The AI never presses send

We get asked, fairly often, why Kauzio does not just place the reorder itself. Or send the campaign. Or cut the price. The system already worked out what to do. Why is there still a human in between?

Because that gap is not a missing feature. It is the feature.

What "drafts, never executes" actually means

When Kauzio decides your mailing list needs a re-engagement campaign, or your Bristol store needs stock moved to Brighton, it does not fire the campaign or move the stock. It builds the real thing, the actual email, the actual transfer order, and stops. Someone has to look at it and press send.

That is not a UI choice we bolted on. It is in the code: the agentic execution engine's own internal notes describe the handoff as "result, human approves", the same checkpoint on every action the system can propose. There is no path from recommendation to execution that skips a person.

Every decision has a name attached, not just a status

A decision in Kauzio is not just "approved" or "pending." It has an owner, and it can have an approver, a specific person on your team, by ID, not a role, not a queue. Assigning someone as approver notifies them directly: a decision needs your approval. If something goes wrong six months later, the record does not say "the system decided this." It says who decided it, and who signed off.

That sounds like a small distinction. It is not. It is the difference between a tool that acts for you and a tool that argues with you until you act.

Some decisions get held whether you like it or not

Not every call carries the same weight. A pricing tier change and a decision to close a location are not the same category of mistake if you get them wrong. Kauzio scores every decision, easy, costly, or irreversible, and the harder it is to undo, the longer it gets held before it can be actioned. A genuinely irreversible call does not go through in the heat of a bad afternoon. It waits. You look at it again the next morning, with a clearer head, and then you decide.

We did not build that to slow you down for the sake of it. We built it because the worst decisions we have seen business owners make were rarely wrong on the facts. They were right on the facts and made in a rush.

Why we are not chasing full autonomy

There is a version of this product that removes the human entirely. Connect your data, let the system run the business, check in once a week. We are not building that, and we do not think anyone honestly serving small retail and hospitality businesses should be.

The reason is simple: when an autonomous system gets something wrong, whose mistake was it? If the answer is unclear, you have not built a tool, you have built a liability nobody signed up for. We would rather Kauzio be the sharpest possible second opinion, one that does the analysis, makes the case, drafts the action, and then steps back, than a system that quietly took the decision away from the person who has to live with the outcome.

The AI does the thinking. You still press send. That is not a limitation we are working to remove. It is the whole point.