Google said the account made 816,270 conversions between 7 August and 9 October 2026. The store had 303 orders. Nobody lied. A page view was set up to count as a purchase, and Google's bidding chased page views.
So an audit starts with tracking. Every other number reads from it. Below are the 40 checks from one real audit of an Indian D2C (direct-to-consumer) store, in the order we ran them. The free template is at the end.
How do you audit a Google Ads account?
Audit it in seven stages, in this order: tracking, account structure, campaigns, assets, products, landing page and store, then measurement. A wrong conversion count poisons every check after it. Most teams start with keywords. Start with the scoreboard.
Our audit logged 145 findings in these stages, with waste at three levels: campaign, asset and product. For market-wide numbers, see Google Ads statistics 2026.
| Stage | Checks | What our audit found |
|---|---|---|
| Foundation and tracking | 1–8 | 816,270 "conversions" vs 303 orders |
| Account structure | 9–13 | Campaigns bid for 6 goals; 1 was a sale |
| Campaign | 14–22 | Targets raised to 7.3–11.5; Google sales fell 40% |
| Asset | 23–27 | 64 of 79 images made no sale |
| Product | 28–33 | 792 of 1,027 products sold nothing through ads |
| Landing page and store | 34–36 | A 2-day, 20% sale lifted daily orders 2.6× |
| Measurement and reporting | 37–40 | Ad spend +96%, online revenue +31% |
How do you check Google Ads conversion tracking? (Checks 1–8)
Check that Google counts only real sales, at their real value, with no gaps. Get this wrong and Google optimises for the wrong thing, perfectly.
- Only real sales are Primary goals. Where: Goals › Conversions › Summary › Edit goal. Red flag: 1 or more page-view or engagement actions set to Primary. Google says Primary actions are "used for bidding". Fix: Set them to "Secondary (observe only)". Our audit: 816,270 "conversions" vs 303 orders. See 12 common Google Ads mistakes.
- Google never counts more purchases than you have orders. Where: the Purchase row, by week, beside your order export. Red flag: a ratio above 1.0 in any week. Fix: Find the inflating action before you touch a bid. Our audit: one Performance Max campaign (Google's all-channel type) showed 32,485 conversions in a week. It made 5 sales.
- Purchase value equals order value. Where: the Conv. value column for Purchase, by day. Red flag: a gap above 5% against store revenue, or a fixed ₹1 per conversion. Fix: Send the real order value in the tag. Our audit: each page view counted as a ₹1 purchase.
- Google Analytics sends only sales to Ads. Where: Analytics › Admin › Data display › Events. Starred events are key events (conversions). Red flag: 1 or more non-sale key events imported as Primary. Fix: Unstar them, or keep them Secondary in Ads.
- No gaps in purchase history. Where: Goals › Conversions › Summary, by month, last 24 months. Red flag: any week with spend and 0 purchases. Fix: Repair the tag. Flag the gap in reports. Our audit: no purchases recorded March to May 2024.
- Enhanced conversions are on. They match more sales with hashed customer data. Where: Goals › Settings › Customer data use, one switch since June 2026. Red flag: off, or no customer data in 7 days. Fix: Turn it on.
- Every click and order carries a source. Where: Admin › Account settings › Auto-tagging (adds a click ID to ad links) and your order export. Red flag: auto-tagging off, or over 20% of orders with no source. Fix: Tick auto-tagging. Record a source on every order. Our audit: 91% of orders had no source.
- Offline sales are counted. Where: Tools › Data manager. Red flag: over 10% of revenue from phone, UPI or shop sales, with none of it in Google. Fix: Upload offline sales weekly, or check them in every test.
One wrong setting from 24 Sep 2025 was still live a year later. Want agents to run these 40 checks every day? See the workflow →
How should a Google Ads account be structured? (Checks 9–13)
One goal per campaign, and campaigns that live long enough to learn. Our audited account failed by changing too much.
- Each campaign bids for one goal: purchases. Where: Campaigns › Settings › Conversion goals. Red flag: more than 1 goal. Fix: Use campaign-specific goals. Keep Purchases only. Our audit: campaigns bid for 6 goals; 1 was a sale.
- Campaigns that sell are left alone. Where: Campaigns › Change history, which keeps 2 years. Red flag: replacing a campaign that sold in the last 30 days. Fix: Freeze the lineup. Judge after 8–12 weeks. Our audit: campaigns with 131 sales in May 2026 were replaced. The latest replacement spent ₹21,600 for 5 sales.
- Each product has one home. Where: Campaigns › Asset groups › Listing groups. Listing groups are the product filter of an asset group (one set of Performance Max ads). Red flag: a product in 2 or more campaigns. Fix: Give each product one campaign. Split by feed label.
- Bestsellers have their own asset group. Where: Campaigns › Asset groups. Red flag: 100% of asset groups on "All products". Fix: Label your top 50 (check 31). Add one bestseller asset group to a running campaign. Our audit: the top 10% of products brought 94% of ad sales.
- Auto-apply is off for bids and budgets. Where: Campaigns › Recommendations › Auto-apply settings. Red flag: 1 or more bidding, budget or keyword types ticked. Fix: Untick them. Review recommendations weekly.
Which Google Ads settings waste the most money? (Checks 14–22)
Targets, channel split and reach. This is where the biggest swings hide.
- Target ROAS sits where sales were best. Target ROAS is the sales value Google must find per ₹1. Where: Campaigns › Settings › Bidding. Red flag: a raise above 20% in one step. Fix: Go back to your best months' level. Our audit: targets jumped to 7.3–11.5 in August 2026. Google sales fell 40%; cost per sale rose from ₹648 to ₹988.
- One change per campaign every 2–3 weeks. Where: Campaigns › Change history. Red flag: 2 or more target, goal or budget changes inside 14 days. Fix: Change one thing, then wait. Our audit: target and goal changed the same week, so nobody can say which did the damage.
- Search gets its share of Performance Max money. Where: Insights and reports › Channel performance, with data from 6 June 2025. Red flag: Search under 30% of spend. Fix: Lower inflated targets; they set the Search share. Our audit: product ads returned ₹10.60 per ₹1 on Search and ₹0.01 on YouTube, yet YouTube got 56% of product-ad money. See Performance Max vs Search.
- You lose searches to rank, not budget. Impression share is the share of eligible searches where your ad showed. Where: Campaigns › columns icon › Competitive metrics. Red flag: over 10% lost to budget, or over 50% lost to rank. Fix: Fix rank with targets, price and photos, not money. Our audit: ads showed on 11–23% of eligible searches for 2 years.
- Search terms match what you sell. Where: Insights and reports › Search terms. Performance Max has its own view. Red flag: unrelated terms above 10% of spend. Fix: Add negatives. Performance Max takes 10,000 per campaign. Our audit: up to 17% of showings were unrelated. A Doctor Ads vendor study (84 advertisers, 2023–2025 data) found a median 39.4% of search-term spend on terms that never converted.
- Location targets people who are there. Where: Campaigns › Settings › Locations › Location options. Red flag: "Presence or Interest", the default, when you serve only some places. Fix: Select "Presence". Our audit: 96% of orders for the 1-hour-delivery products came from one city. Their ads ran across India.
- Night spend earns its keep. Where: Campaigns › Ad schedule › Day and Hour. Red flag: any time block above 2× the average cost per sale. Fix: Cap that block. Our audit: night sales (midnight to 8 am) cost about ₹500 each at ₹9,900 a month, and ₹2,000–2,600 each at ₹19,000.
- Search partners pay their way. Where: Campaigns › Settings › Networks; in Performance Max, the Search partners row in Channel performance. Red flag: cost per sale 20% or more above Google Search. Fix: Untick search partners.
- Budgets move in small steps. Where: Change history, budget changes. Red flag: a change above 20% at once. Fix: Step by 10–20%. Keep steps that pay. Google can spend 2× the daily budget in one day.
How do you audit Google Ads images, videos and ads? (Checks 23–27)
Most creative sells nothing. Leaving it running is the mistake.
- Images that never sell get swapped. Where: Asset groups › View details › Assets, plus the Conversions column. Red flag: over 50% of images at 0 sales in 90 days. Fix: Swap up to 5 every 4–6 weeks. Our audit: 64 of 79 images sold nothing. Google credits every asset in an ad, so asset numbers find duds but do not add up.
- Your own images outnumber Google's. Where: Asset optimization in campaign settings, and each asset group. Red flag: under 20 of your own images per asset group. Fix: Upload 20 of your own, plus videos. Our audit: Google's 101 self-made images got 3,533 views in 8 months. The store's own got 1.08 million.
- Every video shows a product and sells. Where: video assets in each asset group; the YouTube row in Channel performance. Red flag: any video with spend and 0 sales in 90 days. Fix: Switch it off for 3–4 weeks. Remove it if sales hold. Our audit: 13 of 16 videos sold nothing. One took ₹17,400 for ₹95.
- Ad strength is Good or Excellent. Where: the Ad strength column on Ads and on Asset groups. Red flag: any spending ad rated below Good. Fix: Add headlines, descriptions and images. Google says moving a search ad from Poor to Excellent brings 15% more clicks and conversions on average.
- Nothing is disapproved or limited. Where: Tools › Troubleshooting › Policy manager. Red flag: 1 or more disapproved or limited items that spend. Fix: Fix and appeal fast. Since 21 July 2026, decisions over 6 months old cannot be appealed from the account.
How do you audit products in Shopping and Performance Max? (Checks 28–33)
The quiet waste lives here. Since June 2026, the Products report shows cost and conversions on all Performance Max networks, so you can see it.
- Products with clicks and no sales are excluded. Where: Campaigns › Products, last 90 days, by cost. Red flag: zero-sale products above 20% of product spend. Fix: Exclude them in Listing groups. Spare new products for 30 days. Our audit: 792 of 1,027 products took 41% of product-ad money, ₹16,900 a month, with no ad sales. 381 still sold through other channels, so check first.
- No single product burns money alone. Where: Campaigns › Products, filtered to 0 conversions. Red flag: any product past ₹500, or one average order, with 0 sales in 90 days. Fix: Exclude it from ads, not from the store. Our audit: a ₹99 add-on took ₹33,644 and 324,155 clicks in a year. It sold 5 units.
- Product types return more than they cost. Where: the Products report, by product type. Red flag: under ₹1 back per ₹1 over 3, 6 and 12 months. Fix: Exclude the losing types. Our audit: the store's top product line returned ₹11.90 per ₹1 on 6% of the money.
- The feed labels your bestsellers. The feed is your product list in Merchant Center. Where: attributes custom_label_0 to custom_label_4. Red flag: 0 labels in use. Fix: Tag your top 50 "bestseller". Give them one asset group.
- The feed is clean. Where: Merchant Center › Products › Needs attention. Red flag: over 5% of products can't show, or any price mismatch. Fix: Fix prices first, then images, then broken pages. Our audit: 74 of 761 products could not show. 34 had a price that differed from the store.
- Bestsellers are in stock. Where: Merchant Center availability, against your top-20 list. Red flag: a top-20 product out of stock over 7 days. Fix: Check weekly. Push only products with 2 or more weeks of stock. Our audit: 11 of 40 bestsellers could not be bought.
Teams skip checks 28–33 because 1,027 products do not fit in a quick look. Agents check every product, every day. See the 55+ automations →
Does your landing page or store hurt your Google Ads? (Checks 34–36)
The ad buys the click. The page, the price and the offer close the sale.
- Every ad lands on a working mobile page. Where: Insights and reports › Landing pages; in Performance Max, also Final URL expansion, on by default. Red flag: a mobile-friendly click rate under 100%, or any broken page. Fix: Fix broken pages. Exclude pages that cannot sell.
- Your price can win the click. Where: Merchant Center › Analytics › Pricing, and Auction insights. Red flag: a top seller over 10% above the benchmark price. Fix: Fix price, photo and title before you add budget. Our audit: even at a target of 4, about 70% of searches were lost to rank.
- Buyers get a reason to buy now. Where: your store's discount history, outside Google Ads. Red flag: no short sale in 90 days. Fix: Run a 2–3 day sale at 15–20% off, every 2–3 months at most. Our audit: a 2-day, 20% sale lifted daily orders 2.6×. A 19-day sale at 12–15% gave 1.1×.
How do you know if Google Ads is actually working? (Checks 37–40)
Judge the ads by what the store sold, not by Google's scoreboard.
- Spend and store revenue grow together. Where: monthly Google Ads cost beside monthly store revenue. Red flag: spend growing over 2× faster than revenue across 12 months. Fix: Cut no-sale spend before you add more. Our audit: ad spend rose 96% year on year. Online revenue rose 31%.
- Weekly sales have an undo trigger. Where: Purchase conversions, by week. Red flag: 20% or more below the 8-week average, 2 weeks in a row. Fix: Undo the last change. Our audit: Google sales fell from 28 a week to 17 after 7 Aug 2026, at about the same spend.
- Big channels have passed a test. Where: Campaigns › Experiments, for Search, Display and Video. Red flag: a channel above 20% of spend, never tested. Fix: Switch it off in one campaign for 3–4 weeks. Compare all orders. Our audit: YouTube took ₹433,000 from October 2025 for 45 traced sales. Its later effect is unproven until tested.
- Waste is one number, tracked monthly. Where: the template below. Red flag: over 20% of spend on things with 0 sales in 90 days. Fix: Cut the biggest first. Our audit: 48% of a ₹71,800 monthly budget sold nothing. Caveat: the best months had the same waste. Cutting it saves money; it does not bring sales back.
Is there a free Google Ads audit template?
Yes: this one. Select the table, copy it, and paste it into Google Sheets or Excel. Mark each row Pass or Fail. Put a date in Fix by.
| # | Check | Red flag | Pass / Fail | Fix by |
|---|---|---|---|---|
| 1 | Only sales are Primary | Any non-sale Primary | ||
| 2 | Purchases vs orders | Ratio above 1.0 | ||
| 3 | Purchase value | Gap above 5% | ||
| 4 | Analytics key events | Non-sale event imported | ||
| 5 | Purchase history gaps | Week with spend, 0 sales | ||
| 6 | Enhanced conversions | Off, or no data 7 days | ||
| 7 | Click and order source | Over 20% unsourced | ||
| 8 | Offline sales | Over 10% missing | ||
| 9 | One goal per campaign | More than 1 goal | ||
| 10 | Selling campaigns kept | Seller replaced | ||
| 11 | One home per product | Product in 2+ campaigns | ||
| 12 | Bestseller asset group | 100% "All products" | ||
| 13 | Auto-apply | 1+ bid or budget type on | ||
| 14 | Target ROAS changes | Raise above 20% | ||
| 15 | Change pace | 2+ changes in 14 days | ||
| 16 | Search share | Under 30% of spend | ||
| 17 | Lost impression share | Budget over 10%; rank over 50% | ||
| 18 | Search terms | Unrelated over 10% | ||
| 19 | Location option | "Presence or Interest" | ||
| 20 | Time of day | Block over 2× cost per sale | ||
| 21 | Search partners | 20%+ dearer than Search | ||
| 22 | Budget steps | Change above 20% | ||
| 23 | Zero-sale images | Over 50% of images | ||
| 24 | Your own images | Under 20 per asset group | ||
| 25 | Videos | Any video, 0 sales in 90 days | ||
| 26 | Ad strength | Below Good | ||
| 27 | Policy status | 1+ disapproved or limited | ||
| 28 | Zero-sale products | Over 20% of product spend | ||
| 29 | Single-product burners | Over ₹500, 0 sales | ||
| 30 | Product-type return | Under 1× in 3 windows | ||
| 31 | Bestseller labels | 0 labels | ||
| 32 | Feed health | Over 5% can't show | ||
| 33 | Bestseller stock | Top-20 out over 7 days | ||
| 34 | Landing pages | Under 100% mobile-friendly | ||
| 35 | Price vs benchmark | Over 10% above | ||
| 36 | Reason to buy now | No short sale in 90 days | ||
| 37 | Spend vs revenue | Spend growth over 2× | ||
| 38 | Undo trigger | 20% down, 2 weeks | ||
| 39 | Channel tests | 20%+ of spend, untested | ||
| 40 | Waste share | Over 20% of spend |
How long does a Google Ads audit take?
A first pass takes 30 minutes. A manual review of a mid-sized account takes 3–4 hours. Joining ads to store orders takes days. PPC consultant Emily Wood teaches a 30-minute framework (July 2025). A TripleDart vendor guide puts a manual mid-sized review at three to four hours (updated July 2026).
The time goes into data, not clicks. Settings checks (6, 7, 9, 13, 19, 21) take minutes. Checks 2, 3, 8 and 37 need store orders. Our audit joined every campaign since April 2023 to 5,788 orders. Tools speed up the settings checks; see the best AI tools for Google Ads.
How often should you audit Google Ads?
Run all 40 checks every quarter, the money checks monthly, and the sales check weekly. Do not change things faster than Google learns: a bid strategy needs "a few conversion cycles (1-2 typically)" to adjust.
- Weekly: checks 2, 33 and 38.
- Monthly: checks 17, 18, 20, 28–30, 32 and 40.
- Every 4–6 weeks: check 23.
- Every quarter: all 40. Export Change history too; it keeps only 2 years.
- Within 7 days of any tracking or goal change: checks 1–5.
Once a year is too slow. The reports our agents send cover the weekly and monthly rows. Deciding who runs them? Compare agency, freelancer and AI agent costs.
How we know this
The audit numbers come from one real audit of an Indian D2C store. We keep the brand private. Data: Google Ads for every campaign since April 2023 by day, place, hour and asset; Google Analytics; account screenshots from 10 Oct 2026; 5,788 store orders with line items and the current catalog; the Google Ads product report. Data runs from January 2023 to 9 Oct 2026. The audit logged 145 findings in seven stages. We kept the 40 checks any account can run. Every audit number here is a direct count from the account, the store or a screenshot; where an effect is not proven, we say so. The red-flag thresholds are our working rules, not Google's. We checked every menu path on Google's help pages on 14 Oct 2026. Third-party numbers are vendor studies, named and dated.
Questions people ask
What is a Google Ads audit?
A Google Ads audit checks your tracking, settings, ads, products and results against real sales. It finds money that buys nothing and sales you miss. A full audit also reads your store orders, because Google's own numbers can be wrong. In one real audit, Google counted 816,270 conversions against 303 orders.
Can I audit my own Google Ads account?
Yes. Use the 40 checks and the template on this page. Start with tracking (checks 1 to 8), because every other check depends on it. The hard part is not the settings. It is joining Google's numbers to your store orders, product by product.
What access does a Google Ads audit need?
Read-only access is enough. You need Google Ads, Merchant Center and Google Analytics, plus an export of store orders with dates, values and products. Nothing needs to change during the audit.
How do you audit a Performance Max campaign?
Use four reports. Channel performance shows where the money went, with data from 6 June 2025. The search terms report has a Performance Max view. The Products report shows cost and conversions on all networks since June 2026. Listing groups let you exclude products that never sell.
What does a Google Ads audit usually find?
Spend that buys nothing. In our audit, 48% of a ₹71,800 monthly budget sold nothing. A vendor study by Doctor Ads found the median advertiser spent 39.4% of search-term cost on terms that never converted.
Sources
- Google Ads Help: About primary and secondary conversion actions
- Google Ads Help: Updates to your enhanced conversions settings (2026)
- Google Ads Help: About auto-tagging
- Google Analytics Help: Mark events as key events
- Google Ads Help: About Data manager
- Google Ads Help: About campaign-specific conversion goals
- Google Ads Help: Change history
- Google Ads Help: Manage auto-apply recommendations
- Google Ads Help: About the channel performance report for Performance Max
- Google Ads Help: Get impression share data
- Google Ads Help: About the search terms report
- Google Ads Help: About the search terms report in Performance Max
- Google Ads Help: More goals and image controls coming to Performance Max (10,000 negative keywords)
- Google Ads Help: About advanced location options
- Google Ads Help: Prevent clicks outside of your geo-targeted locations
- Google Ads Help: Set up an ad schedule
- Google Ads Help: About average daily budgets
- Google Ads Help: Duration of the learning period for campaigns
- Google Ads Help: Conversion reporting by asset in Performance Max campaigns
- Google Ads Help: About responsive search ads (Ad Strength)
- Google Ads Help: Fix a disapproved ad or appeal a policy decision
- Google Ads Help: About Product Reporting
- Google Ads Help: Listing groups in Performance Max
- Merchant Center Help: Custom label [custom_label_0–4]
- Merchant Center Help: The Needs attention tab
- Merchant Center Help: Price benchmarks (Analytics, Pricing)
- Google Ads Help: Evaluate the performance of your landing pages
- Google Ads Help: About Final URL expansion in Performance Max
- Google Ads Help: Use auction insights to compare performance
- Google Ads Help: Set up a custom experiment
- Doctor Ads (vendor), Igor Ivitskiy: How much Google Ads budget is wasted (updated 28 Sep 2026)
- TripleDart (vendor): How to use AI to audit your Google Ads account (updated 8 Jul 2026)
- Emily Wood: How to audit a Google Ads account (17 Jul 2025)