Amazon PPC Audit Checklist: What to Check Before Touching Bids

1.Start With the Question the Audit Must Answer

A PPC audit is useful only when it answers a business question.

For a mature account it could be: Where are we paying above our allowable acquisition cost?

For a new product launch it could be: Which search terms have proved enough purchase intent for tighter Exact control?

For a multi-product brand: Which product line can take more spend, and which one is being subsidized by the rest of the catalog?

For rising ACoS: Did CPC rise, conversion fall, placement mix change, or did a few high-spend targets deteriorate?

Do not start by sorting the ACoS column. Start with product economics and campaign purpose.

An account can have 28% ACoS and lose money. Another can run at 42% and still profitable. Price, Amazon fees, landed cost, returns, promotions, repeat purchase behavior, and the profit target decide that.

2.Build The Audit from The Right Reports

Amazon PPC Campaign Manager is useful for daily checks. A full audit needs downloadable reporting because the loss often sits below the campaign total.

Amazon currently lists Sponsored Products reports for search terms, targeting, advertised products, placements, performance over time, and purchased products.

Amazon also supports bulk operations for editing campaign elements through a downloaded file and re-uploading the changes.

Use more than one date range. A 30-day view is good for current behavior, while 60 or 90 days gives more evidence for slower targets.

3.Set Break-Even ACoS

ACoS is ad spend divided by attributed ad sales. Before you label 35% ACoS good or bad, calculate how much advertising cost the SKU can carry.

Take a sample product selling for £29.99:

Break-even ACoS is:

If the business wants to keep £4.99 contribution after advertising, the allowable ad cost per order drops to £7.50.

Target ACoS becomes:

Now convert that allowable ad cost into a CPC reference.

If paid traffic converts at 12%:

The £0.90 figure is not a universal bid. A target converting at 20% can carry a higher CPC. A target converting at 5% will carry high CPC. Placement also changes the effective CPC because bid adjustments can raise the bid for selected placements.

A second mistake is applying one target ACoS across the whole catalog. A replenishable product with repeat purchases may justify a different first-order acquisition cost than a one-time product. A launch can also run above steady-state targets for a limited period if the seller has set a clear spend limit and ranking goal.

4.Audit Portfolios and Product Lines

Start broad enough to see where the account is making or losing money. If a brand sells several distinct product lines, compare them separately. Do not let a profitable line hide an inefficient one.

One UK based account we audited had four Sponsored Products portfolios:

The interesting point is not simply that Line D had the lowest ACoS.

Line B converted at 24.5%, far above Line A’s 9.5%, but still carried a higher ACoS because the product value and CPC relationship were different. Line C had both weak conversion and the highest ACoS. Those require different responses.

If the portfolios were built badly, portfolio labels may not match the products inside them. Map campaigns to actual product lines first. One audit even had a historical Exact campaign sitting outside the correct portfolio, which weakened reporting outcome.

5.Separate Ad Types Performance

Audit Sponsored Products, Sponsored Brands, and Sponsored Display separately before using a blended total.

Sponsored Products received most of the budget, which is common in seller accounts because Sponsored Products can cover manual keywords, Auto targeting, categories, competitor products, and individual advertised SKUs.

The important point is not that Sponsored Products had 77.1% of spend.

The important point is that both major ad types were running around 50% ACoS and below 4% conversion.

That told us the problem was broader than a single Sponsored Brands campaign.

We had to go deeper into campaign performance, traffic type, placements, and product conversion.

6.Find Campaign Budget Concentration

Sort campaigns by spend before you start changing bids. A few campaigns usually control a large share of the account outcome.

In one audit, a single campaign spent $4,699.03, equal to 59.4% of Sponsored Products spend and 45.8% of all ad spend. Its conversion rate was only 4.21%.

In another account, one Broad campaign used £1,311.09, or 27.6% of Sponsored Products spend. The top five Sponsored Products campaigns accounted for about 77% of Sponsored Products spend.

That tells you where to spend audit time.

Review those campaigns first:

• 1. What target or targets drive the spend?
• 2. Branded and non-branded terms performance?
• 3. Which placements receive the spend?
• 4. Is one weak ASIN sharing a bid with stronger variants?
• 5. Does the campaign run out of daily budget?
• 6. Are converting search terms isolated elsewhere?

A high-spend campaign is not automatically bad. The campaign may be the account’s main sales source. The point is that bid optimization on a $5,000 campaign matters far more than a $50 campaign.

Spend concentration is normal when one SKU is the clear sales leader. Do not spread budget evenly for cosmetic account balance. Protect the proven SKU while checking whether weaker products deserve their own controlled tests.

7.Audit Match Types and Targeting by Job

Compare Exact, Phrase, Broad, Auto, category, and product targeting by the job each one is doing. The audit should show where spend is concentrated and whether each traffic source is earning that spend.

A large US audit produced this mix:

Broad carried 46.1% of spend at 28.6% ACoS, close to Exact at 27.5%.

But the structure still had a control gap. Broad and Auto together held most of the discovery traffic. Proven queries needed to be separated where tighter bidding and routing would improve control.

Broad is useful when it finds relevant demand at acceptable economics.

Phrase is useful when the word order and surrounding intent matter.

Exact is useful when a proven keyword deserves direct bid and placement control.

Auto is useful for discovery and for matching patterns Amazon sees around the product.

Product targeting is useful when category or individual product-page traffic has acceptable economics.

Match-type comparisons can be distorted by branded traffic. A Phrase campaign full of brand searches can post an extremely low ACoS while a generic Exact campaign looks expensive. Split brand from non-brand before making account-wide conclusions.

8.Audit Customer Search Terms before Changing Targets

Move from targets to the customer queries that actually received the clicks. This is where you find proven demand, expensive converters, zero-order spend, and searches that need tighter routing.

A Broad keyword can generate dozens or hundreds of different search terms. An Auto target can generate search terms and product-detail-page traffic. A category target can generate product queries that need product-level negatives rather than keyword negatives.

Start with the customer search term report.

Sort by spend, then classify each consolidated search term into four practical groups:

Profitable and proven. Enough orders, acceptable ACoS, relevant intent.

Converting but expensive. It sells, but the CPC, conversion rate, or placement mix pushes cost above target.

Clicks with no orders. It has spent enough to justify a decision.

The word consolidated matters. If the same customer query appears across four campaigns, reviewing each row separately can make every row look too small to act on.

One audit found the same anonymized query across four campaigns, with 231 clicks, £217.38 spend, £495.43 sales, 16 orders, and 43.9% ACoS after consolidation. Another appeared in two campaigns with 202 clicks, £168.48 spend, £576.19 sales, and 18 orders.

A one-campaign view would have missed the full economics.

Consolidating across the account can also hide product differences. The same query may work for one ASIN and fail for another. Consolidate first to see total exposure, then split by advertised product before applying account-wide negatives.

9.Build Negative Targeting from Evidence

Build negatives from evidence, not from a zero-order filter. Audit click count, spend, relevance, product conversion, and existing negatives before deciding what to block.

In one account, 1,612 search-term rows produced clicks but no order and consumed £2,049.86. Only 13 rows had reached 10 or more clicks, worth £134.46, and seven had reached 15 or more clicks.

Use the 10 to 15 click rule as a screening rule:

• • 15+ clicks, zero orders: usually block the exact query or product target unless there is a clear reason to keep testing.
• • 10 to 14 clicks, zero orders: review relevance, CPC, listing state, and economics.
• • Below 10 clicks: keep collecting data unless the query is plainly irrelevant.

A larger US account spent $11,825.63 across 12,056 zero-sale terms and 18,803 clicks. But $5,956.37 came from one-click terms. Only 129 terms had 10 or more clicks with no sale, accounting for $1,177.74.

When low-volume waste is spread across thousands of queries, run one-word, two-word, and three-word N-gram analysis. Repeated irrelevant modifiers can expose a pattern that individual search terms never reach enough clicks to show. Check relevance before using negative Phrase because it can block profitable searches too.

10.Harvest Proven Search Terms into Tighter Campaigns

Use discovery campaigns to identify search terms that have already proved they can sell. Then decide which terms deserve direct Exact control because their order volume and ACoS are strong enough to manage separately.

In one 30-day account, 20 search terms discovered through Broad, Phrase, or Auto traffic were not already active as Exact targets. Together they produced $10,234.74 in attributed sales from $1,542.17 spend, 560 orders, 22.8% conversion, and 15.1% ACoS.

In another account, 35 search terms met a stricter rule of at least two orders and ACoS at or below 30%, while not already existing as Exact keywords in the same product-line portfolio. They generated £2,813.36 in sales and 121 orders.

Those are the terms worth reviewing for Exact isolation.

A practical harvesting process is:

• 1. Consolidate the customer search term across campaigns.
• 2. Confirm that orders came from the intended advertised product.
• 3. Separate branded, generic, and competitor intent.
• 4. Check whether the search term already exists as Exact in the relevant portfolio.
• 5. Create the Exact target if it needs direct control.
• 6. Start near an observed CPC that fits product economics.
• 7. Add negative Exact in the discovery campaign only when you want to control routing.
• 8. Watch placement and CPC after the move.

Do not remove every winning term from discovery automatically. A Broad or Auto campaign can still find profitable variants around the same theme. The negative Exact blocks only that exact customer query, while discovery continues around it.

11.Audit Placements Campaign by Campaign

Compare Top of Search, Rest of Search, and Product Pages inside each campaign before changing the base bid. A campaign can look expensive overall while one placement is highly efficient and another is causing most of the loss.

Amazon currently supports Sponsored Products placement bid adjustments up to 900%. A placement modifier should be earned by placement economics.

One anonymized Exact campaign had:

That campaign had a clear case for testing more Top of Search exposure without raising the base bid across every placement.

Another Exact campaign was even more extreme:

Raising the base bid there would pay more for Product Pages and Rest of Search. The best thing here would be to increase top of search placement and other two to zero.

A placement can look excellent on ten clicks. Do not add a large modifier from a tiny sample. Test in controlled increments and recheck CPC, conversion, and total spend after the change.

12.Audit Advertised ASINs and Variants

Break campaign performance down by advertised ASIN. The audit should show which products are carrying sales, which variants convert poorly, and whether shared campaigns are forcing the same bid logic onto products with very different economics.

In one large account, an anonymized high-spend ASIN spent $2,047.40, generated $11,596.70 in sales, converted at 21.1%, and ran at 17.7% ACoS.

Another high-spend ASIN spent $1,528.16, generated only $2,193.14 in sales, converted at 2.0%, and ran at 69.7% ACoS.

Audit each advertised ASIN for:

• • spend
• • attributed sales
• • orders
• • conversion rate
• • CPC
• • ACoS
• • advertised-SKU sales
• • other-SKU sales after the click
• • inventory state
• • price
• • rating and review count
• • Buy Box
• • variation relationship

The purchased-product view matters too. In the same account, $112,489.18 of sales came from the advertised SKU and $10,327.45, or 8.4%, came from other SKUs after the ad click.

Do not call all cross-SKU sales bad. A shopper may click one variation and buy another. The audit question is whether those purchases support the campaign’s commercial goal.

ASIN-level attribution can look strange when variations share traffic or shoppers move between related SKUs. Review parent-child structure and purchased-product data before cutting an advertised product because direct SKU sales look weak.

13.Split Automatic Targeting into its Subtypes

Split Auto into Close Match, Loose Match, Substitutes, and Complements before judging automatic targeting as one block. The audit should identify which subtype is producing useful discovery and which subtype is consuming spend without enough orders.

Sponsored Products automatic targeting can behave very differently across Close Match, Loose Match, Substitutes, and Complements.

A real account showed:

.

If all auto match types are in single campaign, split them into their own dedicated campaign for more control over bids and budgets. Then harvest proven search terms and product targets from the stronger groups.

14.Audit Product Targeting Separately

Review product targeting apart from keyword targeting. Compare category targets, individual competitor products, own-ASIN defense, and the product-page traffic each group is buying.

Check individual product targets, category targets, category refinements, competitor targets, own-ASIN defense, product-page placement, and purchased products after the click.

In one portfolio, ASIN targeting produced £606.76 spend, £2,568.34 sales, 83 orders, and 23.6% ACoS. Category targeting produced £718.75 spend, £2,431.27 sales, 81 orders, and 29.6% ACoS.

Those results supported keeping both methods but not using identical bids.

Competitor targets may support new-customer acquisition or cross-SKU sales. Include purchased-product and New-to-Brand data when those metrics match the campaign goal.

15.Separate Branded and Non-branded Traffic

Separate branded demand from generic acquisition before using either to judge account efficiency. Brand traffic often converts far more cheaply and can make generic campaigns look healthier than they really are.

One audit had branded Phrase traffic at 3.3% ACoS while the main Broad campaign ran at 37.1%.

Report branded, generic, competitor, and own-ASIN defense separately when spend is material. A 5% branded ACoS does not justify raising bids on generic terms when all are mixed together.

A new brand may have too little branded volume to justify separate structure. Split it once branded demand can distort generic reporting.

16.Check Overlap without Deleting Duplicates Blindly

Check overlap at the customer-query level rather than deleting duplicate targets automatically. The audit should show when several campaigns are buying the same intent and whether each campaign has a clear reason to do so.

The same keyword can exist in more than one campaign for valid reasons:

• • different advertised ASINs
• • different match types
• • different placements
• • different bidding strategies
• • ranking versus profit campaigns
• • separate brand and generic routes
• • different marketplaces

The problem starts when the same customer query receives spend through several campaigns and nobody can explain why.

In one audit, the same customer query appeared in four campaigns and accumulated 231 clicks and £217.38 spend. Another appeared in three campaigns and carried 206 clicks and £151.02 spend.

Audit overlap at the customer query + advertised product + campaign purpose level.

If two campaigns are bidding on the same intent with different economics, decide which campaign should own the query. Use Exact isolation and negative Exact routing where appropriate.

17.Check Retail Conditions before Blaming PPC

Check the retail offer whenever relevant ad traffic is converting poorly. Price, reviews, Buy Box, stock, images, delivery, and variation setup can make the same click far more or less valuable.

One 60-day account had 10,491 clicks and 397 orders, a 3.78% paid conversion rate, with $10,250.54 spend and $19,671.14 in attributed sales.

The audit still found useful PPC changes but bid work alone could not explain the low conversion.

Before cutting traffic, check:

• • Buy Box ownership
• • stock availability
• • price versus the current competitive set
• • coupon or promotion state
• • rating and recent review pattern
• • main image
• • title relevance
• • image stack
• • variation selection
• • delivery promise
• • A+ Content
• • category compliance or suppression
• • recent listing changes

This is where PPC and listing work meet. A search term can be highly relevant and still run at bad ACoS because the detail page cannot convert the shopper.

For related listing work, Ecom Brainly also covers Amazon SEO and listing optimization as a separate operating area.

Do not blame the listing every time conversion is low. Weak search intent can also cause low conversion. Compare conversion by search term, target, and placement. If only one traffic source converts badly, fix the traffic before rebuilding the page.

18.Audit Budgets and Campaign States

Review budgets only after you know whether the campaigns deserve more spend. Then check campaign states so paused, budget-limited, or underfunded campaigns are interpreted in the context of their historical performance and current business conditions.

A profitable target can still lose sales if the campaign runs out of budget early. An inefficient campaign can also consume its full budget every day and crowd out better campaigns.

Check:

• • daily budget
• • actual daily spend
• • budget utilization
• • budget changes during the period
• • campaign state
• • portfolio budget caps
• • campaigns that frequently exhaust budget
• • profitable campaigns with limited spend
• • paused campaigns that still matter to the product line

Campaign states deserve special attention.

One audit found an entire product line with strong historical conversion but all major campaigns paused. Auto Close had run at about 25.4% ACoS, while other parts of the line were materially more expensive. Before reactivating anything, the account needed an inventory, Buy Box, price, and business-intent check. The sensible restart order was the strongest Auto Close traffic first, followed by selected Exact and category tests.

A paused campaign is not automatically a missed opportunity. The seller may have paused it because inventory is low or the product is being discontinued.

19.Review Sponsored Brands and Sponsored Display

Judge Sponsored Brands and Sponsored Display by their own campaign goals and traffic. Audit Sponsored Brands search terms separately, and for Sponsored Display check whether campaigns exist before deciding whether there is anything to fix or expand.

In the UK audit above, the only Sponsored Brands campaign with delivery produced £149.87 sales from £37.75 spend at 25.2% ACoS. Yet £29.16 of spend came from zero-order search terms.

If that campaign resumes, the first thing would be to negative targeting and theme separation. Keep brand, generic, and competitor intent out of one broad stream when they behave differently.

In another account, Sponsored Brands had 2,213 customer search terms with clicks, and 2,128 produced zero orders. Those zero-order terms consumed $1,710.52, equal to 73.0% of Sponsored Brands spend.

For Sponsored Display, audit product targeting, audience targeting, retargeting, spend, sales, New-to-Brand where relevant, and placement or audience performance available in the account.

Sponsored Brands and Sponsored Display can have goals beyond direct last-click efficiency. If the seller is funding upper-funnel acquisition, report the agreed KPI rather than judging everything by Sponsored Products ACoS.

20.Add TACoS Only when Total Sales Data Is Trustworthy

Add TACoS only after matching ad spend to trustworthy total-sales data for the same marketplace and date range. The audit should never create a precise TACoS from advertising sales alone.

Review ad spend, ad sales, total sales, ACoS, TACoS, and organic sales share on matching date ranges. ACoS can improve while the account becomes more dependent on ads, so both views matter.

TACoS can rise during a launch or ranking push even when the campaign is doing its assigned job. Judge it against product stage.

21.Use Bulk Operation Instead of Manual Optimization

Use bulk operations when account size makes manual review risky or too slow. The audit should turn large datasets into controlled action sheets for bids, negatives, states, placements, and new targets.

One US audit contained 106 campaigns, 14,071 unique search terms, 12,056 zero-sale terms, 129 immediate 10+ click negative candidates, and 44 mapped advertised-product groups.

For an account that size, keep an untouched source file, add analysis columns, create separate action sheets for bids, negatives, states, placements, and new targets, then upload controlled batches and review Amazon’s processing report.

Third-party tools such as Helium 10, Pacvue, Perpetua, and DataDive can speed analysis. The logic should still be explainable from product economics and ad data.

Bulk edits make mistakes faster too. A bad negative Phrase or incorrect bid formula can hit hundreds of rows. Keep the source file and validate the processing report.

How Often Should You Run An Amazon PPC Audit

Match audit frequency to the speed and volume of the account. Fast-moving budgets and stock need frequent checks, while deeper search-term, product-line, and structural decisions need enough clicks and orders to be reliable.

Run a deeper 30- or 60-day audit when taking over an account, after a material ACoS change, before a large seasonal period, before a large spend increase, or when years of campaigns have accumulated.

What A Good Amazon PPC Audit Should Leave You With

A good audit should leave the seller with specific decisions, not a generic instruction to lower ACoS. The final output should show where the money went, what performed, what failed, and what changes deserve priority.

You should know which product lines deserve more or less spend, which campaigns control most of the account outcome, which search terms are proven, which zero-order terms have enough evidence to block, which expensive converters need bid correction, which placements deserve different bids, which ASINs are carrying the account, and what should be changed first.

A useful Amazon PPC audit gets down to the level where product economics, customer query, advertised ASIN, bid, placement, and campaign purpose point to a clear action.