How does a Google Shopping priority funnel actually work?

Three Standard Shopping campaigns at three priority levels, with negative-keyword lists that route queries down to the right cost. HIGH ($0.05) catches junk that has no negatives blocking it. MEDIUM ($0.85) catches quality terms. LOW ($2.50) catches brand and competitor terms. Same product set in all three — what differs is the bid and the negative list, which together act as an intent router.

Why does the "highest priority, lowest bid" rule feel backwards?

Because Google's naming optimizes for a different use case (product ranking inside a single campaign). Practitioners who run priority funnels flip the model: same products in every tier, priority + negatives route queries by intent. The naming convention makes it counter-intuitive, which is part of why so few accounts run this structure.

What are the five common settings that break the funnel?

HIGH out of budget mid-day (junk overflows to MEDIUM at 17x cost), MEDIUM out of budget (quality terms fall to LOW at 3x cost), device adjustment mismatch (mobile −100% in HIGH only routes mobile junk to MEDIUM), ad-schedule mismatch (HIGH paused off-hours sends off-hours junk to MEDIUM), product-set mismatch (HIGH has 80 SKUs while MEDIUM has 100, so the missing 20 serve in MEDIUM at full price). All five are silent — standard reports do not flag them.

How do you maintain a priority funnel over time?

You cannot, by hand alone. The funnel breaks the moment a teammate adds a campaign-level negative, the feed drops a SKU from one tier, or budget caps mid-day. The reliable pattern is an Ads Script that runs hourly and emails on drift across budget, schedule, device, product set, and negative-list parity.

B2B SHOPPING · LIVE SIMULATOR

Most B2B Shopping accounts overpay 100x on junk clicks.

It's structural, not a bidding problem. Five common settings break the fix and reports won't tell you which one. Toggle the simulator. See your account in the gap between two ledgers. Take the PDF to your boss.

Watch the 3-minute explainer

flip between modes — same query stream, two universes

junk · cheap stickers

quality · ul listed labels

brand · brady m611

competitor · dymo

Speed1.00×

HIGH PRIORITY · $0.05

$0.00

junk catcher

MEDIUM · $0.85

$0.00

quality terms

LOW · $2.50

$0.00

brand + competitor

LEAK OVERAGE

$0.00

paid above the right tier's bid

LIVE LEDGER

Same junk. Two ledgers.

0 junk clicks priced both ways · 4% baseline neg-list leak modeled

NO FUNNEL · 1 campaign @ $5.00

$0.00

WITH FUNNEL · actual cost

$0.00

Give it a few seconds. The green bar is your funnel. The red bar is most accounts.

$0.00SAVED · LIVE

FIVE WAYS THIS BREAKSclick any card. watch the cost stack up live.

HIGH · BUDGET OUT

HIGH dies at 11am

Junk catcher offline rest of day
Junk clicks jump $0.05 → $0.85
Backup HIGH-B at $0.01 fixes it

MEDIUM · BUDGET OUT

MEDIUM caps mid-day

Quality terms tumble to LOW at $2.50
3x overpay on category research
Increase budget or run a backup

MOBILE · MISMATCH

HIGH excludes mobile, others don't

Phone users bypass HIGH entirely
Mobile junk lands in MEDIUM at full bid
Mirror device adjustments across all 3

SCHEDULE · MISMATCH

HIGH paused 7pm–7am, MEDIUM 24/7

Off-hours junk skips dark HIGH
Routes to MEDIUM at $0.85 till morning
Same dayparting on all 3 tiers

PRODUCTS · MISMATCH

HIGH has 80 SKUs, MEDIUM has 100

Missing 20 SKUs can't match HIGH
Their queries serve in MEDIUM at full bid
Identical product sets + inventory filters

FIX · BACKUP CAMPAIGN

Duplicate HIGH at $0.01

Same negatives, same products
Catches junk when primary runs out
Cheap insurance against the #1 break

FUNNEL · CLEAN

Junk in HIGH at $0.05. Quality in MEDIUM. Brand in LOW. Toggle anything below to see what breaks.

FREE · 2-PAGE PDF

Send this to your director.

Funnel diagram. Savings math. Five break scenarios. The exact campaign-structure recommendation. On 2 pages a director will actually read.

No list. One email. Single attachment. Unsubscribe link inside.

PROOF · ONE QUERY

"heat shrink label printer" — Brady owns the SERP.

6 Sponsored Products. Top of Sponsored Result. Zero competitor presence. This is what a working funnel buys you on a brand-adjacent query.

SCREENSHOT PLACEHOLDERDrop your SERP screenshot at /heat-shrink-serp.png in /public to swap this out.

Now break it. Five ways. One of them is happening in your account today.

HIGH hits its daily cap at 11am
Feed adds 4 SKUs to MEDIUM and skips HIGH
Someone sets mobile −100% in HIGH only
Dayparting on HIGH not mirrored in MEDIUM/LOW
A teammate adds a campaign-level neg by hand

Reports won't flag any of these. The funnel quietly stops being a funnel. You find out at quarterly review.

THE FIX · ADS SCRIPT

$99FOUNDER'S PRICE · FIRST 100 BUYERS

A 90-second drop-in that watches all five.

Google Ads script. Runs hourly inside your account. Emails you the second a tier drifts from spec.

Five checks: budget, schedule, device, product set, neg drift
Drop into Tools → Bulk Actions → Scripts
Email alerts on first failure
Source code commented + 12 months of updates
PDF setup guide included

$25one-time · lifetime updates

Pay Pay

card · paypal · ach also accepted at checkout

Refund if it doesn't pay for itself in week one

3 LISTS · 1 JOB EACH

Stop adding negatives at the campaign level.

SHARED LISTS · EDIT ONCE · APPLIES EVERYWHERE

UNIVERSAL JUNK
APPLY TO
HIGH + MEDIUM + LOW
JOB
Block terms you'd never bid on at any price
EXAMPLES
free diy amazon login torrent
PROMOTE-TO-MEDIUM
APPLY TO
HIGH only
JOB
Route quality terms down to MEDIUM
EXAMPLES
ul listed nfpa thermal transfer
PROMOTE-TO-LOW
APPLY TO
HIGH + MEDIUM
JOB
Route brand + competitor terms down to LOW
EXAMPLES
brady bradyid mecco dymo

Edit one list. Every campaign using it updates. No drift. No "did I add this to MEDIUM too?"

FAQ

Four questions everyone asks first.

How does this work with audiences?

Audiences and query routing are separate dimensions. They don't fight each other.

Overlay pattern. Add audience bid adjustments to all 3 tiers (+20% converters, −15% cold). Funnel logic untouched.
Parallel pattern. Clone the 3-tier funnel as RLSA-only. Run it next to your prospecting funnel. Each tuned for its own traffic.

Funnel cares about what was searched. Audiences care about who searched. Layer freely.

How do you scale across 10,000+ products?

Same architecture. More plumbing.

– Inventory filters per tier (forced parity)
– Custom labels for margin / velocity splits
– Supplemental feeds to lock product sets together
– Pivot-table audits for feed-vs-tier drift

The free Custom-label coverage planner walks through the schema. The Feed Pulse extension audits feed health so the routing stays clean.

Why doesn't everyone do this?

Priority levels were originally built to prioritize products, not queries:

– Best-sellers in HIGH
– Mid-tier in MEDIUM
– Long-tail clearance in LOW

Most accounts still use them that way. Google's docs read like that's the only way.

We flip it: same products in every tier. Priority + negatives route queries by intent, not products by importance.

"Highest priority carries the lowest bid" violates the naming convention. That's why 9 of 10 accounts miss it.

How do you maintain this over time?

You can't, by hand. The funnel breaks the moment one of these happens:

– Teammate adds a campaign-level neg by mistake
– Feed update drops a SKU out of one tier
– Budget caps mid-day
– Device adjustment changes in HIGH and not MEDIUM/LOW

Reports won't flag any of it. Jump to the script ↑ for the hourly drift watcher.

B2B Shopping · negative priority funnel · live demo. Companion extension: Feed Pulse · SKU Mapper.Standard Shopping · negative priority funnel · live demo