These Claude Enterprise features let you track AI ROI – what’s new
Is your expensive Claude Enterprise account worth the cost? Anthropic is introducing three tools to help businesses understand what they get from using it.
ZDNET’s key takeaways
- Many businesses still want clearer proof that their spending on AI is worth it.
- Claude Enterprise now offers three generally available features that track work results and bring budget information together.
- Anthropic needs to keep growing its enterprise business ahead of its planned stock market listing.
Some reports suggest that businesses are finally starting to get value for their money from expensive AI tools. But many companies still want better information to help them decide whether the productivity improvements are worth the cost.
Trying to give its business customers a clearer picture of how they use AI, Anthropic has made three features generally available to Claude Enterprise customers. All three were released in beta in September. They are smart reports, pooled budgets, and in-app surveys.
Smart reports show which tasks employees use Claude for, how often they do those tasks, and how much the tasks cost. The reports are turned off by default. They also show where people have problems while using Claude and point out tasks that employees repeat often and that could be turned into regular, automated workflows, the company said.
AnthropicAnthropic explains in its documentation that smart reports are not meant to measure individual employees' performance or help make decisions about their jobs. This warning matters because AI is increasingly affecting workplace decisions.
The pooled budgets feature lets administrators set a monthly spending limit for a group within a company, alongside the separate monthly limit for each user account. Administrators can set limits for individuals and groups based on how much they spent in previous months. The goal is to manage spending more effectively, so one person who uses Claude a lot does not run out of budget while other users leave some of their budgets unused. Users can be part of several pooled budgets, and these budgets can be ranked by priority.
With in-app surveys, administrators can create surveys that appear inside Claude Chat. These surveys ask employees about how they use Claude. Administrators can then view combined results in the Admin console.
AI spending and Anthropic’s IPO
Claude Enterprise is much more expensive than a Claude Teams account. Claude Enterprise costs $20 per user account each month, plus fees based on API token use. Claude Teams also costs $20 per user account each month, but it includes tokens, so users do not pay a separate fee for their usage. It also limits accounts to 150 users, making it a more affordable choice for smaller teams.
Companies that have already decided to spend large amounts on AI may be able to afford Claude Enterprise. However, the more they spend, the more important it becomes to show that the investment is worthwhile. Making these tools available to all eligible customers may provide useful information and encourage larger customers who are starting to have doubts to keep using the service.
Anthropic has been a leader in the business AI market, but OpenAI has caught up in recent months. Both companies are preparing for stock market listings, known as initial public offerings, or IPOs. New data from Ramp also shows that spending on Anthropic reached its highest point in July, then stayed slightly lower in October so far.
The market is changing quickly, and some people worry that AI may be in a bubble. Anthropic needs to keep gaining customers, especially since it filed the paperwork for its long-awaited IPO on Monday. In its prospectus, Anthropic warned: “If the market for our models, platform, and applications develops more slowly than we expect or declines, our growth may slow or stall, and our business, financial condition, and results of operations could be adversely affected.”