TL;DR
Tool sprawl is the build-up of overlapping, underused and disconnected software. Unused licences are the visible cost; the bigger costs are re-keyed data, security gaps and decisions made on partial information. AI tools have accelerated it, because employees can adopt them in minutes. The fix isn't simply fewer tools. Consolidate where tools genuinely overlap, connect the ones that stay so data flows between them, retire the rest, and put light governance in place so it doesn't grow back.
No organisation sets out to run three project tools, two document stores and a dozen AI assistants. Each was added for a good reason by a team solving a real problem. The sprawl only becomes visible later, when a simple question such as "how many customers raised this issue?" needs data from four systems that don't talk to each other.
What tool sprawl really costs
- Licences nobody uses. Zylo, which tracks SaaS spend for its customers, found that the average organisation uses only 54% of the SaaS licences it pays for.
- Re-keyed and conflicting data. When systems aren't connected, people copy data between them. Copies drift, and every report starts with an argument about which number is right.
- Security and offboarding gaps. Every tool is another set of accounts to provision and remove, and another place sensitive data can sit. Tools outside single sign-on are the ones leavers keep access to.
- Slower work. People spend time hunting for information across tools and switching between them, and new starters have more to learn.
- Harder AI. AI assistants and agents can only use the data they can reach. Fragmented data means fragmented answers.
Why AI has made it worse
AI tools spread faster than any software category before them, because most need nothing more than a browser and a credit card. In a Gartner survey of cybersecurity leaders, 69% said they suspect or have evidence that employees use prohibited public generative AI tools. Gartner predicts that by 2030, more than 40% of enterprises will have security or compliance incidents linked to unauthorised AI.
AI sprawl also has a quieter form: overlapping AI features inside tools you already pay for. Salesforce, Atlassian, Microsoft and Adobe all now include AI assistants and agents, often with usage allowances and extra charges. Without a view across them, organisations pay for the same capability several times, and data ends up in places nobody approved.
In Australia, this has a legal edge. The OAIC's guidance on AI products recommends against entering personal information into public AI tools, and from 10 December 2026 the Privacy Act's automated decision rules require organisations to disclose significant decisions made by computer programs. You can't disclose what you don't know is running.
Step 1: Audit what you actually have
Most organisations underestimate their tool count, because so much is bought outside IT. Build the inventory from several sources:
- Single sign-on and identity logs show the apps people sign in to with company accounts.
- Finance and expense data shows subscriptions paid by card or reimbursed, which is where most unapproved tools hide.
- Browser and network data, where you have it, shows AI tools used without a company account.
- A short survey of team leads fills the gaps, and tells you why each tool was adopted.
For each tool, record the owner, cost, number of active users, what data it holds, and what it connects to. That last column is usually the most revealing.
Step 2: Decide what to consolidate, connect or retire
Not every overlap needs fixing. A useful rule: consolidate where tools do the same job, connect where they do different jobs on the same data, and retire what nobody would miss.
| Decision | When it fits | Example |
|---|---|---|
| Consolidate | Two or more tools do the same job for different teams | Three project tracking tools merged into one Jira site with business and software spaces |
| Connect | Different tools are each best at their job but share data | Salesforce for customers and Jira for delivery, with escalations and status flowing between them |
| Retire | Low usage, duplicated by something you already own, or a security risk | A standalone AI note-taker replaced by the assistant already included in your collaboration platform |
| Keep as is | Specialist, well used and self-contained | A design tool used only by the creative team |
Consolidation works best on platforms you already rely on. Vendors now bundle more into their core platforms, such as Atlassian's Collections, which package Jira, Confluence, Loom and Rovo AI, so the cheaper option is often to use more of what you own rather than buy another point tool.
Service desks are one of the most common cases. IT, HR, facilities and finance often each run their own ticketing tool or shared inbox, with separate portals, SLAs and reports. Consolidating them onto one platform such as Atlassian ITSM and Jira Service Management gives employees one place to ask for help, and gives you one set of SLAs, automation rules and reports to manage.
Step 3: Connect what stays
After consolidation, most organisations still run several core platforms: a CRM, an ERP or finance system, a work management platform, a service desk and a marketing platform. That's healthy, as long as they're connected. Prioritise the connections that remove the most re-keying or the most risk:
- Customer data that sales, service and delivery teams all need.
- Joiner, mover and leaver events from HR into identity and every tool with accounts.
- Financial data such as orders, invoices and project budgets that operational teams rely on.
Design each connection with a clear system of record for each kind of data, so you know which system wins when they disagree. Our guide to integrating Atlassian with your tech stack shows what this looks like in practice.
Step 4: Stop it growing back
- A light approval path. A short form for new tools that checks for overlap, data handling and single sign-on. Make approval fast, or people will route around it.
- An approved AI list. Name the AI tools staff may use and what data may go into each. People use AI either way; a list gives them a safe option.
- Single sign-on as the default. If a tool can't use your identity provider, it needs a good reason.
- A quarterly licence review. Reclaim unused seats before renewals, not after.
- A named owner for every tool, responsible for its cost, data and retirement.
How to know it's working
- Licence utilisation rising towards 80–90% on your major tools.
- Fewer manual data transfers between systems, and fewer reconciliation arguments.
- Faster onboarding and offboarding, with leavers removed from every tool on day one.
- A complete, current list of AI tools in use and the data each one touches.
How Great Minds helps
We're an integrator first. We help organisations audit their platforms, consolidate where it makes sense, for example onto Atlassian, Salesforce or Microsoft, and connect what stays so data flows between systems instead of being copied. Our Integration Hub has templates for common connections such as Salesforce to Jira.
Frequently asked questions
What is tool sprawl?
The accumulation of overlapping, underused or disconnected software tools, usually adopted team by team without a shared plan. It wastes money and fragments data.
Is consolidating onto one vendor always better?
No. One vendor rarely does everything well, and over-consolidation creates lock-in. The goal is fewer overlapping tools and well-connected core platforms, not a single tool.
What is shadow AI?
AI tools used by employees without approval, often with company data. It's the fastest-growing form of tool sprawl, and the riskiest, because data entered into public AI tools can leave your control.
How long does a tool audit take?
For a mid-sized organisation, two to four weeks to build a reliable inventory from identity, finance and survey data. Deciding and acting on it takes longer and is best done in waves, starting with the biggest overlaps.
Where to start
Export last year's software spend from finance and the app list from your identity provider, and compare them. The tools that appear in one but not the other, and the categories with three or more tools, are your first conversations. We're happy to help you work through them.