You doubled your Google Ads budget last quarter. Your cost-per-sale did not budge. You are not bad at advertising.
You are bidding on keywords that attract browsers, job seekers, and bargain hunters instead of buyers.
Google’s Keyword Planner suggested "women’s dresses" because it has 100,000 monthly searches. You added it as broad match.
Now you pay for clicks from "women’s dresses jobs," "women’s dresses wholesale pricing," and "women’s dresses return policy exchange." Those searchers never intended to buy anything.
Most PPC keyword guides tell you which tools to open first. They skip the step that actually recovers wasted money: auditing what your campaigns are already paying for.
How do I find high-converting keywords for my ecommerce PPC campaigns?
Open your Google Ads search query report for the last 30 days. Filter for any term with over 15 clicks and zero conversions. Add every single one as a negative keyword at the campaign level.
Then sort remaining terms by cost-per-conversion. Pause anything with a CPA over twice your target. This one-hour audit recovers 20–35% of wasted spend before you research a single new keyword.
Most ecommerce operators do keyword research backward. They start in Keyword Planner. They type in their product name.
They copy the suggestions with the highest search volume. They check their dashboard daily. They see clicks going up.
They assume that means progress.
What this actually costs: You bid "running shoes" as broad match. You pay for "running shoes donation." You pay for "running shoes career" and "how to clean running shoes."
A running gear store spent $847 in one month on non-purchase searches. Their search query report showed 43 irrelevant terms eating budget. They had never opened that report once.
They saw 340 clicks in a month. They thought the campaign was working.
The move that actually works: Build your keyword list from your own conversion data first. Start with exact-match terms that already produced sales. Use those as seeds for phrase-match variations.
Only then consult external tools for expansion ideas.
A Shopify pet supply store doing $28,000/month did exactly this. They exported their search query report. They found 22 terms with 10+ clicks and zero sales.
They added them as negatives. They redirected budget to their three top-converting exact-match keywords. Conversion rate rose from 2.1% to 3.8% in five weeks.
Cost-per-sale dropped from $28 to $19. They did not increase their budget by a dollar.
What’s the difference between broad match, phrase match, and exact match keywords for ecommerce ads?
Match types control how closely a user’s search must match your keyword for your ad to appear. Exact match triggers ads only for searches that share the same meaning as your keyword. Phrase match includes searches that contain your keyword in the correct order.
Broad match shows your ad for any search Google considers related. For ecommerce, choosing the wrong match type is the single biggest source of wasted ad spend.
Google has made broad match more aggressive every year. A broad match keyword like "leather wallet" triggers ads for "fabric wallet." It also triggers for "wallet repair near me" and "wallet making classes."
Google calls this "close variant matching." Ecommerce operators call it a budget fire.
Here is how each match type behaves with a real product:
Exact match on [vegan leather crossbody bag]: Your ad shows when someone searches "vegan leather crossbody bag" or close variants like "vegan leather crossbody purse." You pay for high-intent searches. CPC runs higher per click.
Conversion rate runs higher too.
Phrase match on "vegan leather crossbody": Your ad shows for "vegan leather crossbody bag" and "best vegan leather crossbody for travel." It also shows for "vegan leather crossbody pattern" and "vegan leather crossbody wholesale." This is the middle ground.
Use it when you need more reach but still want guardrails.
Broad match on vegan leather crossbody: Google decides what relates. Your ad shows for "fake leather purse," "leather-free bag," "crossbody satchel," and occasionally "leather repair kit." Highest reach.
Highest waste.
Negative keywords block specific terms. Add "pattern," "wholesale," "repair," "jobs," "career," "internship," "rental," "used," "second hand," "free," "DIY," "how to," and "what is" at the campaign level. This stops non-buyer searches from ever triggering your ads.
A WooCommerce electronics store switched their top-spending keywords from broad match to phrase match. They added 67 negative keywords found in their search query report. Impressions dropped 40%.
Orders stayed exactly the same. They cut $1,200 in monthly ad waste overnight.
Which free or affordable tools can I use for ecommerce PPC keyword research?
You already own the best keyword tool. It is your Google Ads search query report. The terms people typed before buying from you are more valuable than any tool’s volume estimates.
Mine your own conversion data before opening third-party software.
Most small ecommerce operators need exactly three tools. Two of them are free.
Google Ads Search Query Report (free, built in): This report is your primary keyword research tool. Sort by conversions descending. The top 10–15 search terms are your highest-intent keywords.
Build campaigns around these first. No third-party tool can replicate this data. It comes from your actual customers, not industry averages.
Google Keyword Planner (free, inside Google Ads): Use it to estimate volume on new keywords you want to test. Do not use it as your primary keyword source. Ignore its "competition" metric entirely.
That metric measures how many advertisers bid on a term, not how hard it is to win. High competition often signals a high-converting commercial keyword worth bidding on.
Keywords Everywhere (free browser extension): Shows search volume, CPC, and competition data directly in Google search results. The cheapest paid tier costs $1.25 per month for 100,000 keyword lookups. Install it.
Search your product category in Google. Scroll the "People also ask" and "Related searches" sections. The extension shows volume for each suggestion.
This replaces hours of manual Keyword Planner searches.
For stores spending over $50,000/month on ads, add one paid tool.
Ahrefs or SEMrush ($130–$140/month): Enter a competitor’s domain. The tool reveals every keyword driving their organic traffic plus estimated PPC keywords. A mid-market home goods store discovered their top competitor was bidding on 14 comparison terms like "Brand A vs Brand B."
They had never considered those keywords. Those comparison terms produced a conversion rate 3x higher than their product-name keywords.
Run the search query audit from the first section before subscribing to any paid tool. Reclaim your wasted spend first. Then use the recovered budget to test keywords from these tools.
How do I organize keywords into effective ad groups for my product categories?
Group keywords by purchase intent and search specificity, not by product name. An ad group called "leather bags" with 40 mixed-intent keywords performs worse than three separate groups. Those groups: "buy leather bag," "leather bag reviews," and "vegan leather tote under $100."
Each group gets its own ad copy and landing page.
Most small operators create one ad group per category. They dump every keyword into it. Their ads say "Shop Our Leather Bags."
That message shows whether someone searched "buy leather messenger bag" or "how to condition leather bag." The second searcher never clicks. The first searcher clicks.
They land on a grid of 80 products. They leave.
Here is a structure that works for stores with 3–50 product categories:
Tier 1: Brand-defending campaigns. One ad group per branded term. Use exact and phrase match only.
These protect your brand from competitors bidding on your store name. Your Quality Score stays high because your brand terms are highly relevant. CPC stays low for the same reason.
Tier 2: Product-specific buying-intent campaigns. One ad group per product or tight product family. Keywords include "buy [product]," "[product] for sale," "discount [product]," "[product] online."
Use phrase match. Ads include price, shipping speed, and one differentiator. Landing page is the exact product page.
These produce your highest ROAS.
Tier 3: Category-intent campaigns. One ad group per category with modifier keywords. A furniture store example: "modern sofa under $1,000," "sectional with storage," "apartment sofa small space."
These searchers know their need but not the exact product. Ads showcase selection. Landing page is a curated category view, not the full catalog.
Tier 4: Problem-aware campaigns. One ad group per pain point. Keywords: "best shoes for plantar fasciitis," "backpack for heavy laptop commute," "gift for new mom who has everything."
Ads name the problem and offer the product. Landing page is a guide with products embedded. Conversion rates run lower here.
But these campaigns introduce new customers who buy later.
A multi-category home goods store doing $80,000/month reorganized from 12 messy ad groups into this four-tier structure. They wrote fresh ad copy for each tier. Their average Quality Score across all campaigns rose from 5.2 to 7.4 in eight weeks.
Average CPC dropped 18%. Conversion rate stayed flat. Same conversions, significantly lower cost.
What are the best strategies for finding long-tail keywords for ecommerce products?
Long-tail keywords come from two sources most operators never check: their own search query report and competitor product reviews. The search query report reveals the exact multi-word phrases buyers type before purchasing. Competitor reviews reveal the language customers use.
They describe what they want. They describe what they hate about existing products.
Long-tail keywords are phrases of four or more words. Examples: "lightweight waterproof hiking boots women size 8" or "organic cotton baby pajamas no flame retardant." Search volume runs low for each phrase individually.
Combined, they often drive more total conversions than short-tail keywords. The searcher knows exactly what they want. CPC runs lower.
Competition runs lower. Conversion rate runs higher.
Three methods for finding long-tail keywords without paying for a tool:
Method 1: Mine your search query report for 4+ word phrases. Export the last 90 days. Filter for queries with four or more words and at least one conversion.
Sort by conversions descending. These are proven long-tail keywords. Add them as exact match keywords in a dedicated campaign with higher bids.
Method 2: Read Amazon reviews for your top competitor’s product. Sort by "Most recent." A review says "finally a yoga mat that doesn’t slip during hot yoga."
That gives you "yoga mat that doesn’t slip hot yoga." A review says "perfect diaper bag for twins with lots of pockets." That gives you "diaper bag for twins lots of pockets."
These are the words actual buyers use. Keyword Planner would never suggest them.
Method 3: Use Google autocomplete as a discovery engine. Type a product category plus a modifier. "Running shoes for" — Google suggests "running shoes for flat feet," "running shoes for wide feet," "running shoes for knee pain."
Install Keywords Everywhere to see volume for each. Add suggestions with volume above 100 per month as phrase match keywords.
A children’s clothing store used Method 2 for their bamboo pajama line. They read 200 Amazon reviews for the top three competing brands. They extracted 27 long-tail phrases parents used to describe what they wanted.
Examples: "two-way zipper bamboo pajamas toddler" and "tagless bamboo sleeper sensitive skin." They built a campaign around these exact phrases. Conversion rate hit 4.7% versus 1.9% on their broad-match category campaign.
Average CPC dropped to $0.52 versus $1.84.
What results should I expect after fixing my keyword strategy?
Expect to recover 20–35% of wasted ad spend in month one from the search query audit and negative keyword additions. Expect conversion rates to rise 30–50% when you switch high-spend keywords from broad to phrase or exact match. Full ad group reorganization takes four to eight weeks.
Quality Score improves by one to three points.
Google’s algorithm needs time to relearn after major keyword changes. Here is a realistic timeline:
Week 1: Export search query report. Add negative keywords for zero-conversion queries. Pause keywords with CPA over 2x target.
Wasted spend drops within 48–72 hours.
Weeks 2–3: Switch top-spending broad match keywords to phrase or exact. Create new ad groups for your highest-converting exact-match terms. CPC rises slightly.
Conversion rate rises more. Net ROAS improves.
Weeks 4–6: Build Tier 3 category-intent and Tier 4 problem-aware campaigns. Launch long-tail keyword campaigns from the Amazon review method. Give these new campaigns two to three weeks to accumulate data before judging them.
Weeks 7–8: Full Quality Score improvement becomes visible. Campaign structure stabilizes. You now own a repeatable keyword workflow.
Track cost per conversion, not total conversions. Twenty conversions at $45 each costs more than twelve at $18 each. Total conversions trick operators into keeping expensive keywords alive.
One store selling premium blenders tracked this for 60 days. Before the audit, they spent $6,200 per month on ads. They got 82 conversions at $75.60 CPA.
After the full reorganization, they spent $5,800 per month. They got 119 conversions at $48.74 CPA. Same product.
Same landing pages. Same prices. The only variable was which keywords they paid to appear for.
The uncomfortable truth about ecommerce PPC: Google’s default settings are designed to spend your budget, not to maximize your profit. Broad match is the default match type. Automated bidding targets maximum clicks, not maximum conversions.
The search query report that shows you what actually happened sits buried three menus deep.
Your job is not to outsmart Google’s algorithm. Your job is to give it cleaner inputs. Better keywords.
Tighter match types. A growing negative keyword list. Ad groups organized around intent, not around product names.
This week, do exactly one thing. Export your search query report for the last 30 days. Filter for any term with over 15 clicks and zero conversions.
Add every one as a negative keyword. It takes one hour. It produces the highest return of any hour you spend on advertising this month.
