Why hasn’t workflow automation worked for Corporate IT teams?

Automation in IT has been a goal for decades: eliminate repetitive tasks, enforce policies consistently, and prove ROI. The promise is clear, and IT teams invest heavily to achieve it.

At Stitchflow, we work with Cloud Native Corporate IT teams who rely on iPaaS platforms like Okta Workflows, BetterCloud, Workato, and Dell Boomi. They adopt these tools to save time, reduce errors, and maintain control over complex IT processes.

But even with these platforms, automation often falls short. Teams struggle with:

• Workflows that depend on multiple apps and APIs, where a single change can break a chain
• Limited visibility across hundreds of automations
• Inconsistent governance when multiple teams build workflows independently

These challenges prevent the efficiency they expected and leave IT leaders fire-fighting automation instead of benefiting from it. Stitchflow’s mission is to close that gap: making automation reliable, measurable, and scalable, so IT teams finally get the efficiency they invested in.

TL;DR

• Most IT teams automate less than 10% of workflows, and those are usually the easiest, least impactful tasks.
• Failed automation often creates more overhead, with debugging, rigid workflows, and hidden costs leading to negative ROI.
• Exceptions and limited API endpoints make complex workflows fragile, forcing IT to maintain scripts and spreadsheets instead of scaling automation.
• Some workflows—like onboarding and offboarding—are good candidates for automation, but disconnected apps and non-SCIM systems remain the hardest and riskiest to automate.
• Modern SaaS management tools like Stitchflow fix this by giving IT full visibility across every app and user (including non-SCIM ones) and turning that into reliable automation.

Why most IT teams automate less than 10% of workflows

IT automation rarely lives up to the promise. While the vision is effortless workflows, fewer errors, and measurable efficiency, the reality is very different.

Before Stitchflow, our team built atSpoke, an ML-enabled IT Service Management tool, giving us visibility into tens of millions of IT tickets. We spoke to over 100+ IT managers, and they reported that automation typically handles only 5–10% of their work. This is mostly “low-hanging fruit”: the easiest tasks, but often the least impactful.

There are exceptions. One IT leader we work with has nearly 1,000 Okta workflows and manages IT operations for a 1,500-person company with just three IT staff. But these cases are rare. More commonly, teams have few or no automated workflows, leaving them buried in repetitive work.

‼️Even leading companies face similar struggles. Vercel, managing over 100 SaaS apps under strict PCI DSS compliance, ran quarterly access reviews that involved 50+ stakeholders across Finance, IT, Security, and business teams. The process was slow, manual, and high-risk. They used:

• Notion to track access reviews, chasing follow-ups manually
• CSV exports to compare against Okta data
• Excel to identify ex-employees and outdated access

What should’ve been a repeatable, controlled process instead consumed weeks of senior team time, introduced audit risk, and relied on disconnected tools and outdated workflows.

So while the promise is near-complete automation and efficiency, the reality is fragmented workflows and minimal impact.

The cost of failed IT automation

Failed IT automation often creates more problems than it solves. Teams can end up spending more time managing broken systems than they would on manual processes.

• Time sink: When automation fails, it often demands more attention than manual processes would. Teams end up spending hours troubleshooting errors, patching broken workflows, and juggling alerts.
• Debugging complexity: Layered workflows turn routine maintenance into high-risk operations. Simple fixes become time-intensive investigations across multiple systems and platforms.
• Organizational inflexibility: Rigid automated processes lock teams into inflexible workflows. When business requirements change—new regulations, mergers, or security incidents—automation becomes a hurdle rather than a help.
• Resource drain: Teams stuck in firefighting mode have no capacity for strategic projects, process improvements, or new technology adoption—the activities that actually drive business value.

‼️The negative ROI from failed IT automations

Failed automation projects can easily cost 2–3 times their initial investment. This often includes:

• Maintenance overhead: Ongoing updates, workflow tweaks, and monitoring automated systems
• Staff troubleshooting time: Hours spent diagnosing and fixing failures instead of productive work
• Opportunity cost: Strategic projects delayed or neglected due to firefighting broken automation
• Downtime losses: Productivity and revenue are lost when automated processes fail

A project showing $100K in annual savings may actually cost $250K when all hidden and indirect costs are included—making negative ROI one of the biggest consequences of failed automation.

What makes IT workflow automation challenging

We’ve identified five reasons for the ineffectiveness of existing IT workflow automation solutions:

Business process nuance

On paper, automating a task may sound simple, but in reality, a lot of the workflow depends on specifics about the teams involved, their location, an employee’s manager, department, and specific requirements around the task.

‼️For example, the onboarding process for a sales team member in San Francisco vs. London will differ by the Okta groups and tools, the Slack channels, and the Google groups they need to be a part of.

And that’s for one department and 2 locations. Multiply your number of departments, locations, and managers, and you have 100s of workflows just for onboarding.

Every workflow has tons of exceptions

Exceptions are one of the most difficult things to manage when setting up automated workflows. Let’s take an example—you want to ensure that only active employees in Google have paid Zoom licenses.

Simple, right? You run a workflow that checks Google status for every licensed Zoom user. Unfortunately, there are usually going to be corner cases that you need to account for:

• Contractors in certain departments, like marketing, need active Zoom licenses, but they are not added to Google, which is only for FTEs
• IT test accounts to ensure Zoom is working fine don’t exist in Google
• You may have some suspended Google users who still have access to licensed Zoom because of an HR or legal requirement
• All active users in Google, but not part of the engineering team, don’t need paid licenses to Zoom

Tracking and incorporating exceptions adds significant overhead to each workflow you set up.

📚Also read: SAML, OIDC, and SCIM: The IT Leader's Guide to Modern Identity Management

Limited data and action endpoints in workflow tools

Even if you’re willing to deal with business process complexity and exceptions, a major challenge in existing workflow tools is limited access to the data and action endpoints you need to actually implement a workflow.

‼️For example, most workflow tools that support JAMF, a popular Apple device MDM, support fewer than 20 data conditions and actions. In contrast, the JAMF API has close to 2000 endpoints!

So when you’re trying to implement a nuanced business workflow with exceptions, you more often than not run into gaps in the data you need.

IT workflow automation tools aren’t easy to use

Existing IT workflow tools are not known for their user-friendly design. Users often complain about the complexity and unintuitiveness of current tools.

And there’s an unfortunate tradeoff between the complexity of the IT tool and the number of data endpoints it supports (to point 3 above). So you have a difficult choice of using an easier tool with limited capabilities, or struggling through an unintuitive, cumbersome tool if you want more fine-grained control.

It’s not just set-up but maintenance as well

IT teams must account for not only workflow setup but also ongoing testing and updates whenever business processes or exceptions change. This hidden operational overhead makes ROI harder to realize and explains why automation often focuses on low-impact, stable tasks.

As multiple IT managers on Reddit point out, an IT team isn’t sitting on a new automation request because the idea is bad. But, because even small workflows create ongoing maintenance and compliance work.

Source: Reddit

Factoring in all of these points is why automation has been limited and primarily focused on unchanging, low-hanging fruit.

📚Also read: Why don't existing IT tools help with visibility?

What should (and shouldn't) you automate in IT?

In practice, IT automation succeeds when processes are predictable, high-volume, and supported by integrated systems. But it struggles when apps are disconnected, lack APIs, or require ongoing manual oversight.

Good candidates for automation include:

• Onboarding and offboarding: Provisioning Google Workspace accounts, email, calendar, and Drive access, or revoking access when offboarding employees
• Routine IT tasks: Managing Google Groups, standardizing email signatures, and handling password resets
• Project setup and assignment: Creating project frameworks, migrating files, and assigning access to new project members
• Customer support workflows: Automatically generating Jira Service Desk tickets from feedback forms
• Security monitoring: Exporting Google Vault data, suspending users after unusual activity, or triggering urgent file reviews

Some processes, on the other hand, are inherently fragile, high-risk, or too complex to automate reliably:

• Disconnected apps or non-API systems: Workflows that span apps without direct integrations are brittle and error-prone
• Non-SCIM/SSO identity apps: User provisioning becomes complex and difficult to maintain
• Processes with high variability: Irregular steps or constantly changing logic require manual judgment
• High-risk operations: Finance, compliance, or security actions that demand human oversight

At the same time, the processes that are hardest to automate—disconnected apps, non-standard identity systems, high-variability workflows—also carry the most value when done right.

And if you want to go all in on automation, you need to be deliberate:

• Redesign the process before automating: Fix inefficiencies, simplify steps, and remove unnecessary exceptions first. Automation amplifies what already works; it can’t patch broken workflows.
• Choose tools built for the hard cases: Not all workflow platforms can handle disconnected apps or non-standard identity systems. Pick one that’s specifically built for this.

For example, you might use Okta Workflows to automate provisioning in SCIM-enabled apps like Slack or Zoom. But non-SCIM tools still require manual tracking and cleanup. By using a tool like Stitchflow alongside Okta, you cover the gap.

Stitchflow continuously reconciles Okta data against every app—via APIs, CSVs, and even non-API methods—so orphaned ex-employee accounts, hidden contractor logins, and unused licenses are automatically flagged.

📚Also read: How to get 100% visibility with Okta + Stitchflow

Types of workflow automation tools for IT

From managing projects to securing SaaS accounts, different workflow tools solve different problems.

• Task management (like Trello): Track individual and team-level tasks, manage to-do lists, assign responsibilities, and monitor progress
• Database management (like Airtable): Handle data storage, backups, and reporting while ensuring data integrity across large datasets
• Employee onboarding and offboarding (like Sapling): Streamline account provisioning, access setup, document collection, and exit procedures
• IT service management (like ServiceNow): Manage incidents, change requests, and problem resolution workflows efficiently
• Vendor and procurement workflows (like Coupa): Simplify supplier onboarding, purchase approvals, and invoice processing for smoother collaboration
• SaaS user management (like Stitchflow): Gain visibility into orphaned, hidden, or underutilized accounts across all apps, including disconnected/non-SCIM apps, and handle remediation automatically

👉Just getting into workflow automation? Here are some free tools we built at Stitchflow that can help.

How unified visibility reduces complexity in IT automation

Most IT automation struggles not because the vision is wrong, but because the foundation is missing. Without full visibility, automation is built on shaky ground—blind spots, exceptions, and disconnected apps ensure workflows remain fragile.

At Stitchflow, we tie automation directly to visibility. It starts with an IT Graph that unifies all your apps—including disconnected, non-SSO/SCIM apps, and shadow IT.

From there, you create your IT policies: setting role-based access controls (RBAC) for all apps and defining the rules for detecting orphaned accounts and unused licenses. Finally, Stitchflow runs continuous checks to ensure every app stays within your pre-set policies.

When violations are detected, you can remediate instantly with one-click automation or route them to ticketed workflows in your ITSM tool for offboarding, license downgrades, and reassignments.

And it doesn’t stop there. Every action syncs with your compliance tools, like Vanta, providing strict compliance and fully auditable logs.

Automation works best when visibility comes first—and Stitchflow makes that possible. Instead of fragile scripts and spreadsheets, you get continuous auditing, one-click remediation, and automation that adapts as your environment changes.

Book a demo and see how Stitchflow can help you go from 10% to true automation confidence.