The 10 biggest questions in IT right now (Answered by 525 IT leaders)

Every IT team is wrestling with the same questions. Should we move faster on AI? Are we already behind? How do we govern tools we can’t even see? 

During State of SaaS LIVE, we tackled five of the biggest questions facing IT today using data from our annual survey of 525 IT and security professionals. Here are those answers, plus five more questions the report answered.

1. Is IT actually ready for agentic AI, or are we deploying blind?

Short answer: Not yet, and IT knows it.

We asked webinar attendees what scares them more, data loss from AI or falling behind on AI adoption. 58% said data loss. The survey backs that up almost exactly: 2 in 3 IT professionals fear AI-related data loss more than they fear moving too slowly. At the same time, 61% say they’re under real pressure from executives to deploy agentic AI now.

That’s the bind IT is currently in. Just over half of the applications running in the average organization today are actually IT sanctioned. The rest is shadow AI, tools individuals and teams picked up on their own because they wanted the same capabilities they use in their personal lives.

As BetterCloud President Raj Kunnath put it during the session, data loss isn’t the disease, it’s the symptom. The real problem is that most IT teams can’t see everything running in their environment. And you can’t govern what you can’t see.

2. What do you do about shadow AI that’s already inside your company?

Short answer: Banning it doesn’t work. It never has.

David Politis has watched this movie before, going all the way back to the early days of SaaS, when locking down file sharing just pushed employees toward their personal Gmail accounts instead. The same pattern is playing out with AI. Shut a tool down without a plan and people don’t stop using AI, they just stop telling you about it.

The fix isn’t more restriction, it’s visibility plus a fast, safe path to “yes.” Discover what’s already in use, bring it into the light, and give teams a sanctioned way to keep doing the work they were already doing. Governance that only knows how to say no just drives the behavior further underground.

3. Is automation actually freeing IT up for strategic work?

Short answer: Not yet, but it’s still the top priority.

For the third year running, automating IT and SaaS operations work ranked as the number one priority for the next 12 to 18 months, followed closely by strengthening IT and business collaboration.

What’s changed is the shape of the ask. Automation used to mean scripts and workflows. Now it means agents doing the work, and agents only earn their keep if you can govern them.

Raj’s read on the data: the things IT ranks lowest on the priority list, like governance, are quietly the obstacle standing between them and the automation they want most. Solve governance and automation accelerates on its own.

4. Are we finally seeing ROI from SaaS and AI spend?

Short answer: Yes, and the app count going up is actually part of the win.

For the first time in two years, the average number of SaaS applications per organization went up instead of down. App counts climbed 11% in 2026 and jumped 44% in the enterprise segment specifically.

The naysayers who spent the last year predicting AI would kill SaaS had it backwards. AI adoption is driving SaaS growth, not replacing it.

More importantly, look at what’s inside those apps. Nearly every SaaS app out there is either adding AI features or integrating an AI agent into the platform itself.

Raj’s framing is the one worth remembering: stop counting apps. One app can spin up dozens of agents, meaning the real unit of sprawl isn’t the application anymore, it’s the actor. The orgs that win won’t be the ones with the leanest app count, they’ll be the ones who can see what every agent inside every app is actually doing.

If you want a quick win on cost, don’t try to boil the ocean. The top 10% of your applications by spend typically account for the majority of your total SaaS bill. Start there.

5. Will IT get a bigger seat at the strategy table, or is that wishful thinking?

Short answer: The seat is there. Readiness is the open question.

According to the survey, 80% of IT professionals expect their strategic influence to grow and 84% say IT is the function actually leading agentic AI deployment inside their company.

The optimism is great, but the report’s AI Readiness Index puts a number on the gap between ambition and reality: 61 out of 100. Most teams have deployed tools and written policies. Very few have built the automated infrastructure to actually enforce those policies in real time, and data protection scored the lowest of any category measured.

The seat at the table is there for the taking. Whether IT is actually ready to sit in it is a separate question, and right now the data says most teams aren’t, quite yet.

6. Who actually owns AI governance?

Short answer: In practice, no one. Ownership is fragmented, and that’s the problem.

Some teams are heads-down building and enabling agents. Others are quietly provisioning seats and calling it adoption, with no one tracking what those agents can actually access or do.

The clearest answer is a solid framework. Every organization needs to see what’s running, decide whether it’s allowed to run, be able to act, and prove to an auditor that the company isn’t leaking sensitive data.

Whether that sits with IT, security, or a shared function varies by org. What doesn’t vary is the accountability. Right now, most companies don’t have this.

7. How ready is my organization to actually govern AI?

Short answer: Probably not as ready as the enthusiasm suggests.

The AI Readiness Index score of 61 tells the story: strong on tool deployment and policy-writing, weak on the automated enforcement that turns a policy into something real. Data protection was the lowest-scoring category in the entire index.

A useful gut check to ask:

• Do you have a living inventory of what your AI tools and agents can access?
• Can you shut one off in minutes, not weeks?
• Could you hand an auditor proof of what an agent did last month?

If the honest answer to any of those is no, that’s your starting point, not the org-wide rollout.

8. What should IT actually automate with AI first?

Short answer: The busy work eating the hours your team wants to spend on strategy.

That was the single loudest answer in the survey: automating repetitive IT and SaaS operations work ranked as the number one priority, well ahead of consolidation or governance projects on their own.

The catch is sequencing. Automation and governance aren’t competing priorities, they’re the same project viewed from two angles. Guardrails good enough that nobody has to manually double-check them are what let automation move fast in the first place. Automate the thing, then build the visibility to trust it running unattended, and the “extra day a week” everyone says they want starts to show up for real.

9. How do I convince leadership that governance isn’t slowing AI down?

Short answer: Reframe governance as the thing that lets you move faster, not slower.

Governance has a stigma that it’s slowing things down, and that definition needs to change in your organization.

The reframe that landed in the session: governance is what lets you take your foot off the brake, not what puts it on. Teams that move the fastest aren’t the ones with no rules, they’re the ones with guardrails solid enough that they never have to stop and check. Picture it like bumpers at a bowling alley, or what one CIO called a “walled garden of Eden,” a space where employees can use the tools they want, the way they want, without ever seeing or feeling the walls that keep things safe. That’s a much easier pitch to leadership than “slow down and wait for approval.”

10. Are we solving yesterday’s SaaS problems while missing tomorrow’s AI risks?

Short answer: For a lot of teams, yes, and the data shows exactly where the miss is happening.

IT has spent years building muscle around counting and consolidating applications. That instinct is running into a landscape where the app itself isn’t the risk anymore.

One app can now spin up dozens of agents, each one a new identity, a new data path, a new actor capable of touching sensitive information on its own. Counting licenses tells you almost nothing about that. The teams ahead of this aren’t the ones with the tightest app catalog, they’re the ones who’ve shifted their governance model from tracking software to tracking actors, human and non-human alike.

Want the full picture? The complete 2026 State of SaaS report covers all ten questions in depth, along with the full survey data behind every stat in this post. Get the report here.

Prefer to watch instead? Catch the State of SaaS LIVE webinar with David Politis and Raj Kunnath for a deeper dive into the findings. Watch it on demand here.