Google Ads Ad Group Structure, Enhanced Conversions, and Custom Goals: 3 Questions Answered
These questions came from the Google Ads Masterclass community. Ad group structure, enhanced conversions, and campaign-level custom goals are three areas where the defaults in Google Ads are rarely the right choice. Here is how I actually handle each one across client accounts.
Ad group structure affects quality score, ad relevance, and how the algorithm learns what you are selling. Enhanced conversions affect the accuracy of your bidding data. Custom goals affect which actions Google is actually optimizing for. Get any one of these wrong and the others cannot save you.
These are not abstract settings questions. They connect directly to how much signal your campaigns generate and whether that signal is pointing at the right thing. Here is the concrete logic behind each recommendation. For related questions on campaign structure and PMax budget, see Google Ads PMax Budget, Campaign Structure, and Consolidation: 3 Real Questions Answered.
Brand traffic inflating your ROAS, one asset group blending completely different buyers: this free guide walks through the exact structure issues I find most often, with the fix for each.
01Question 1: Should I Use Enhanced Conversions?
Yes. No qualification. If you have Google Ads conversion tracking on your website, you should have enhanced conversions enabled. The question is not whether to use them but how to implement them correctly.
Enhanced conversions supplement your existing conversion tags by sending hashed first-party data (email address, phone number, name) alongside conversion events. Google uses this to match conversions that were not directly attributed through the standard click tag, such as conversions that happened after a cookie was blocked, a browser was switched, or a tracking gap occurred.
What Enhanced Conversions Actually Fix
Standard Google Ads conversion tracking relies on the click ID being passed from the ad click through to the thank-you page. Any break in that chain (privacy browsers, iOS restrictions, server-side redirects, cross-device journeys) means the conversion does not get attributed. In many accounts, this gap is 15 to 30% of actual conversions.
Enhanced conversions use hashed customer data to reattribute those lost conversions back to the correct campaign and ad group. The result is a more accurate picture of what is actually working, which means smart bidding is optimizing against better data.
How to Enable Enhanced Conversions Correctly
Go to Tools and Settings → Conversions → select your primary purchase conversion → scroll to Enhanced Conversions → turn on. Choose “Automatic” if your checkout page has standard email/name fields. Choose “Manual tagging” if you need to specify which fields to capture.
Do this for each purchase conversion action separatelyEnhanced conversions only work when your Google tag fires on the page where the customer data is present. If your purchase confirmation page does not load the Google tag (common with server-side tracking setups), enhanced conversions will not capture anything.
In the conversion settings, there is an enhanced conversions diagnostics section. It will show you what percentage of conversions have enhanced data attached. Anything below 50% indicates a tagging issue on your confirmation page or a mismatch in the field capture setup.
Target: 70% or higher enhanced match rateEnabling enhanced conversions in Google Ads settings but not updating the tag is the most common implementation error. The toggle alone does nothing. The Google tag on your confirmation page must be updated to capture and hash the customer data fields. Use Google Tag Manager with the enhanced conversions variable to do this without touching site code.
02Question 2: How Should I Structure Ad Groups?
The standard advice is to group keywords by theme and write one set of ads per theme. That advice is fifteen years old and reflects how the old quality score system worked. Today, the question you should be asking is different: how many ad groups do I need to give the algorithm enough signal in each one while keeping keyword-to-ad relevance meaningful?
The practical answer for most accounts is fewer ad groups than you think, not more. One very common pattern is creating a separate ad group for every keyword variation, ending up with dozens of ad groups that each contain one keyword and no historical data. Smart bidding cannot function at that fragmentation level.
Ad Group Structure by Campaign Type
| Campaign Type | Recommended Ad Group Count | Grouping Logic |
|---|---|---|
| Brand Search | 1 to 2 | Brand name variations in one group. Product-specific brand terms (brand + product) in a second if volume justifies it. |
| Nonbrand Search | 3 to 6 | Group by buyer intent stage: high intent (buy, purchase, order), product category, and competitor terms if applicable. |
| Performance Max | 1 to 4 asset groups | One asset group per major product category if you have distinct creative for each. Otherwise one asset group total with all products. |
| Display / Demand Gen | 2 to 4 | Group by audience type: remarketing, similar audiences, interest targeting. Never mix audience types in one ad group. |
The principle across all campaign types is the same: each ad group needs enough conversion volume to be meaningful, and the ads in each group need to be relevant to the keywords or audience signals in that group. Those two requirements usually push in opposite directions. More specificity means better relevance but thinner data. More consolidation means better data but less relevance. The right balance is closer to consolidation than most accounts currently are.
The Single Keyword Ad Group Trap
Single keyword ad groups (SKAGs) were popular because they gave near-perfect keyword-to-ad relevance and appeared to boost quality score. The problem is that quality score is now less important as a standalone metric, and smart bidding needs aggregated signals to function. According to Google’s smart bidding guidance, campaigns need sufficient conversion volume to optimize effectively. An account with 40 ad groups each generating 2 conversions per month is worse than an account with 5 ad groups each generating 16 conversions per month, even if the second account has slightly lower keyword-to-ad relevance scores.
- Keywords have genuinely different buyer intent (high intent vs. research stage)
- Different landing pages are needed for different keyword groups
- You want to run different ad messaging for different product categories
- The ad group already generates 30+ conversions per month on its own
- Just to have cleaner organization in the UI
- To put each keyword in its own group for “control”
- To match keyword match types (broad vs. phrase) into separate groups
- To separate desktop and mobile intent (handled by bid adjustments)
If you are considering splitting an ad group, ask: will each resulting ad group generate at least 15 conversions per month independently? If not, you are fragmenting signal without gaining meaningful strategic control. Keep it consolidated until the volume justifies separation.
I review every submission personally. Structure, tracking, bidding, ad groups. A written breakdown of what I find, at no cost.
03Question 3: Should Lead Forms and Call Bookings Count as Conversions?
This depends entirely on how closely these micro-conversions correlate with the outcome you actually care about. If every lead form submission becomes a paid customer 90% of the time, yes, count it. If 10% of lead form submissions convert to revenue, counting the form submission as your primary conversion will teach the algorithm to optimize for form submissions at the expense of the quality that determines whether they close.
The real question is not “should I track these?” but “should I bid on these?” You can and should track lead forms, call bookings, and other micro-conversions as secondary conversions. The decision point is whether they belong in your primary conversion column, which is the column smart bidding optimizes against.
Primary vs. Secondary Conversions
Google Ads lets you set each conversion action as either included in conversions (used for bidding) or not included (tracked only). Most accounts set everything as included and then wonder why their smart bidding is optimizing for low-quality leads.
The rule I use: include a conversion in bidding only if it is within one step of revenue. For ecommerce, that means the purchase event only. For lead generation, it means qualified lead, booked meeting, or proposal request, depending on how your pipeline works. Exclude anything that is two or more steps from revenue.
- Purchase / completed transaction
- Qualified lead (after manual qualification step)
- Booked consultation call
- Trial signup with credit card
- Paid subscription started
- Contact form submission (unqualified)
- Lead magnet download
- Newsletter signup
- Product page view
- Add to cart (unless it correlates very strongly to purchase)
Call bookings are a special case. If your business closes 40% or more of booked calls, counting the booking as a primary conversion is reasonable. Smart bidding can optimize toward people likely to book and show up. If your close rate is below 20%, the booking is too far from revenue to be a reliable bidding signal. Track it, observe the pattern, but exclude it from primary conversions until the correlation with revenue is established from your own data.
04Campaign-Specific Goals: When to Override Account-Level Conversions
By default, every campaign in your account bids toward the same conversion actions you have set at the account level. This is fine when all your campaigns have the same objective. It becomes a problem when they do not.
The most common scenario: you have an ecommerce campaign running alongside a lead generation campaign in the same account. If both are set to account-level conversions, the ecommerce campaign may be crediting lead form submissions and the lead gen campaign may be crediting purchases. The algorithm gets mixed signals about what to optimize for in each campaign.
When to Use Campaign-Level Conversion Goals
Set campaign-specific goals whenever a campaign has a materially different objective from the rest of the account. The two most common cases are: a remarketing or upper-funnel campaign where micro-conversions (page views, video views) are the intended outcome and should not be mixed with purchase signals; and a lead generation campaign in an account that also runs ecommerce campaigns.
To set campaign-specific goals: go into the campaign settings → scroll to Conversions → click the option to use conversion goals specific to this campaign → add only the conversion actions relevant to that campaign’s objective.
The single most common mistake I see in lead generation accounts is including both a newsletter signup (high volume, low intent) and a booked consultation (low volume, high intent) in the same campaign’s primary conversions. Smart bidding will optimize toward whichever conversion is easier to get, which is almost always the newsletter signup. The result is high volume of low-quality conversions and a reported CPA that looks excellent but does not correspond to actual revenue.
05The Underlying Principle Connecting All Three
Enhanced conversions, ad group structure, and campaign-level goals are all versions of the same problem: getting accurate, concentrated signal into Google’s algorithm so it can make good decisions on your behalf.
Enhanced conversions increase the accuracy of the signal by recovering attribution gaps. Ad group consolidation increases the volume of signal per optimization unit. Clean conversion goal configuration makes sure the signal is pointing at the right outcome. All three work together. Fixing one while ignoring the others produces partial results. The same logic applies to campaign-level consolidation decisions — if you have not read through how that works at the campaign level, this breakdown on PMax budget and campaign consolidation covers it directly.
More of your actual conversions get attributed correctly. Smart bidding is making decisions based on 85% of real conversions instead of 65%, which changes the algorithm’s understanding of which users, queries, and times of day actually convert.
More conversions flow into fewer optimization units. Each ad group or campaign reaches the minimum conversion threshold for smart bidding to learn effectively. Thin data across many ad groups is worse than concentrated data in fewer, well-structured groups.
The algorithm is optimizing toward actions that actually lead to revenue, not toward whichever action is easiest to get. Removing low-intent conversions from primary goals often causes reported conversion volume to drop but revenue to hold or increase, because the algorithm stops wasting budget on low-quality actions.
When an account is underperforming, check these three in order. First: is enhanced conversions enabled and showing a match rate above 60%? If not, fix that before adjusting bids or structure. Second: does each ad group or campaign generate enough conversions to be meaningful? If not, consolidate. Third: are the conversions you are bidding on actually correlated with revenue? If you optimize conversion goals before fixing attribution, you are optimizing a distorted signal.
I check enhanced conversions setup, ad group structure, and conversion goal configuration in every audit. Written findings, no cost, no pitch.
06Action Plan
Here is the specific sequence of actions for an account that has not addressed any of these three areas yet. Do them in this order. Each step makes the next one more effective.
Go to Tools → Conversions and list every action that is currently set to “Include in Conversions.” For each one, write down how directly it correlates to revenue. Anything that is two or more steps from a closed sale should be moved to Secondary (observation only) immediately.
Takes 30 minutes. Do this first.Turn on enhanced conversions for your primary purchase or lead conversion action. Verify the Google tag is correctly capturing and hashing customer data fields on your confirmation page. Use Google Tag Manager to implement without touching site code. Check diagnostics after two weeks to confirm match rate.
Target match rate: 70% or aboveExport your ad group performance for the last 30 days. Any ad group with fewer than 10 conversions per month is a consolidation candidate. List which groups could be merged without losing meaningful relevance, and start with the lowest-volume ones.
If you have campaigns serving different purposes in the same account, override the account-level conversion goals for those campaigns. Assign only the conversion actions relevant to each campaign’s specific objective. This takes five minutes per campaign and immediately improves bidding signal quality.
Check campaign settings → Conversions → Use campaign-specific goals- 1Is enhanced conversions enabled and are you seeing a match rate in diagnostics? If not, your conversion data is incomplete and any bid adjustments are based on partial information.
- 2Are all the conversions in your primary column actually correlated with revenue? Pull a 90-day sample and check whether lead form submissions, for example, actually result in sales at a meaningful rate.
- 3Does each ad group generate at least 10 conversions per month? Below that threshold, the algorithm does not have enough within-group signal to make meaningful bid decisions. Consolidate before optimizing.