Automation phases (Phase 1 to 4)

Picaro.AI is designed to deepen operational automation step by step. Jumping straight to full automation carries both psychological and operational risk, so the path starts with one-shot manual execution and advances as you grow familiar with the operation and accumulate data.

What is available today goes up to Phase 3 (semi-automatic with an approval flow). Phase 4 (fully automatic) is not implemented; it is positioned as a future direction (concept stage).

The four phases at a glance

Phase	State	Typical use	Target timing
Phase 1	One-shot	Send prompts on demand through your AI agent environment for immediate execution	Right after connecting Picaro
Phase 2	Daily-executable	Register the same prompt as a saved prompt and run it every morning with one click	KPI targets registered + 4 weeks of data
Phase 3	Approval-flow (semi-automatic)	Candidates are extracted automatically; you approve with one click in the AI agent side’s approval flow to apply	Automation rules registered
Phase 4 (concept)	Fully automatic (not implemented)	A future vision of executing qualifying proposals automatically. Not offered today	—

Here is what changes in each phase.

Phase 1: One-shot

Every skill starts here. When you send a prompt through your AI agent environment, Picaro responds immediately.

Trigger: the operator sends a prompt manually in your AI agent environment
Parameters: specified inside each prompt (some fields, like monthly KPI targets, are referenced automatically once registered)
Application: bid, budget, and negative-keyword changes do not take effect until “execute” is approved

The point at this stage is to understand the output of each prompt and the meaning of the conditions the prompt specifies (caps on increase or decrease, scope, target count). These caps are specified in the body of each prompt, and you can change the numbers to fit your operation.

Phase 2: Daily-executable

In this phase, the same prompt can be re-run with one click every morning. The prompt is registered as a saved prompt on the Picaro side and invoked from your AI agent environment.

Prerequisites to move from Phase 1 to 2

Monthly KPI targets are registered (no need to retype the target ACoS each time)
At least four weeks of recent data have accumulated (so period-over-period comparison is meaningful)
Output quality has been validated through one-shot execution
The prompt is registered as a saved prompt

Skills that become practical in Phase 2 (examples)

Period-over-period KPI comparison on the dashboard (daily check routine)
Drift alerts for favorite keywords
Anomaly detection (notification of sudden KPI shifts)
Drift monitoring on bid recommendations
Mid-month KPI progress checks

The goal of Phase 2 is a five-minute morning routine that runs “dashboard KPI -> anomaly detection -> favorite keyword drift” in sequence.

Phase 3: Approval-flow (semi-automatic)

Candidate extraction is fully automated, and only the final approval needs a single click from a human. Proposals are generated on a schedule; you review them in your AI agent environment (Picaro connection) and, once you approve, the change is applied.

Prerequisites to move from Phase 2 to 3

Automation rules are registered explicitly (scope, item cap, monetary cap)
An approval flow (review the proposal, then approve) is built into your operation
Caps such as adjustment magnitude are written down (for example, +20% / -30% specified in each prompt)

Skills that move to automatic candidate extraction in Phase 3 (examples)

Bid reductions based on ACoS (weekly, cap of 100 items per week, one-click approval)
Adding negative keywords for high-ACoS / zero-conversion terms (weekly, up to 50 items)
Daily budget adjustment based on spend rate and ACoS (daily, within the monetary cap)
Auto-to-Manual promotion (weekly automatic candidate extraction plus manual approval)
Monthly report auto-generation

Because these actions move money directly, Phase 3 always routes through the approval flow. Nothing applies without approval.

Phase 4: Fully automatic (concept stage, not implemented)

The vision is that proposals with an established approval track record would execute without human involvement. Realizing it would require mechanisms to judge whether automatic execution is appropriate and to roll back automatically on anomalies — these are still under consideration.

Skills envisioned as part of the concept (examples):

Automatic bidding for proposals that meet the conditions
Automatic budget acceleration during sale periods
Automatic campaign pause when inventory runs out
Competitor ASIN monitoring with automatic response candidate generation

Even if full automation is realized, actions tied to strategic decisions — such as creating a new campaign — are intended to remain in human hands. The idea is not to automate “everything,” only the areas where accountability is clearly scoped.

Caps on changes (conditions you specify in the prompt)

As automation deepens, each prompt can specify conditions that prevent overshoot. These are specified in the body of the prompt and can be adjusted to fit your operation.

Item cap: how many targets a single batch can touch (for example, negative keywords up to 50 items)
Monetary cap: a per-proposal amount guideline. A hard ceiling on total ad spend is set in Amazon’s Ad Console portfolio budget
Adjustment magnitude cap: specify that a single adjustment does not exceed +20% / -30%
Scope: limit targets at the account / portfolio / campaign / keyword level
Whether approval is required: decide, per skill, whether to approve each time

These are conditions to hold proposals in check; the AI proposes within the range specified in the prompt. There is no system-enforced “safety device” that hard-stops execution, so state any limits you want respected explicitly in your request (prompt).

Why the phased approach

There are three reasons full automation is not offered from day one.

Data accumulation is required: the precision of proposals improves as approval / rejection history builds up. Operating across Phases 1 to 3 is the foundation for future automation.
Operational alignment: existing workflows, internal approval routines, and monthly reporting cycles do not fit full automation overnight. Organizational practices evolve alongside the phases.
Clear accountability: “the AI decided” leaves accountability ambiguous. Going through propose -> approve -> execute records who decided what at which stage, providing the transparency internal governance requires.

Where to look to gauge the current phase

Signal	Likely phase
Sending prompts on demand in your AI agent environment and reviewing results	Phase 1
Morning routine is fixed and uses saved prompts	Phase 2
Automation rules are registered and proposal approvals are running	Phase 3

Moving from Phase 1 to Phase 2 happens by registering monthly KPI targets and creating saved prompts. Moving from Phase 2 to Phase 3 requires explicit automation rules.

What to do next

Save rules to systematize proposal approval and monitoring — the on-ramp from Phase 2 to 3
Set targets for sales, ad spend, and ACoS — the prerequisite for Phase 1 to 2
View KPIs on the dashboard — the core of the Phase 2 daily routine