Actioning Amazon Ads

Actioning turns insights into changes you can push straight to Amazon Ads — no copy‑pasting between Aliveo and the Amazon Ads console. When you ask Aliveo to optimize budgets, bids, bid strategies, statuses, or keywords across Sponsored Products (SP), Sponsored Brands (SB), and Sponsored Display (SD), it produces an action table: a clear, side‑by‑side preview of current vs proposed values. You review the rows, adjust anything you like, select what you want, and click Apply — Aliveo then sends those changes to Amazon on your behalf and reports back which ones succeeded.

This guide covers:

How action tables work — reviewing and applying changes safely.
What you can action — budgets, bids, bid strategies, statuses, and keywords across SP, SB, and SD.
Creating actions on the fly with /actioning in Analytics Chat.
Building repeatable Custom Actioning Workflows for campaign/ad‑group and keyword optimization.
Example prompts, from simple one‑liners to multi‑step optimization workflows.

Who can action. Actioning is available only for accounts Aliveo manages for you (not read-only connections), and only to users who have permission:

Primary users can action by default.

Secondary users can action only once a primary user grants them permission for that account or platform — see Actioning Permissions.

Read-only users can never action.

Aliveo checks this up front. If you're not eligible for an account, it still analyzes and recommends — but the Apply controls won't appear.

1. The Action Table

Whenever a chat answer or an agent step proposes changes to your account, Aliveo renders them as an action table. This is the single place where you review and apply changes — it looks and behaves the same whether the table came from a chat, a /actioning request, or a scheduled workflow.

Every action table carries the same instruction at the top:

Select rows, adjust the recommended values if needed, and click Apply.

Each row pairs a proposed value with its current value, and you can adjust any proposed value inline before applying. Select the rows you want, and the Apply button reflects how many you've chosen.

Anatomy of the table

Element	What it does
Row checkboxes	Select exactly which recommendations you want to apply. A header checkbox selects/deselects all eligible rows.
Proposed value (editable)	The recommended new value. You can edit it inline before applying — change a bid, round a budget, switch a status, etc.
Current value	Shown beneath the proposed value (e.g. Current: 25.00) so you always see what you're changing from.
Status badge	Pending until applied, then Applied once the change reaches Amazon successfully.
Search / filter / sort	Narrow large tables with the Filter rows… box, per‑column filters, and sortable headers — handy when a table has hundreds of keywords.

Proposed cells adapt to the field type: amounts like bids and budgets use a number input (with valid min/max bounds), while statuses, bid strategies, and match types use a dropdown of allowed values.

Reviewing → applying

Review each row — compare the proposed value against the current one.
Adjust any proposed value inline if you'd like something different from what was recommended.
Select the rows you want with their checkboxes (or use the header checkbox to take everything).
Click Apply (N) — the number reflects how many rows you've selected. The button stays disabled until at least one row is selected.
Aliveo validates the changes, pushes them to Amazon, and shows the result:
- Actions applied — "Selected recommendations were sent for execution."
- Some actions could not be applied — a few rows failed validation; review them and retry.
- Failed to apply actions — try again, or Refresh the table to reload the latest recommendations.

Applied rows flip to an Applied badge and their checkboxes lock, so you can't double‑apply the same change. You can keep working through the rest of the table, or hit Refresh to pull fresh recommendations against the now‑updated account.

Tip — expand for big tables. Use the maximize button to open the table full‑screen for easier review, and the refresh button to reload recommendations after the account changes.

Built‑in guardrails

Before anything reaches Amazon, Aliveo runs each row through safety checks so you don't get cryptic API errors:

No‑op protection — rows where the proposed value equals the current value are skipped.
Drift protection — if the account changed since the recommendation was generated (the "current" value no longer matches Amazon), the row is held back rather than overwriting newer data.
Amazon rule enforcement — Aliveo pre‑checks Amazon Ads' constraints (see Things to know) and rejects rows that would violate them, with a plain‑language reason on the row.
Change log & restore — every applied change is recorded (what changed, from → to, who, and when) in the Action Log, where you can track its impact and restore most changes with one click.

2. What You Can Action

Aliveo can action the most common day‑to‑day optimizations across all three Amazon ad products — Sponsored Products, Sponsored Brands, and Sponsored Display. Campaigns, ad groups, and product ads can be updated; keywords and negative keywords can be both created and updated.

Campaigns

Action	Values
Daily budget	Set a new daily spend cap (in the account's currency)
Status	`ENABLED` / `PAUSED`
Bid strategy	SP: `MANUAL`, `RULE_BASED`, `SALES_DOWN_ONLY`, `SALES_UP_AND_DOWN` · SB: `MANUAL`, `SALES_UP_AND_DOWN` · SD: not available

Ad Groups

Action	Values
Default bid	Set the ad group's default bid (SP and SD — SB ad groups are status‑only)
Status	`ENABLED` / `PAUSED`

Keywords (SP & SB)

Action	Values
Bid	Set a keyword‑level bid
Status	`ENABLED` / `PAUSED`
Create keyword	Add new targeting keywords to an ad group — match type `BROAD`, `EXACT`, or `PHRASE` (defaults to `BROAD`)

Negative Keywords (SP & SB)

Action	Values
Create negative keyword	Match type `EXACT` or `PHRASE` (defaults to `EXACT`)
Pause / re‑activate	`ENABLED` / `PAUSED`

Product Ads

Action	Values
Status	`ENABLED` / `PAUSED` — pause underperforming ASINs or re‑enable them

Ad products differ. Every entity belongs to one ad product, and the levers move with it: Sponsored Display has no keywords and no bid strategy; Sponsored Brands ad groups carry no default bid (bids live on the keywords, so new SB keywords need a bid); Sponsored Products has the full surface. Aliveo knows which product each entity belongs to and only proposes what that product supports.

Note: Out of the box, actioning updates existing campaigns, ad groups, and product ads and creates keywords and negatives within them. Looking to create campaigns or ad groups from Aliveo too? Talk to us — we can set this up as part of a custom workflow for your account.

3. `/actioning` in Analytics Chat

The fastest way to make a one‑off change is right inside a chat. Type / in the chat input to open the mode menu, then pick actioning.

Selecting actioning tells Aliveo you want to change the account, not just analyze it. It then makes the actioning tools available so your request comes back as a ready‑to‑apply action table instead of a plain answer.

You don't always have to type /actioning — Aliveo also recognizes actioning intent from phrasing like "pause…", "reduce the daily budget…", "increase the bid…", or "add … as a negative keyword." Using the slash command simply makes your intent explicit.

Things you can say in actioning mode:

"Pause the 2 Amazon campaigns with the highest spend and lowest sales."
"Reduce the daily budget by 20% for campaigns spending over $20 a day with ACOS above 20%."
"Increase the bid by 15% on the top 5 keywords by conversion rate."
"Add competitor brand as an exact‑match negative keyword."

Aliveo finds the relevant entities, builds the action table, and hands it to you to review and apply.

The other chat modes, for reference: info (answer from context without running analysis), brainstorming (strategy and market exploration), web_search (pull in live external info), and create_agent (draft an automation agent). Only actioning produces apply‑ready action tables.

4. Custom Actioning Workflows

For optimizations you want to run repeatedly — or assemble as a multi‑step routine — use a Custom Actioning Workflow agent instead of a one‑off chat. These are blank‑slate agent templates pre‑wired for actioning: you describe the analysis and the change, and the workflow produces an action table each time it runs.

You'll find them in the agent creation flow under Step 2: Template Selection (look in the Optimization Agents group). There are two, depending on what you want to action:

Custom Actioning Workflow (Campaign / Ad‑Group)

Blank‑slate actioning workflow for Amazon Ads at the campaign / ad‑group level (budget, bid strategy, bid, status) across SP/SB/SD.

Use it for: daily budgets, bid strategy switches, ad‑group default bids, and pausing/enabling campaigns, ad groups, and product ads.
Set up with: Ad Performance data.

Custom Actioning Workflow [Keywords]

Blank‑slate keyword‑level actioning workflow for Amazon Ads (keyword bid, status, negative keywords) for SP/SB.

Use it for: keyword bids, keyword status (pause/activate), and negative keyword management.
Set up with: Keyword data.

How to build one

Both templates drop you into the workflow playground with the actioning tools already switched on, so you assemble the steps yourself:

Shared Context — define anything common across steps (definitions, thresholds, the time window, e.g. "use the last 30 days").
Analyze steps — do the data work: pull performance, compute metrics like ACOS/ROAS/conversion rate, rank entities, and produce the proposed changes as an action table.
Summary step — (optional, always last) logically summarize the combined output of the analyze steps.

Because these are actioning workflows, the analyze step that proposes changes renders an action table in the workflow output — you still review and Apply exactly as you would in chat. (Custom Actioning Workflows are set up as one‑off agents so you stay in control of when changes are proposed.)

For the general agent‑building mechanics — accounts, data context, grouping, steps, preview, and scheduling — see Creating an Agent.

5. Examples — From Simple to Complex

Every example below is a single natural‑language request. Aliveo handles the analysis (finding the entities, computing the metric, ranking) and produces the action table — you review and apply.

Simple, single‑step actions

These map directly to one type of change:

Budget — "Set the daily budget to $100 on my top 2 Sponsored Products campaigns by sales."
Status (campaign) — "Pause the 2 campaigns with the highest spend and lowest sales."
Status (product ad) — "Pause the product ads with zero orders and more than $50 spend in the last 30 days."
Bid — "Find the keywords with ACOS above 40% and reduce their bid by 25%."
Negative keyword (create) — "Add competitor brand as an exact‑match negative keyword to the Brand Defense ad group."

Metric‑driven optimizations

These combine an analysis with a change — Aliveo computes the metric, ranks, filters, then proposes the action:

ACOS budget trim — "Reduce daily budget by 20% for campaigns spending over $20 a day with ACOS above 20%."
Wasted‑spend cleanup — "Identify keywords with over $25 spend and zero orders in the last 30 days, and pause them."
Bid efficiency — "Raise the default bid by 10% on the 3 ad groups with the lowest ACOS."

Keyword discovery → action

These turn search‑term analysis into keyword changes:

Harvest winners — "Add the top 5 converting search terms as exact‑match keywords."
Block losers — "Add the 3 search terms with the most clicks but lowest purchases as exact‑match negative keywords."

Bid‑strategy automation

Protect efficiency — "Switch the Sponsored Products campaigns with ACOS above 20% to the SALES_DOWN_ONLY bid strategy."
Chase volume — "Move my most efficient SP campaigns to SALES_UP_AND_DOWN."
Back to manual — "Switch campaigns with erratic ACOS back to MANUAL bidding."

Multi‑step workflows you can assemble

In a Custom Actioning Workflow, you can chain several of the above into one routine. A few patterns that work well:

Weekly keyword hygiene (Keywords template) — (1) Analyze: find zero‑order, high‑spend keywords → pause them. (2) Analyze: find high‑converting search terms → add them as exact‑match keywords. (3) Analyze: find irrelevant search terms → add them as negatives. (4) Summarize what changed.
ACOS guardrail (Campaign template) — (1) Analyze: rank campaigns by ACOS, trim daily budget on the worst performers. (2) Analyze: rank by ROAS, raise budgets on the best. (3) Analyze: move consistently inefficient campaigns to SALES_DOWN_ONLY. (4) Summarize.
Catalog cleanup (Campaign template) — flag ASINs (product ads) with sustained spend and no orders, and pause them alongside their weakest keywords.

Each analyze step produces its own action table, so you can review and apply each set of changes independently.

Advanced, tailored workflows

Beyond the self‑serve patterns above, we build custom actioning workflows tuned to your account's strategy and scale. If any of these sound like what you need, talk to us and we'll set them up for you:

Sophisticated bid & budget management — rule‑based and goal‑seeking bid/budget logic that reacts to ACOS, ROAS, and conversion trends rather than simple top‑N rules.
Pacing — keep daily spend on track to budget targets across the period, with proposed budget adjustments when an account is over‑ or under‑pacing.
Keyword deduping — detect duplicate or overlapping keywords competing across campaigns and ad groups, and consolidate them so you stop bidding against yourself.
Account structuring — organize keywords into a clean, intentional structure (e.g. discovery vs. brand vs. competitor, by match type), proposing the moves and negatives needed to get there.

These run as the same review‑and‑apply action tables, so even sophisticated logic stays fully under your control — nothing changes until you click Apply.

6. Things to Know — Amazon's Rules

Aliveo enforces Amazon Ads' own constraints up front, so invalid rows are caught before they're sent. If you see a row rejected with a reason, it's usually one of these:

Ad products

Every entity belongs to one ad product — Sponsored Products, Sponsored Brands, or Sponsored Display — and only that product's levers apply: SD has no keywords and no bid strategy, and SB ad groups have no default bid (bids live on the keywords).

Statuses

Only ENABLED / PAUSED can be set. ARCHIVED is read‑only — archived entities can't be changed or un‑archived through actioning, and you can't archive an entity from Aliveo. Everyday phrasing like "activate", "disable", or "turn off" is normalized to the right value automatically.

Bids & budgets

Amounts are in the account's currency and can't be negative. Amazon enforces its own per‑marketplace minimums for budgets and bids — rows below them are rejected by Amazon and Aliveo surfaces its reason on the row.

Bid strategy

Allowed strategies depend on the ad product — Sponsored Products: MANUAL, RULE_BASED, SALES_DOWN_ONLY, SALES_UP_AND_DOWN; Sponsored Brands: MANUAL, SALES_UP_AND_DOWN; Sponsored Display: none.

Keywords

Keywords are Sponsored Products and Sponsored Brands only.
Text and match type are fixed after creation — you can change a keyword's bid and status, but to change its text or match type, recreate it.
New keywords need text and a match type (BROAD / EXACT / PHRASE); new Sponsored Brands keywords also need a bid (SB has no ad‑group default bid to fall back on).
Negative keywords take EXACT or PHRASE match only (no broad negatives).

These guardrails mean you can safely propose broad changes; Aliveo simply skips the rows Amazon wouldn't accept and tells you why.

That's actioning — analyze, review, and apply Amazon Ads changes without leaving Aliveo. Start with a quick /actioning request in chat, and graduate to a Custom Actioning Workflow once you find an optimization worth repeating.