ChatGPT Enterprise: new spend controls and usage analytics

What changed in the June 2026 update

On June 18, 2026, OpenAI has officially published a relevant update for ChatGPT Enterprise. The announcement introduces two major features: the new spend controls and an advanced system of usage analytics . Both features address concrete operational needs of organizations that have already adopted AI in production.

Specifically, spend controls allow administrators to set spending limits per user, per team, or per organizational unit. Additionally, it's possible to configure automatic alerts when predefined thresholds are reached. This reduces the risk of unplanned consumption, a frequent problem during rapid scaling phases.

Usage analytics, on the other hand, offer a centralized dashboard. That way, IT managers and CFOs can check out usage stats broken down by department. As a result, checking on internal adoption is way more accurate than with older tools.

The problem these tools actually solve

Until now, one of the roadblocks to enterprise adoption of generative AI was how hard it was to predict costs. Many B2B organizations ran into situations where usage shot up fast, but without a clear picture of how it was being split up internally. Still, shutting off access just to keep costs down meant missing out on productivity boosts.

According to research by McKinsey on the topic of AI adoption , one of the main barriers to the systematic adoption of AI in businesses is precisely the difficulty of financial governance. Thus, tools like spend controls are not a technical detail: they are a strategic enabler.

Similarly, not having detailed analytics used to make it tough to justify AI spending internally. Now, usage data turns into a real asset for management. Because of this, enterprise AI ROI is much easier to measure and pitch to company boards.

Immediate impact for Italian B2B SMEs

Italian SMEs operating in B2B contexts often find themselves in a unique position. On one hand, they need to adopt AI to stay competitive. On the other, they have leaner IT teams and tighter budgets than large corporations. Therefore, AI cost governance is even more critical for them.

With the new spend controls, an SMB with 50-200 ChatGPT Enterprise users can now set different budgets for each department. For example, the marketing team can have a different allowance than the legal or sales team. Plus, automatic alerts let you step in before hitting your limits, without having to keep an eye on usage manually.

Usage analytics, on the other hand, offer a less immediate but equally important advantage. In fact, knowing which teams use the tool the most—and in what contexts—helps identify high-value use cases. Consequently, you can focus training and optimization where the impact is greatest.

We at SHM Studio we work with SMEs that are integrating AI tools into their operational workflows. Often the problem isn't the technology itself, but the lack of structure around it. These new OpenAI tools go precisely in that direction.

Spend controls architecture: how it works in practice

ChatGPT Enterprise spend controls work at the workspace . The admin can set monthly spending limits overall or for each individual user. When a set percentage is hit—like 80% of the budget—the system automatically shoots out email or webhook alerts.

It is also possible to set a hard cap , or an absolute limit beyond which access is temporarily suspended. However, this option should be used with caution in production environments. Therefore, the ideal setup includes multiple alerts before reaching the automatic block.

Usage analytics, on the other hand, aggregate data into a dashboard accessible to admins. Available metrics include: number of sessions per user, volume of tokens consumed, hourly distribution of access, and type of usage. Among other things, this information can be exported in CSV format for integration with company BI systems.

To dive deeper into the technical architecture, it is also useful to consult the official OpenAI documentation , which constantly updates the specifications of enterprise features.

What to do next: three practical steps

For organizations already using ChatGPT Enterprise, the first step is to head over to the admin section and check out the new spend management options. First off, it's a good idea to map out your active teams and guess how much each one is likely to use. This gives you a solid starting point for setting realistic limits.

Next up, we suggest turning on usage analytics and gathering data for at least four weeks before making any tweaks to save money. Honestly, those first few weeks usually show usage patterns that are still settling down. So, it's better to just watch and wait before jumping in.

Lastly, it's super helpful to share usage reports with the managers of each department. This builds teamwork accountability and encourages people to use the tool more wisely. Plus, this data can really help spark conversations about the real value AI is bringing to every department.

For small and medium businesses checking out ChatGPT Enterprise, this update removes a major roadblock. Still, just having the tools available doesn't guarantee people will actually use them effectively. A structured digital strategy remains the fundamental prerequisite.

The still open construction site: what is still missing

That said, the update has a few areas for improvement. Current spend controls mainly work at the token consumption level. However, you still can't break down costs by task type—for instance, separating GPT-4o usage from lighter models within the same workspace.

Furthermore, native integration with ERP systems or procurement platforms is still limited. SMEs that manage IT spending through centralized tools will therefore have to rely on CSV exports and custom integrations. This represents a significant operational cost for smaller IT teams.

According to an analysis by Gartner on enterprise AI governance , organizations that implement structured AI spending controls achieve an average 23% reduction in unplanned costs in the first year. Therefore, even with current limitations, the value of these tools is measurable.

For strategy managers SEO or google ads campaigns with AI support, keeping tabs on how much generative tools cost becomes a regular part of the marketing budget. So, these features matter just as much outside of IT teams.

Outlook: where ChatGPT Enterprise is heading in the next 12-18 months

The June 2026 update follows a clear path. OpenAI is slowly shifting ChatGPT Enterprise from a simple productivity tool into a corporate AI infrastructure . So, governance features — spend controls, analytics, audit logs — will become increasingly central compared to purely generative capabilities.

Over the next 12-18 months, it's reasonable to expect deeper integrations with IAM (Identity and Access Management) systems, compliance dashboards, and data lineage tools. Plus, European regulatory pressure—especially the AI Act—will push vendors to offer more and more granular tracking and explainability features.

For Italian SMEs, this means that investing today in structuring AI adoption — with governance, training, and metrics — isn't just an optional extra. Instead, it's the key to scaling sustainably when tools become even more powerful.

Those who want to learn more about how to integrate these tools into an overall digital strategy can explore the SHM Studio services or contact the team directly . Furthermore, the SHM Studio blog regularly publishes analysis on AI, web development and B2B LinkedIn strategies . Finally, for those working on optimized content, the service of SEO copywriting already integrates AI-assisted approaches with structured editorial supervision.