AI impact tools

AI impact on my business.

Enter your headcount and a rough split across function buckets. The tool returns a directional read on how much of your work AI could assist, how much it could largely handle, and where to look first. It uses the usage patterns published in Anthropic's Economic Index.

Every number here is a directional estimate built from broad usage patterns, not a measurement of your specific company. Usage is not the same as cutting headcount. Read the caveat box under the calculator before you act on anything.

Your team

Enter the approximate number of people in each function. Leave a box blank or at zero if it does not apply. The tool adds them up for you.

Software and engineering Writing and marketing Analysis and data Admin and back-office Customer service and support Sales and outreach Operations and physical work Management

Typical work hours per person, per week

This calculator needs JavaScript. With scripts off you can still read the framing and the sources below. The core pattern from the Anthropic Economic Index: admin, support, and sales work lean toward automation, writing and management lean toward assisting a person, and operations and physical work show very low AI use today.

Read this before you act. These are directional estimates built from Anthropic's Economic Index usage patterns, not a measurement of your specific company. High usage is not the same as headcount reduction. A function can show a large assist or automate share and still need the same number of people, because AI often handles parts of tasks rather than whole jobs, and it tends to struggle on the most time-consuming work. Treat the output as a starting point for a conversation, not a plan.

How it works

What the buckets mean.

Each function carries a lean drawn from the Economic Index. These are the patterns behind the estimate.

Software and engineering: the heaviest area of AI use. Collaborative in chat, highly automated inside coding tools.
Writing and marketing: high use, leaning toward assisting a person on drafts, edits, and variations.
Analysis and data: medium to high use, mixed, with higher autonomy for straightforward data outputs.
Admin and back-office: automation-heavy, mostly through the API. Email, scheduling, invoice processing, document handling.
Customer service and support: rising automation for routine billing and payment questions, with hard cases staying human.
Sales and outreach: automation-heavy through the API. Cold email, lead qualification, and enrichment doubled in early 2026.
Operations and physical work: very low AI use today. Hands-on, in-person tasks are mostly out of reach.
Management: lower use but rising, leaning toward assisting with analysis and memos.

Sources and method

Where these numbers come from.

The tool assigns each function an assist, handle, and low-exposure split based on the augmentation and automation leans reported in the Anthropic Economic Index, then weights those splits by your headcount. The hours figure multiplies your entered hours by the assist-and-handle share and applies a conservative discount, because coverage of tasks is not the same as doing whole jobs. All outputs are labeled as directional estimates.

Atlas Instinct is an independent firm. It is not affiliated with, endorsed by, or sponsored by Anthropic. Figures here are cited from Anthropic's public Economic Index research, and directional estimates are labeled as such.

Want to go deeper than an estimate?

Check a single role in the job exposure tool, or sort real tasks with the augment or automate classifier.

Talk to Atlas Instinct about what this means for your team