized.
The culprit is rarely the technology itself. More often, it's gaps in support and strategy that's the problem. And these gaps have a way of compounding quietly in the background. Year over year, they cost organizations money, efficiency, and competitive ground. Left unaddressed, they don’t just hold you back: they prevent your technology investments from ever becoming the compounding assets they were meant to be.
In this article, we'll walk through what we think organizations need to create that kind of flywheel effect, why they're struggling to set it in motion on their own, and what the right support model makes possible when everything starts working together.
You’re Probably Getting Less Than Half the Value You Paid For
A few years back, the Aberdeen Group conducted a study analyzing ERP usage patterns across more than 1,000 companies. Their research found that organizations were only using about 27.6% of the functionality available in their ERP systems. And while ERP platforms are the most common example, the same pattern tends to hold true across other major business systems as well, including CRM and HR systems.
In our experience, this lack of adoption is not because the features are bad or the business doesn't need them. More often than not, the issue is much simpler: the teams responsible for rolling out and operating these systems either don’t realize the capabilities exist or lack the time and resources to properly evaluate and implement them.
The downstream effects are predictable. When teams don't know what their core platforms can actually do, SaaS tools start looking like potential cheat codes. A product catches someone's eye, it seems to fill the gap, and before long it's licensed, implemented, and baked into the stack. Then another gap emerges, and the cycle repeats. This cycle has a name: it's called SaaS sprawl.
SaaS Sprawl is an Expensive Band-Aid
When left unchecked, SaaS sprawl tends to become a real problem. According to a recent study by Productiv, the average company now manages somewhere between 200 and 300 SaaS applications. When you factor in licensing costs, maintenance overhead, integration complexity, and the sheer amount of time employees spend navigating disconnected systems, the economics deteriorate pretty quickly.
To make matters worse, many of those third-party tools end up overlapping with functionality you already have. We frequently walk into environments where teams are paying for multiple overlapping solutions that collectively duplicate capabilities already included in their core platforms. In some cases, prior integrators have even been paid to build custom extensions that recreated out-of-the-box features the business had already licensed.
What you end up with is a collection of tools that each work reasonably well on their own but don’t talk to each other in any meaningful way. Data gets entered in one system and manually re-entered in another. Requests fall through the cracks between applications. Reports require someone to pull information from four different dashboards and stitch it together in a spreadsheet. And somewhere in the middle of all of that, there are human beings spending a significant portion of their workday acting as the connective tissue between systems that should be integrated.
Figure 1: Swivel Seat Integration with Human Beings Running Point
When Data Lives in Silos, Nobody Can See the Whole Picture
One of the most consistent complaints we hear from business leaders is that they can’t get a clear line of sight across their organization. It's certainly not for lack of data. If anything, most companies have more data than they know what to do with. The problem is that the data's scattered across a dozen systems, each with their own reporting format and update cadence. Getting a complete answer to even a fairly simple business question can require pulling partial data from multiple systems and reconciling it by hand.
What's missing is a comprehensive data integration strategy and a centralized place to bring all this data together underneath one roof. Most SaaS products offer built-in reporting capabilities, but those reports only reflect what’s happening inside that particular application. Understanding what’s happening across the business requires connecting the dots between systems, and that’s not something that happens automatically.
Figure 2: The Benefits of Having a Clearly Defined Data Strategy
Organizations that solve this problem don’t just get better reports. They gain the ability to pinpoint bottlenecks, spot trends earlier, and make decisions based on a complete picture rather than a fragmented one.
AI Dials Up the Urgency for Data & Integration
As organizations wrestle with all this SaaS sprawl, there's a certain irony in watching those same software vendors come back to the table, this time pushing their next-generation AI-infused solutions. And to add insult to injury, each vendor is spinning their own narrative for why their solution is the only one worth considering.
For leadership teams under pressure to define a corporate AI strategy and deliver meaningful results, it can be incredibly difficult to cut through all the hype. At the same time, standing still isn't an option. The fear of being left behind is real, and it's felt at every level of the organization.
This is where a strategic partner makes all the difference. Someone who can help you cut through the noise, make sense of your landscape, and build a roadmap that's grounded in your specific business context rather than a vendor's agenda. Navigating an AI transformation is genuinely uncharted territory for most organizations, and trying to do it without an experienced guide is a risk few can afford to take.
Regardless of where you are in your AI adoption journey, it's important to recognize a universal truth: AI is only as good as the foundation underneath it. Which brings us back to our data and integration problem. In our experience, most organizations have a lot more work to do here than they initially realize.
It's sobering, in the way that most "eat your vegetables" conversations tend to be. But the reality is that AI models don't operate in a vacuum. They rely on access to quality data, well-defined workflows, and the right tools and integrations to drive real automation. Without those things in place, even the most sophisticated AI solution will underdeliver. The vendors pitching you their next-generation platform won't always lead with that, but a good advisor will.
The good news is that the work of getting your data and integration house in order isn't wasted effort. It pays dividends well beyond AI. And for organizations willing to do it right, it becomes the foundation for everything that comes next.
AI really is powerful, and the productivity gains it can create are very real. But AI also works best when it’s connected to clean, organized, well-integrated data and supported by strong governance and security practices. Without that foundation, things can get messy fast. Sensitive information can leak into the wrong places, outputs can become unreliable, and organizations can end up creating compliance and security problems they didn’t anticipate.
That’s why the support and integration gaps we’ve been talking about matter so much right now. Getting your application landscape in order isn’t just an operational improvement anymore. It’s becoming the foundation for doing AI responsibly and effectively. The companies putting those fundamentals in place today are the ones best positioned to take advantage of automation and AI at scale. The ones skipping that work may be taking on more risk than they realize.
The In-House Handyman Problem
In our experience, this realization tends to follow a predictable pattern. Once organizations recognize the notable gaps in their software support model, the first instinct is almost always the same: try to hire their way out of it. We're frequently asked to review (Copilot-based) job requisitions for what we’d describe as “unicorn” resources—a single individual who is expected to serve as a project manager, systems analyst, developer, integration specialist, and data engineer all in one.
We understand the instinct. It feels cost-effective, and it keeps things simple from an org chart perspective. But in practice, we find that this approach rarely delivers the results organizations are hoping for. Proper business transformation—the kind that actually moves the needle—requires a multi-disciplinary team with complementary skills, not a single generalist stretched across everything.
At minimum, meaningful transformation work typically requires access to:
Project managers who can coordinate across stakeholders and keep work moving
Business analysts who can translate process requirements into technical specifications
Solution architects who understand how systems should fit together
Developers who can build integrations and custom functionality
Data engineers who can create the pipelines and repositories needed for unified reporting
Quality engineers who can validate that changes work correctly before they reach production
Platform specialists who stay current on the capabilities of the core systems already in use
Of course, you don’t need all of these resources engaged full-time. But you do need access to all of them at the right moments. That’s a difficult thing to replicate with a single hire, and it’s one of the key reasons why organizations that try to manage this work internally tend to see slow progress, inconsistent results, and high turnover among the people assigned to lead it.
The Managed Services Parallel
Here's a curious pattern we see regularly. Organizations that wholeheartedly embrace managed services for their hardware, networking, and cybersecurity will often draw a hard line when it comes to their software and data estate, convinced that function is better handled in-house. As we explored in a recent article on why business transformation isn't DIY, the logic that drives smart infrastructure decisions rarely gets applied with the same consistency to the application layer.
And yet the parallel is hard to ignore. Network infrastructure and cybersecurity get outsourced because the expertise required is deep, the stakes are high, and the cost of getting it wrong is obvious. Software application support isn't so different, but for some reason it's often treated as though it is. Organizations that wouldn't dream of building their own cybersecurity practice from scratch will happily spend months trying to hack together application improvements with a part-time internal resource and a collection of AI tools.
We’ve seen this play out many times. A business champion takes on an improvement initiative alongside their existing responsibilities. They make some early progress. Then priorities shift, or a vendor release changes something, or the original scope turns out to be more complex than expected. A year later, the initiative is still in progress. Two years later, it might still be in progress. Or, the business champion has moved to a different role and the whole thing is starting over.
The compounding cost of that kind of stalled progress is hard to quantify precisely, but it’s very real. Every month spent not capturing the available value from existing systems is a month of efficiency foregone. Every integration that doesn’t exist means more manual work and more exposure to data errors. Every AI initiative that gets delayed waiting for the data foundation to be ready is a competitive gap that widens.
Transformation is a Journey, Not a Project
One of the most important mindset shifts we try to help customers make is moving away from thinking about this as a series of discrete projects and toward thinking about it as an ongoing capability. Your application estate isn’t something you configure once and leave alone. The platforms evolve, the business evolves, and the opportunities available through automation and AI are going to keep expanding. Maintaining and building on those investments requires sustained attention.
The organizations that are getting the most out of their technology investments are the ones that have found a way to make continuous improvement a structural part of how they operate. In most cases, that means working with a partner who can provide consistent support, bring in the right specialized skills at the right time, and keep the broader strategy in view even as individual workstreams evolve.
When this model works well, something interesting happens. IT stops being viewed as a cost center and/or an expense line on a budget that leadership tolerates but would rather minimize. Instead, it becomes a genuine driver of business value: a function that compounds efficiency gains over time, enables better decisions through integrated data, and creates the conditions for AI to actually deliver on its promise.
Closing Thoughts
Most of the challenges we've outlined in this article aren't new. Organizations have been wrestling with SaaS sprawl, data silos, and software support gaps for years. What's new is the urgency. AI has raised the stakes considerably, and the foundational work that was easy to defer a few years ago is now directly in the critical path of initiatives that leadership teams are under real pressure to deliver.
What makes the difference is getting the flywheel moving. When your application landscape is well-managed, your data is clean and connected, and you have the right partner providing consistent support, something shifts. IT stops being a source of friction and starts becoming a genuine driver of business value.
Integrations that used to require manual workarounds run automatically. Decisions that used to require someone to pull data from four different systems get made faster and with more confidence. And as that foundation matures, AI stops being an aspirational talking point and starts delivering the productivity gains it's actually capable of. Each improvement builds on the last, and over time, the compounding effect becomes significant.
That's the goal worth aiming for: a technology operation that isn't just keeping the lights on, but actively driving innovation on a continuous basis. The organizations building that capability today, getting their foundations in order while others chase the next vendor pitch, are the ones that will be best positioned to capitalize on everything AI has to offer. The flywheel takes effort to get started. But once it's turning, it's a powerful thing.
