Where does the time go?
Take one piece of customer feedback, from the email landing in the inbox to the customer hearing what happened with it. Support knows triage takes minutes, product knows tagging takes a little more, engineering can estimate its part. The total is rarely known, because the flow crosses several teams and the clock keeps running between them.
What do recent numbers say?
In Glean's Work AI Index, published in June 2026, workers say AI saves them about 11 hours a week, yet only 13 percent say it has significantly improved their organization's performance. Anthropic's Economic Index from the same month notes that Claude sessions increasingly consist of long-running agentic tasks. Individual tasks are getting faster; the effect on the organization as a whole is much smaller.
How do you see the time between teams?
Map the process across all the teams involved, with the duration of each step and each handoff. It then becomes clear which part of the waiting nobody owns. That is the picture kawaru.ai builds: AI agents interview the people who do the work and assemble the flow, with the numbers on it.
Sources
- Glean, Work AI Index 2026, 10 June 2026. Survey of 6,000 full-time digital workers in the US, UK and Australia.
- Anthropic, Anthropic Economic Index report: Cadences, 26 June 2026. Data from April to June 2026.