
The SaaS Mess: How Enterprises Lost Control Of Their Software
Pankaj Goel is the technology leader and cofounder of Opkey, an Agentic AI-powered ERP lifecycle optimization and assurance platform.
Five years ago, Gartner released a seminal paper that introduced the concept of a transformative world: composable ERP.
The paper promised enterprise software that would function like LEGO bricks: modular, buildable and easily reconfigurable. They called it 'an adaptive technology strategy' that 'enables…[you] to keep up with the pace of business change.' But the reality of ERP in 2025 has proved quite different. What was meant to be an agile, efficient software ecosystem is instead an unmanageable web of disconnected applications and fragmented data. The status quo today drives enterprise costs higher than they need to be. In a recent conversation I had with a CIO, they put it bluntly: "Composable ERP, without a proper lifecycle management strategy, is like building a race car with mismatched parts. Sure, it might look cool in the garage, but good luck winning any races."
Composable ERP has delivered better solutions for individual processes but also adds new layers of complexity. What was formerly a single-vendor ERP environment is quickly becoming a sprawling SaaS mess of multiple applications. IT teams find themselves tasked with the near-impossible: keeping everything connected and running smoothly.
Consider the steps required to execute a typical 'procure-to-pay' process. First, a requisition might begin in Workday, a spend management platform. Then, a purchase order is processed through Oracle Fusion. Interactions with suppliers are in SAP Ariba, inventory is managed through Blue Yonder, invoicing goes through Stampli and finally, payments flow through Tipalti.
This approach offers the flexibility you need for functional options but comes with a heavy price. IT teams can spend more than a third of their time maintaining custom integrations between applications like these, costing enterprises anywhere from $5,000 to $20,000 per connection. Often dubbed 'SaaS tax,' many companies are paying dearly for their SaaS ecosystem.
What's worse, each of these SaaS vendor updates runs on its own schedule, forcing companies into a never-ending testing cycle and depleting valuable IT team resources.
End-users, your non-IT employees, need a great deal of training to carry out workflows on so many apps, which costs an immense amount to carry out manually. An even greater problem is the cost of fixing integration-related defects. It takes an IT team an average of thirteen days to resolve an integration defect, slowing down other business priorities and taking up valuable resources.
Enterprises don't need to rush to abandon their composable architectures. Enter: ERP Lifecycle Optimization, a complete framework for proactively managing change across fragmented systems. Let's dive into how they help:
Sorting through your SaaS mess requires that you get ahead of issues. To do this, your IT team needs excellent system visibility. More specifically, they need real-time monitoring of configuration updates, API changes and metadata modifications to anticipate disruptions and avoid downtime. A misstep I've often observed—assuming manual alerts and ticketing systems are enough. Here's why they fall short: By the time a problem shows up in a support ticket, it's already customer-facing. I don't know about you, but I want to catch issues before users ever feel them.
In an enterprise, no app is an island. A minor update in one SaaS app can completely derail another, causing a cascade of downstream issues across multiple workflows. With AI-enabled analysis, teams can visualize the impacts of cascading issues instantly. Manual methods of impact analysis are not wise, as teams often realize that they've missed critical cross-module impacts too late. My advice: Invest early in systems that surface and analyze these connections automatically and proactively. The hidden issues are usually the ones that hurt the most.
Providers update the software frequently, which is viewed as one of the advantages of SaaS applications, but it is unsustainable if your testing methods are still manual. Enterprises need automated test frameworks to validate integrations and business processes in real time. Adopting robust test automation can mean 70% faster release cycles, with a 90% reduction in the occurrence of post-release defects. For this, I recommend agentic systems. They not only execute these tests but also decide which tests to run. They adapt testing based on what's changing, so you can rest assured you're covered.
With every update, employees need to be trained on how to adapt to each change, but too many training programs are static PDFs or outdated videos created manually. The most valuable enablement happens in the flow of work—giving users contextual help and guidance as they are navigating their tasks. For this, I recommend AI-driven training delivered through in-platform guidance. Agentic digital trainers can personalize training to specific user behaviors and adapt in real time, resulting in faster adoption, fewer errors, less workflow disruption and drastically fewer support tickets for your IT team to address.
To truly solve problems, IT teams need more than just error logs. AI-powered observability tools bring real-time visibility to your team across the ERP lifecycle. These tools can tell you not only where the fire is burning but also how it started and where it might spread. Agentic-powered observability tools allow your team to shift from putting out fires to preventing them in the first place.
Composable ERP is finally within reach. Enterprises that treat their ERP landscape as a unified whole will be the ones thriving in the Agentic AI era.
This is where ERP Lifecycle Optimization becomes essential. Instead of reacting to problems as they arise, enterprises can take a proactive approach to the core system of their IT ecosystem. ERP Lifecycle Optimization enables enterprises to orchestrate change, enhance existing systems and mitigate risks before they materialize.
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
Hashtags

Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles


Gizmodo
an hour ago
- Gizmodo
Wikipedia Tries to Calm Fury Over New AI-Generated Summaries Proposal
The denizens of the open web don't want anything to do with AI. The Wikimedia Foundation, the organization behind Wikipedia, made the unfortunate decision to announce the trial of a new AI-fueled article generator this week. The backlash from the site's editors was so swift and so vengeful that the organization quickly walked back its idea, announcing a temporary 'pause' of the new feature. A spokesperson on behalf of the Foundation—which is largely separate from the decentralized community of editors that populate the site with articles—explained last week that, in an effort to make wikis 'more accessible to readers globally through different projects around content discovery,' the organization planned to trial 'machine-generated, but editor moderated, simple summaries for readers.' Like many other organizations that have been plagued by new automated features, Wikipedia's rank and file were quick to anger over the experimental new tool. The responses, which are posted to the open web, are truly something to behold. 'What the hell? No, absolutely not,' said one editor. 'Not in any form or shape. Not on any device. Not on any version. I don't even know where to begin with everything that is wrong with this mindless PR hype stunt.' 'This will destroy whatever reputation for accuracy we currently have,' another editor said. 'People aren't going to read past the AI fluff to see what we really meant.' Yet another editor was even more vehement: 'Keep AI out of Wikipedia. That is all. WMF staffers looking to pad their resumes with AI-related projects need to be looking for new employers.' 'A truly ghastly idea,' said another. 'Since all WMF proposals steamroller on despite what the actual community says, I hope I will at least see the survey and that—unlike some WMF surveys—it includes one or more options to answer 'NO'.' 'Are y'all (by that, I mean WMF) trying to kill Wikipedia? Because this is a good step in that way,' another editor said. 'We're trying to keep AI out of Wikipedia, not have the powers that be force it on us and tell us we like it.' The forum is littered with countless other negative responses from editors who expressed a categorical rejection of the tool. Not long afterward, the organization paused the feature, 404 Media reported. 'The Wikimedia Foundation has been exploring ways to make Wikipedia and other Wikimedia projects more accessible to readers globally,' a Wikimedia Foundation spokesperson told 404 Media. 'This two-week, opt-in experiment was focused on making complex Wikipedia articles more accessible to people with different reading levels. For the purposes of this experiment, the summaries were generated by an open-weight Aya model by Cohere. It was meant to gauge interest in a feature like this, and to help us think about the right kind of community moderation systems to ensure humans remain central to deciding what information is shown on Wikipedia.'


Wall Street Journal
an hour ago
- Wall Street Journal
Treasury Yields Fall Amid Concerning Labor Data, Mild Inflation
0900 ET – U.S. labor and inflation data deepen a decline in Treasury yields. Weekly jobless claims were unchanged from the previous week's upwardly revised pace, at 248,000. Economists surveyed by WSJ expected 246,000. Continuing claims, a measure of the unemployed population, was 1.96 million, the highest level since November 2021. May's wholesale price inflation was 0.1%, accelerating from April's 0.2% deflation and below consensus of a positive 0.2%. The combination of slower-than-expected inflation and concerning labor data underscores bets that the Fed may need to change its hawkish position. Yields were already declining and fell further after the data. The 10-year Treasury yield is at 4.360% and the two-year at 3.891%. ( @ptrevisani) 0614 GMT – A downside surprise in U.S. CPI data gave only a small boost to Treasurys, probably because tariff-driven price hikes still look imminent, says Capital Economics' James Reilly in a note. That said, these price hikes look discounted in markets, shielding Treasury yields from rising pressure, the senior markets economist says. 'We don't expect much upwards pressure on Treasury yields even as the inflationary impact of tariffs eventually feeds into U.S. consumer prices,' he says. Capital Economics expects core inflation to rise in coming months but it thinks that investors are already braced for a broadly similar outcome on tariffs, he says. (


Forbes
an hour ago
- Forbes
iOS 26 Is About To Save You Money On Your Energy Bill
Apple hasn't given the smart home much love on stage at WWDC 2025 this week, but if you dig a little deeper into the dev sessions you'll find a feature that could actually make a dent in your energy bill. Dubbed EnergyKit, it's coming as part of iOS 26, which will arrive later this year. It's all a bit technical right now, but it points toward eventually turning your Apple Home system into a money-saving energy manager for your house. I say system, rather than app, because it sounds as if Apple is going to allow app developers to bake this tech into their own apps, even if those device types aren't currently supported. EnergyKit is a developer framework lets apps tap into Apple's Home energy data, and things like your rate plan and a forecast of when the grid is running cleaner or cheaper, which it will use to shift when your devices draw power. For example, it could allow your EV charger to schedule itself to run when grid rates are low or solar energy is peaking, or have your smart thermostat pre-cool your house before prices spike. If you're hooked up to PG&E (the first and only energy provider supported so far), your Apple Home app can already show this kind of info, but EnergyKit will supercharge things and open it up to developers to build smarter automations on top. Apple says the framework is aimed at residential use, for things like HVAC systems and EV charging. In a video introducing the new tools, we're told that EnergyKit can provide personalized guidance for when to use electricity based on environmental impact and cost. Apple is actually pretty late to the smart energy party, with platforms like SmartThings and Homey pushing energy optimization for a while now; and devices from the likes of Ecobee, Eve and Tado already doing this kind of thing on their own. But this feels like Apple finally putting down a foundation to make its Home app more than just a pretty interface for turning off your lights. The Cupertino tech giant doesn't actually support energy monitoring or EV chargers natively at the moment, but obviously Matter makes this sort of thing easier. If you're a dev, you can read more technical info on EnergyKit over on the Apple Developer website.