Google ShoppingPerformance MaxE-CommerceBicycle RetailOngoing
670 Purchases/Month at 15.7× ROAS: Bicycle Store Scales with Google Shopping & PMax
A bicycle e-commerce retailer was running basic Google Search campaigns with zero Shopping presence — leaving the highest-intent purchase queries entirely untapped. We built Google Merchant Center from scratch, launched structured Shopping campaigns by product category, and layered in Performance Max with video and display assets. The result: 670 monthly purchases, a 15.7× ROAS, and a $0.86 cost per conversion on an ongoing and improving basis.
670
Monthly Purchases
×15.7
ROAS
$0.86
Cost Per Conversion
1.91%
CTR
The Situation
Search Traffic Only. Shopping Intent Untapped.
The store had a solid product catalog — road bikes, mountain bikes, urban commuters, accessories, and components — and existing Google Search campaigns that drove some traffic. But Search campaigns capture users who already know what they're searching for with text. Google Shopping captures users who are visually browsing products, comparing prices, and ready to purchase — and that audience was seeing the store's competitors, not the store itself.
There was no Google Merchant Center account, no product feed, and no Performance Max campaigns. The cost per conversion on Search alone was running above viable margins for mid-ticket items. The store was competing on branded queries and a handful of generic keywords while its entire product catalog sat invisible in Shopping results.
Before · Search Only
0
Shopping Campaigns
None
Google Merchant Center
High
Cost Per Conversion
Limited
Purchase Intent Reach
After · Full Shopping Stack
670
Monthly Purchases
×15.7
ROAS
$0.86
Cost Per Conversion
1.91%
CTR
Root Causes
Three Gaps That Kept the Store Invisible
The audit revealed that the underperformance wasn't a bidding problem — it was a structural absence. The store simply didn't exist in the channels where bicycle buyers actually shop:
No Google Merchant Center — the entire product catalog was invisible in Google Shopping, meaning competitor stores appeared at the top of results for every product-level query the store's customers were making
Search-only campaigns were limited to text-based queries — missing the massive pool of purchase-intent traffic that arrives through product image browsing, price comparison, and Shopping tab searches
No Performance Max meant no cross-channel reach — YouTube (where bicycle review content drives strong purchase intent), Display (retargeting site visitors), and Gmail were all completely untapped
Bidding was not segmented by product category or margin — high-ticket bikes and low-margin accessories competed for the same budget with no tROAS differentiation, diluting return on the most valuable categories
Conversion tracking was incomplete — not all purchase events were firing correctly, which meant the bidding algorithm had insufficient signal to optimize toward actual buyers rather than general traffic
Running Google Search campaigns for your e-commerce store with no Shopping presence?
A free audit shows you exactly what your competitors are capturing that you're missing.
What We Did
Five Workstreams. One Unified Shopping Engine.
We built the full Google Shopping stack from scratch — starting with the foundation (Merchant Center and feed quality) before touching campaigns. A clean feed is the prerequisite to everything else; campaign structure built on a poor feed just amplifies wasted spend.
1
Google Merchant Center Setup and Feed Optimization
Created and verified the Merchant Center account, then built a structured product feed from the store's catalog. Feed optimization focused on five areas: (1) Title rewriting — leads with brand + model + key spec + variant (e.g., "Trek Marlin 7 29″ Mountain Bike — Matte Black 2025"); (2) GTIN/EAN matching — matched all branded bicycles to manufacturer GTINs for Shopping eligibility; (3) Custom labels for campaign segmentation — tagged by category, price bracket (under $70 / $70–$186 / $186+), and margin tier; (4) High-quality images — sourced manufacturer clean-background images and added lifestyle images as additional_image_link; (5) Real-time inventory sync to prevent serving ads for out-of-stock products.
Category-Segmented Shopping Campaigns with tROAS Bidding
Structured Shopping campaigns by product category rather than running everything in a single campaign. Road bikes, mountain bikes, urban/city bikes, children's bikes, and accessories each received dedicated campaigns with individual tROAS targets calibrated to their margin profiles. High-ticket bikes ($186+) received higher tROAS targets (allowing higher CPC bids) because the revenue per conversion justified it. Accessories ran with conservative tROAS to protect margin on lower-value orders. This segmentation is what drove the $0.86 blended cost per conversion — without it, accessories would have dragged down spend efficiency on bikes and vice versa.
Performance Max Launch with Video, Display, and Search Assets
Launched PMax campaigns alongside Shopping, using the existing Search campaign conversion history as the algorithm's starting signal. Built asset groups for each major category with: 15–20 image assets (product + lifestyle), 5 video assets (manufacturer brand videos + custom short clips for YouTube), 15 headlines, 5 descriptions, and sitelinks. PMax audience signals were populated from GA4: all purchasers (past 180 days), product page viewers (90 days), and cart abandoners (30 days). This gave the algorithm a strong ICP seed to expand from rather than starting blind.
Performance MaxVideo assetsGA4 audience signals4 asset groups
4
Conversion Tracking Rebuild and GA4 Integration
Audited and rebuilt conversion tracking via Google Tag Manager — fixed the broken purchase event firing, added enhanced conversions (hashed email + phone for better match rates), and imported GA4 purchase events into Google Ads as the primary conversion action. Also tagged micro-conversions: add-to-cart, checkout initiation, and product detail page views — used as observation signals without bidding pressure. Without accurate conversion data, Smart Bidding optimizes toward ghost traffic. Fixing the tracking was a prerequisite to all bidding improvements.
Ongoing Optimization: Negatives, Bids, and Feed Iteration
Weekly Search Terms report review and negative keyword additions — common wasters for bicycle stores include "repair", "rental", "used", "second hand", "DIY", brand competitors, and non-purchase modifier terms. Monthly feed audits for new products and seasonal catalog updates. Quarterly tROAS target adjustments based on actual ROAS trends by category. Seasonal bid increases ahead of peak demand periods (spring cycling season, gift season). This ongoing management cadence is what sustains the 15.7× ROAS at growing volume rather than seeing returns erode as spend scales.
670 Monthly Purchases. $0.86 Per Conversion. Ongoing.
The results reflect an ongoing engagement — these numbers are not a snapshot from a single month but a sustained performance level achieved through continuous optimization. The 15.7× ROAS is maintained at 670 monthly purchases, meaning scale hasn't eroded efficiency, which is the usual pattern when Shopping campaigns are structured incorrectly. Correctly segmented campaigns allow spend to concentrate in the highest-return products automatically.
Metric
Before
After
Change
Monthly Purchases
Limited (Search only)
670
↑ Scaled
ROAS
Below margin threshold
×15.7
↑ ×15.7
Cost Per Conversion
Above viable margin
$0.86
↑ Efficient
CTR
Search-only baseline
1.91%
↑ Strong
Shopping Presence
Zero
Full catalog live
↑ Launched
Performance Max
Not running
Active (4 asset groups)
↑ Active
Conversion Tracking
Incomplete
Full purchase + micro
↑ Fixed
Feed Quality
No feed
Optimized + synced
↑ Built
Key Google Shopping Insight
The $0.86 cost per conversion is maintained at 670 monthly purchases — meaning volume did not erode efficiency. This is only possible with proper category segmentation. When bikes and accessories compete in the same campaign, the algorithm averages their performance. Separated, it can push budget aggressively to the categories where high-ticket items justify high CPCs — and pull back automatically on lower-margin SKUs. Campaign structure is what protects ROAS at scale.
Campaign Breakdown
Category Campaigns. Margin-Aware Bidding.
Each product category operates in its own campaign with a tROAS target calibrated to its actual margin profile. This prevents the most common Shopping mistake — running high-margin and low-margin products in competition for the same budget:
Road & Mountain Bikes ($186+)
×22.4
ROAS (high-ticket)
tROAS 1800%Priority High spend
↑ Core revenue driver
Urban & Kids Bikes ($70–$186)
×12.1
ROAS (mid-ticket)
tROAS 1100%Volume High
↑ Volume + margin balance
Performance Max (All Categories)
×14.3
Blended ROAS
4 asset groupsYouTube + Display
↑ Cross-channel reach
Is Your E-Commerce Store in the Same Position?
This case study is directly relevant if your situation matches any of the following:
You're running Google Search campaigns but haven't launched Google Shopping — your product catalog is invisible in Shopping results while competitors capture purchase-ready customers
You have a Google Merchant Center account but the feed is unoptimized — products with missing GTINs, generic titles, or poor images are suppressed or underperforming in Shopping auctions
All your products are in a single Shopping campaign — high-margin and low-margin products are competing for the same budget, and average ROAS is hiding which categories are profitable and which aren't
You haven't launched Performance Max — you're missing YouTube, Display, and Gmail reach for the product-aware audiences that convert after multiple brand touchpoints
Your cost per conversion is above your viable margin threshold and you're not sure whether the problem is bids, feed quality, product mix, or tracking — all four need to be checked together
Conversion tracking is incomplete or you're using only GA4 goals without proper Google Ads enhanced conversion setup — Smart Bidding is optimizing toward incomplete data
30 min · Free · We'll review your feed, campaigns, and tracking gaps
Frequently Asked
Questions About This Case Study
Launching Google Shopping for a bicycle retailer requires four steps: (1) Set up and verify Google Merchant Center, then submit a clean product feed with accurate titles, descriptions, GTINs, and high-quality images; (2) Structure Shopping campaigns by product category — road bikes, mountain bikes, accessories, and spare parts each convert at different margins and need separate management; (3) Set initial tROAS targets conservatively to let the algorithm build conversion data without underbidding; (4) Launch Performance Max once Shopping conversion history is established. Feed quality is the foundation — poor titles and missing attributes are the most common reason Shopping campaigns underperform at launch.
ROAS benchmarks vary significantly by category. High-ticket bicycles ($500–$3,000+) typically achieve 8×–18× ROAS because high purchase values relative to ad spend create favorable ratios. Accessories and consumables typically achieve 4×–10× ROAS. The 15.7× overall ROAS in this case reflects a healthy mix of high-ticket bikes and accessories in a properly segmented campaign structure. Importantly, this is measured ROAS — conversion tracking included returns and cancellations to avoid inflating the number with unconfirmed orders.
For retailers with an established conversion history, Performance Max outperforms standard Shopping by expanding reach across Search, Display, YouTube, and Gmail simultaneously. However, PMax needs at least 30–50 monthly conversions to optimize effectively. The recommended approach is to start with standard Shopping campaigns to build conversion data, then layer in Performance Max once monthly conversions exceed 30. In this case, PMax was launched alongside Shopping because the account had prior Search history — the algorithm wasn't starting completely blind. Both now run in parallel and serve complementary roles.
Feed optimization for bicycle retailers focuses on five areas: (1) Titles — lead with brand + model + key spec + color/size; (2) GTINs — bicycle manufacturers almost always have GTINs and missing them hurts Shopping eligibility; (3) Product type and custom labels — tag by category, price bracket, and margin tier so campaigns can target by profitability; (4) Images — use clean product shots and add lifestyle images as additional_image_link; (5) Availability sync — ensure real-time inventory sync to avoid serving ads for out-of-stock items, which wastes budget and damages Quality Score. Feed quality directly determines whether your products are competitive in Shopping auctions — it's not a one-time setup but a continuous maintenance task.
Reducing cost per conversion in Google Shopping involves three levers: (1) Negative keywords — add query-level negatives from Search Terms reports weekly; common wasters for bicycle stores include "used", "repair", "rental", "DIY", and competitor brand names; (2) Product-level bid management — lower bids on low-margin accessories while increasing bids on high-margin bikes where the ROAS math supports higher CPC; (3) Audience layering — apply GA4 audiences (previous purchasers, cart abandoners, product page viewers) as bid modifiers to push spend toward segments that convert at higher rates. In this case, the $0.86 cost per conversion was achieved primarily through aggressive negative keyword management combined with category-separated campaigns with margin-calibrated tROAS targets.
Ready to Launch Google Shopping for Your E-Commerce Store?
We'll audit your current Google Ads setup, Merchant Center feed quality, and conversion tracking — and show you exactly what's missing and what it's costing you. Free, 30 minutes, no obligation.