BigCommerce Google Ads: The ArmorGarage Rebuild, Step by Step

Quick Answer: ArmorGarage, a BigCommerce store selling garage flooring, was losing money to broad-match waste and conversion tracking optimizing for page views instead of purchases. We rebuilt tracking, expanded negatives across 14 ad groups, fixed GTINs and titles, then launched Performance Max with customer-list signals. The account reached 1,500%+ ROAS, peaking at 12.84x. Validate tracking, first-party evidence, the landing-page promise and the business outcome before acting. Use the comparison table to choose the appropriate route, then document assumptions, decision thresholds and review dates so future updates remain accurate, attributable and easy to audit.

Agencies love to present Performance Max wins as if the campaign type did the work. It did not. In every strong PMax account I have built, the result was decided before the campaign launched, in the tracking, the feed and the account structure underneath it. PMax just amplified what was already fixed.

ArmorGarage is my clearest example. A BigCommerce store selling garage flooring, running Google Ads with results nobody was proud of. After a rebuild, the account reached 1,500%+ ROAS on Performance Max, peaking at 12.84x. Nothing about the process was secret. It was a sequence, and the order of the steps mattered as much as the steps themselves.

This post walks through that rebuild step by step, so you can run the same sequence on your own BigCommerce store. At the end, I will cover the bonus round: taking the same platform to Microsoft Ads, where a BigCommerce epoxy brand hit 1,430% ROAS.

Table of Contents

1. The Setup: What Changed and Why
2. What I Inherited: Broad-Match Waste and Tracking That Lied
3. The 4-Day Audit: Finding the Real Problems
4. Step 1: Fix the Conversion Tracking
5. Step 2: Restructure and Expand Negatives Across 14 High-Spend Ad Groups
6. Step 3: Fix the Feed, GTINs and Titles
7. Step 4: Launch Performance Max With Customer-List Signals
8. Bonus: 1,430% ROAS on Microsoft and Bing Ads, Same Platform
9. Meta Advantage+ Shopping for BigCommerce Stores
10. Ecommerce Campaign Kit for BigCommerce
11. Common Mistakes We See on BigCommerce Accounts
12. What This Rebuild Means for Your BigCommerce Store
13. Practical Growth Playbook for Bigcommerce Google Ads
14. Practical Comparison for Bigcommerce Google Ads
15. AI Buyer Questions
16. FAQs
17. Further Reading
18. Related Hustle Marketers Research and Proof
19. About Ishant Sharma

The Setup: What Changed and Why

Metric	Before	After
Conversion tracking	Optimizing toward page views instead of purchases	Rebuilt so purchases with real order values are the only primary signal
Account structure	Broad-match keywords bleeding spend across 14 high-spend ad groups with no negative discipline	Restructured with a full negative keyword expansion across those same 14 ad groups
Feed quality	Missing GTINs, titles not matching buyer search language	GTIN coverage implemented, titles rebuilt to lead with buyer-first terms
Campaign type	Standard Shopping running on corrupted data	Performance Max launched with customer-list audience signals
ROAS	Varies by account and must be verified	1,500%+ ROAS, peaking at 12.84x

What I Inherited: Broad-Match Waste and Tracking That Lied

The inherited ArmorGarage account had two core problems: broad-match keywords bleeding spend on irrelevant queries, and conversion tracking that was optimizing for page views instead of purchases. Every automated bid decision Google made was built on false data. Until that was fixed, no campaign change could work.

Read that tracking problem again, because it is more common than most advertisers believe. The conversion action Google’s algorithms were optimizing toward counted page views. So Smart Bidding dutifully found users who viewed pages. Cheaply. In volume. The dashboard looked busy while actual revenue efficiency went nowhere.

Layered on top of that, broad match terms were pulling queries that had nothing to do with buying garage flooring. Broad match is not evil, but broad match plus corrupted conversion signals is a money incinerator: the algorithm expands into junk queries and then gets told the junk converted, because a page view fired.

If your account has flat results despite constant “optimization”, check what your primary conversion action actually measures before you touch anything else. In my experience auditing hundreds of accounts, wrong-signal tracking is the single most common root cause of stuck performance.

The 4-Day Audit: Finding the Real Problems

I ran a 4-day audit before changing anything. The audit covered conversion tracking end to end, search term reports across every ad group, feed and Merchant Center health, and account structure. Diagnosis before treatment. Most turnarounds fail because the fixing starts before the finding is finished.

What the audit process looked like:

• Tracking verification. Fire test conversions, inspect what the tag actually sends, confirm which conversion actions are set as primary, and check whether values pass correctly. This is where the page-view optimization problem surfaced.
• Search term autopsy. Pull search terms for every high-spend ad group and classify them: converting, relevant but not converting, and irrelevant. The irrelevant column told me exactly where broad match was leaking.
• Feed review. Product titles, GTIN coverage, disapprovals, and how the feed data mapped to real buyer queries for garage flooring.
• Structure review. How ad groups and campaigns were organized, and whether budget could flow to what worked.

Four days is short. It is enough when you know what kills accounts. The output was a written fix sequence, ordered by dependency. Tracking first, always, because every later step relies on honest data.

Step 1: Fix the Conversion Tracking

We rebuilt conversion tracking so purchases, with accurate values, became the only primary conversion signal. Everything else moved to secondary. This is step one because Smart Bidding is a mirror of your conversion data. Show it page views and it buys page views. Show it revenue and it buys revenue.

On BigCommerce this means verifying the purchase event fires on the actual order confirmation, passes the real order value, and deduplicates properly. We removed the page-view conversion from the primary slot so it could no longer steer bidding.

Then comes the uncomfortable part: waiting. After a tracking rebuild, the algorithms need conversion history built on the new, honest signal. Expect a learning period where results wobble before they climb. Clients who understand this upfront stay calm. Clients who do not will panic and undo good work.

This is the same tracking-first principle behind a rebuilt tracking setup that delivered 280% more leads for a driving school client, a different vertical, same root cause: fix what the algorithm is actually being told before touching anything else.

Step 2: Restructure and Expand Negatives Across 14 High-Spend Ad Groups

We restructured the account and ran a negative keyword expansion across 14 high-spend ad groups. The search term autopsy from the audit became a negative list that stopped the broad-match bleeding, while the restructure grouped products so budget could concentrate on proven winners.

The negative expansion was not a one-time list paste. For each of the 14 ad groups we:

1. Classified every meaningful search term from the audit period.
2. Built shared negative lists for globally irrelevant themes, applied account-wide.
3. Added ad-group-level negatives for cross-contamination, so queries reached the ad group with the most relevant ad and landing page.
4. Tightened match strategy where broad match had no data to justify itself.

Restructuring followed the money. High-intent product themes got their own ad groups with dedicated budgets instead of sharing space with low-intent traffic. None of this is exotic. It is disciplined query hygiene, done thoroughly instead of occasionally.

Step 3: Fix the Feed, GTINs and Titles

We implemented GTINs across the catalog and restructured product titles to lead with the terms flooring buyers actually search. On BigCommerce, feed quality decides how Shopping and PMax match your products to queries. GTINs improve matching and unlock competitive data, and buyer-first titles put you in the right auctions.

GTIN implementation sounds trivial. At catalog scale it rarely is: sourcing manufacturer barcodes, filling the fields, and handling genuinely exempt products correctly. It is worth the grind. Products with valid GTINs match better and gain eligibility Google reserves for well-identified items.

Title restructuring followed the same logic I use on every platform: front-load what buyers type. For garage flooring, that means the product type, material and key attribute up front, not marketing names. This mirrors the rebuild I later ran on P-REX Hobby’s Shopify catalog, where model-first titles across 400+ SKUs drove a 9x ROAS. Different platform, same principle.

BigCommerce merchants can manage this through the native Google integration or a feed tool. The tool matters less than the rule: your feed title is a search-matching asset, not a website label.

Step 4: Launch Performance Max With Customer-List Signals

Only after tracking, structure and feed were fixed did we launch Performance Max, seeded with customer-list audience signals. PMax amplifies whatever foundation exists. On the rebuilt foundation, the account reached 1,500%+ ROAS, with a peak of 12.84x. On the old foundation, the same campaign would have failed.

Customer lists were the launch accelerant. Uploading the buyer list as an audience signal tells PMax who real customers look like, so it starts from a sketch of your actual buyer instead of a cold guess. For a considered purchase like garage flooring, where buyers research before ordering, that head start matters.

The launch discipline that made it work:

• Honest conversion data from step 1, so the algorithm optimized for purchases and real values.
• Clean feed from step 3, so PMax matched products to high-intent queries.
• Customer-list signals, so learning started warm.
• Patience through learning, with judgment reserved for stable post-learning windows.

Getting the asset group structure right matters as much as the signals you feed it. See how to structure PMax asset groups for the framework I use across accounts, BigCommerce or otherwise.

The sequence is the strategy. Tracking, structure, feed, then PMax. Run it out of order and each step undermines the next.

Bonus: 1,430% ROAS on Microsoft and Bing Ads, Same Platform

BigCommerce stores can extend the same rebuilt foundation to Microsoft Ads, where competition is thinner and CPCs run lower. For an epoxy ecommerce brand on BigCommerce, that produced a 1,430% ROAS on Microsoft and Bing Ads. The import takes minutes, and the audience skews toward exactly the buyers home-improvement brands want.

The playbook:

1. Set up Microsoft Merchant Center and sync your product feed. All the feed work you did for Google, titles and GTINs included, transfers directly.
2. Use Microsoft’s Google Ads import to copy campaigns, negatives and structure.
3. Adjust bids downward to reflect Bing’s cheaper auctions, then let its data accumulate separately. Do not mirror every future Google change blindly.

Bing’s demographic skews older and higher-income, which for garage, flooring and epoxy products is the homeowner doing a considered project. The volume is smaller than Google. The efficiency, in my accounts, is frequently better. For ten minutes of import work plus light ongoing management, it is one of the easiest profitable expansions in ecommerce PPC.

Meta Advantage+ Shopping for BigCommerce Stores

Google and Bing were never the whole story for ArmorGarage. The same rebuilt data discipline carries over to Meta, and BigCommerce merchants who skip this step are leaving a channel on the table that reaches buyers earlier in the research phase, before they are typing branded terms into Google.

Catalog sync starts with BigCommerce’s native Facebook and Instagram channel, built on Meta’s Commerce Manager integration, or a dedicated feed app if you need more control over attribute mapping. Either path pulls your BigCommerce product catalog, price, availability, images and variants included, into Meta’s catalog on a scheduled sync, typically every 24 hours through the native channel, faster through a feed app with its own refresh cadence. The mapping question that trips up most BigCommerce merchants is category structure.

BigCommerce’s category tree does not automatically become Meta’s product category taxonomy, so someone has to map categories manually or through the feed app’s rules engine. Custom fields matter here too. If GTIN, MPN or brand live in BigCommerce custom fields rather than the standard catalog fields, confirm the sync tool actually reads them, because a silent mapping gap means Meta falls back to weaker identifiers and your catalog under-matches on search.

Advantage+ Shopping setup considerations differ from a standard catalog sales campaign. Meta’s algorithm wants creative variety. Plan for several image and video variants per product group rather than a single static image, since Advantage+ tests combinations automatically and starves creative sets that only give it one option. Conversions API matters as much here as tracking did on the Google side. BigCommerce’s webhook and API integration can pass server-side purchase events to Meta, recovering conversions the pixel alone misses to iOS privacy restrictions and ad blockers. Margin exclusions belong in the setup too. Exclude clearance and loss-leader SKUs from the Advantage+ product set so the algorithm is not optimizing spend toward your lowest-margin items.

Creative for a considered-purchase category like industrial epoxy is not the same as creative for apparel or beauty. Buyers researching garage flooring or epoxy coatings want proof: before-and-after coverage shots, application video, coverage-per-gallon specifics, not lifestyle imagery alone. That approach pairs directly with the buyer-first feed language covered in Step 3 above. The same discipline, front-load what buyers actually search for and care about, applies to ad copy and creative testing on Meta, not just Shopping titles.

None of this replaces the Google and Bing work covered earlier in this post. It is the same foundation, honest tracking, a clean catalog and disciplined structure, applied to a channel that reaches buyers while they are still comparing options rather than actively searching with intent.

Ecommerce Campaign Kit for BigCommerce

Use these templates as the starting point for your BigCommerce feed, campaign build and ad copy, then adjust them to your catalog.

Feed title formulas by category

Product Category	Title Formula	Example
Apparel	Brand + Gender + Product Type + Color + Size	Ridgeline Men’s Trail Jacket Navy L
Parts and components	Brand + Part Type + Compatibility Model + Part Number	ProDrive Roller Bracket for LiftMaster 8500 PD-2210
Beauty	Brand + Product Type + Key Attribute + Size	GlowLab Vitamin C Serum Fragrance Free 30ml
Home goods	Brand + Product + Material + Dimension	ShieldCoat Garage Floor Epoxy Kit Gray 2 Car
Consumables	Brand + Product + Count/Size + Flavor/Variant	PurePet Salmon Dog Treats 500g Grain Free

Manage these through the native Google integration or a feed tool, and keep the rule in mind: the feed title is a search-matching asset, not a website label.

Campaign structure

Campaign	Type	Purpose	tROAS approach
Branded Search	Search	Protect brand terms and capture buyers who already know you	High tROAS, this traffic should stay cheap
Shopping by category	Standard Shopping	Query visibility, negatives and feed insight per category	Moderate tROAS set from each category’s margin
PMax per category group	Performance Max	Scale proven categories with customer-list signals	Start at the ROAS Shopping proved, raise in small steps
Dynamic Remarketing	Display	Bring back product viewers and cart abandoners	High tROAS with a small, capped budget
Bing Shopping import	Microsoft Shopping	Cheaper clicks, strong fit for home improvement buyers	Mirror Google targets, then adjust down for Bing CPCs

The Bing row matters more on BigCommerce than most platforms: the import reuses everything you built for Google, and the audience fits considered home projects.

Merchant Center pre-flight checklist

Feed quality is also what stands between you and a suspension. If disapprovals or policy flags have already hit your account, see how to recover from a Google Merchant Center suspension before working through the list below.

• GTIN coverage across the catalog, with identifier_exists set correctly for exempt items
• Titles front-loaded with buyer search terms in the first 70 characters
• Images clean: no watermarks, no promo text, main image shows the product clearly
• Price and availability in the feed matching the live product page exactly
• Shipping settings configured in Merchant Center and matching checkout rates
• Tax settings configured for every state you target
• Returns policy and contact pages published and linked in the footer
• Claims wording scanned on coatings, supplements and beauty products, no exaggerated or cure claims
• Supplemental feed or feed-tool rules in place for title rewrites
• Merchant Center diagnostics reviewed weekly

Ecommerce negative keyword starter list

free, cheap, used, second hand, refurbished, diy, how to make, how to build, how to install, repair, manual, instructions, pdf, review, reviews, complaints, wholesale, alibaba, aliexpress, amazon, ebay, walmart, jobs, careers, salary, template, tutorial, course, warranty claim, recall

For a DTC BigCommerce store, keep “wholesale” as a negative unless you sell trade. Watch “how to install” carefully: it filters research traffic, but remove it if installation queries convert for your products.

10 responsive search ad headlines for branded and category search

1. Shop [Brand] Official Store
2. Top Rated [Category] Online
3. [Offer] On Your First Order
4. Fast Free Shipping
5. New [Category] In Stock
6. [Brand] Sale Ends Soon
7. Shop [Category] Deals Now
8. Easy Returns, Fast Help
9. Save [Offer] Today
10. Official [Brand] Site

4 descriptions

1. Shop [Brand] [Category] with fast shipping, easy returns and secure checkout.
2. Order today and get [Offer]. Fast dispatch and tracked delivery on every order.
3. Browse the full [Category] range. Clear pricing, real stock and quick support.
4. Find the right [Category] fast. Simple filters, clear specs and checkout in minutes.

Common Mistakes We See on BigCommerce Accounts

Mistake	Why It Happens	The Fix
Category feed mismatched to Google or Meta taxonomy	BigCommerce’s category tree is built for site navigation, not for Google’s product category system or Meta’s catalog taxonomy, so merchants leave the default mapping in place	Map each BigCommerce category to the correct taxonomy node manually, or through a feed app’s rules engine, and re-check the mapping whenever categories change
GTIN or MPN stuffed into the wrong custom field	BigCommerce lets merchants create arbitrary custom fields, and product teams often put identifiers there instead of the dedicated GTIN, MPN or UPC fields the feed export expects	Move identifiers into BigCommerce’s native identifier fields, and confirm the feed export, native or app, actually reads them before assuming coverage is complete
Multi-storefront or multi-channel sync drift	Merchants running multiple storefronts or sales channels off one BigCommerce catalog assume price, stock and status stay in sync automatically	Audit each channel’s feed against the source catalog on a schedule, and document channel-specific overrides like price rules or exclusions somewhere other than someone’s memory
App stack bloat slowing page speed	Every marketing tool wants its own app installed, and BigCommerce’s app ecosystem makes it easy to add several tracking and personalization apps without ever removing one	Audit installed apps quarterly, remove anything not actively used, and check page speed impact before adding a new script-heavy app
Checkout and tax config mismatched with the feed	Tax settings get configured once at launch and never revisited, while the feed’s price field assumes a tax treatment that no longer matches checkout	Reconcile Merchant Center tax settings against actual checkout tax collection per state, especially after nexus changes or new product categories
Conversion value passed as order subtotal instead of true order value	Default BigCommerce tracking setups sometimes pass subtotal before shipping, tax or discounts are applied, so Smart Bidding optimizes toward a number that does not match real revenue	Verify the conversion tag fires on the confirmed order total, not subtotal, and reconcile a sample of tracked values against actual order records monthly

What This Rebuild Means for Your BigCommerce Store

The ArmorGarage numbers came from sequence and discipline, not a secret tactic. Verify what your conversions measure, stop the query bleeding, make your feed speak buyer language, then let automation scale a foundation worth scaling. Any BigCommerce store can run this sequence. Most just never do it in order.

If your account has been “optimized” monthly for a year and the trend line is flat, the odds are high that one of the foundations, tracking, query hygiene or feed, is quietly broken. Those problems do not show up in a dashboard screenshot. They show up in a proper audit.

I offer that audit free, the same $500-value, 4-day style review that started the ArmorGarage rebuild. You get the findings and the fix sequence whether or not we work together afterward. Request it at ishantsharmamarketer.com.

Practical Growth Playbook for Bigcommerce Google Ads

Use approved first-party evidence where it exists, then adapt the sequence to the business baseline, market, offer, budget and operational capacity. The aim is to turn the article into a sequence of measurable decisions instead of a list of disconnected tactics.

90-Day Execution Roadmap

Phase	Priority actions	Required output	Decision gate
Days 1 to 14	Audit unit economics, product feed accuracy, Merchant Center eligibility, purchase tracking, new-customer identification and returns.	A product-level baseline for margin, availability, conversion value and feed health.	Do not scale automation while price, availability, value or product identifiers are unreliable.
Days 15 to 30	Segment products by demand, margin, inventory and business priority, then align Search, Shopping or PMax ownership.	Campaign and custom-label structure tied to product economics.	Fund product groups only when the account can explain what they contribute and why.
Days 31 to 60	Test feed titles, creative, offers, landing pages and new-customer bidding with one documented variable at a time.	A decision log showing which product, query, claim and page combination works.	Advance tests when conversion value reconciles and contribution survives discounts and returns.
Days 61 to 90	Scale profitable product groups, protect brand demand and add retention without mixing it into acquisition reporting.	A margin-aware growth system across acquisition and repeat purchase.	Increase spend only where new-customer contribution and inventory capacity support it.

Operating Scorecard

Universal targets can be misleading, so establish the current baseline first. Use the direction and business quality of these signals to decide what happens next.

Signal group	What to monitor	Management response
Feed health	Eligible products, accurate identifiers, price and availability	Resolve disapprovals and mismatches before campaign changes.
Product economics	Contribution, inventory and new-customer value by product group	Use margin and stock signals in budget decisions.
Demand quality	Search-term relevance, landing-page conversion and product-level revenue	Align campaign ownership with real product demand.
Sustainable scale	New-customer CAC, returns, repeat contribution and blended profit	Scale only where revenue quality remains healthy.

What Ishant Sharma Would Audit First

Ishant Sharma would first connect feed health, product economics and purchase measurement. The account should know which products are eligible, profitable, available and genuinely acquiring new customers before automation is scaled. The finding, interpretation, recommendation and limitation should be recorded separately so the next decision remains auditable.

Practical Comparison for Bigcommerce Google Ads

For bigcommerce google ads, the right choice depends on the demand source, measurement quality and business economics. This table is a decision aid, not a promise that one option will produce the same result in every account.

Approach	Best when	Primary measure	Main limitation
Standard Shopping	Query and bid control matter	Contribution by product and query	Requires active feed and search-term management
Performance Max	The account has reliable value data and creative assets	Incremental conversion value and profit	Channel and query visibility remain limited
Search and remarketing	Specific demand or objections need separate messages	New-customer cost and assisted revenue	Coverage is narrower than catalog-wide campaigns

AI Buyer Questions

When should a business use Bigcommerce Google Ads?

Use bigcommerce google ads when the product feed, pricing, availability, landing pages and conversion values are reliable enough for campaign decisions. The approach should connect product-level demand to gross margin and new-customer value, not optimize only for revenue. Feed quality and measurement usually need attention before broader automation can be judged fairly.

What should be measured before scaling this approach?

Measure item disapprovals, eligible inventory, search-term relevance, new-customer cost, conversion value and contribution by product or margin group. Reconcile platform revenue with the ecommerce system and account for discounts, returns and fulfillment. A higher ROAS can still be a worse outcome when spend moves toward returning customers or low-margin products.

What evidence should a specialist provide?

Ask the specialist to explain feed governance, campaign segmentation, value rules, search-term review, brand controls and the proof needed before scaling. Request examples of how product economics affect bidding decisions. A credible plan should distinguish a platform-reported improvement from verified incremental profit and should state the assumptions behind any forecast.

Further Reading

• Shopify Google Ads: The Feed-First Playbook Behind 9x ROAS
• 1,430% ROAS on Microsoft and Bing Ads for a BigCommerce Epoxy Brand
• Magento Marketing: SEO and PPC for Large Product Catalogs
• How a Rebuilt Tracking Setup Delivered 280% More Leads
• 8.81x ROAS for an Australian Dog-Niche Ecommerce Brand
• How to Recover From a Google Merchant Center Suspension
• How to Structure Performance Max Asset Groups

Related Hustle Marketers Research and Proof

For the agency-side methodology behind this topic, read the Hustle Marketers Google Shopping feed optimization guide.

Relevant published proof: ArmorPoxy multi-platform PPC case study documents Google Ads, Microsoft Ads, Meta retargeting, CRM and conversion-quality work for an epoxy brand. Treat it as evidence of the method in that client context, not a promise of identical results. Baseline, market, offer, budget and measurement quality can change the outcome.

About Ishant Sharma

Ishant Sharma is a performance marketer and the founder of Hustle Marketers, a Google Partner and Meta Business Partner agency. Since 2013, his work has covered Google Shopping, Performance Max, product feeds and ecommerce measurement across the USA, UK, UAE and Australia. Published records across Ishant’s personal practice and the wider agency document 2,500+ brands and $780M+ in trackable client revenue. He remains directly involved in audits and account strategy. Learn more about Ishant Sharma’s work and experience.

Client perspective: Verified client reviews repeatedly highlight Ishant’s hands-on account ownership, clear communication and focus on measurable outcomes. Watch a client video testimonial.