How to evaluate an Amazon PPC restructure before you approve one

Ask what problem the restructure is actually solving

"The account is messy" is not a problem. It is an aesthetic complaint. A restructure is only justified when the current structure prevents a specific decision you need to make.

There are really only a handful of structural problems worth a rebuild: you cannot tell which search terms drive profitable sales because everything sits in a few broad campaigns; budget caps are throttling your best converting terms while cheap discovery spend runs free; branded and non-branded traffic are blended, so your ACOS number is meaningless; or match types are competing against each other and inflating cost-per-click on terms you already own.

If the proposal cannot name which of those is happening in your account, with the report that shows it, you are being sold housekeeping.

Demand the baseline before anything changes

You cannot judge a restructure without a frozen snapshot of what came before it. Ask for a documented baseline over a window long enough to smooth out promotions and seasonality — usually the trailing 60 to 90 days.

The baseline should include, per ASIN and per campaign type:

• Spend, sales, ACOS and TACOS, so you can see paid performance in the context of total revenue.
• Conversion rate and click-through rate, separated for branded and non-branded terms.
• The share of sales coming from your top 20 converting search terms.
• Organic rank for the keywords you are defending, so a rank slide during the rebuild does not go unnoticed.

Find out what happens to your search-term history

This is the question most proposals skip. Amazon's algorithm rewards campaigns with conversion history. When you pause a mature campaign and rebuild it in a fresh one, you do not carry that history over — you start bidding for placements you were previously winning cheaply.

A competent plan protects history. It migrates in stages, keeps proven exact-match terms running in their existing campaigns while new structure is built alongside, and only sunsets the old campaigns once the new ones have demonstrated equivalent efficiency. If the plan is "pause everything Monday, launch the new build Tuesday," expect a dip and make sure the agency has told you how deep and how long.

Insist on a defined dip window and a rollback trigger

Every honest restructure has a learning period. What separates a professional plan from a gamble is whether the dip is quantified in advance.

Before approval, agree in writing on three numbers: how long the learning period is expected to last, how much efficiency loss is acceptable inside it, and the threshold at which you stop and roll back. Something like: two to three weeks of learning, ACOS up to a defined ceiling, and a rollback conversation if spend is above baseline with sales flat by day 21.

An agency that will not commit to a rollback trigger is asking you to absorb all the downside of their hypothesis.

Check that the restructure is tied to the listing, not just the campaign

PPC efficiency is capped by conversion rate. If your main image, title, bullets and A+ content are weak, a better campaign structure buys more expensive traffic for a page that still does not convert.

So the right question is not only "how will the campaigns be organised?" but "what changes on the detail page in the same window, and how will we tell the two effects apart?" Sequencing matters: fix obvious listing gaps first or run the changes in a deliberate order, otherwise you learn nothing from either.

Ask who is actually doing the work each week

Restructures are not a one-off build. The first month is daily bid and negative-keyword work, then weekly search-term harvesting. That is hands-on operator time, not a deliverable you receive once.

Get a name, a cadence, and a reporting format you can read in five minutes. If the person presenting the plan is not the person doing the work, ask to meet the person who is.