Most businesses hiring a Google Ads management agency assume the agency is doing everything right. Often, they're not. Campaigns run on autopilot, budgets drift, and reporting gets polished to look better than the underlying numbers.
This isn't a list of nice-to-haves. These are seven specific strategies a competent agency should be executing on your account right now. Use it as a checklist when evaluating your current partner or deciding whether to bring one on. Every item here is concrete and testable. If your agency can't explain how they're doing these things in your actual account, that's your answer.
When campaigns are organized purely by keyword theme, informational traffic competes for budget alongside high-intent transactional queries. Someone searching "how does Google Ads work" and someone searching "Google Ads agency pricing" are not the same prospect. Treating them identically wastes budget and produces misleading performance data.
Campaigns should be segmented by funnel stage: branded, transactional, and informational traffic each behave differently and convert at different rates. Branded campaigns protect your own name and typically convert at lower cost. Transactional campaigns target high-intent queries and deserve tighter bidding control. Informational traffic, if targeted at all, belongs in separate campaigns with appropriate budget caps.
This structure also directly affects Quality Score. Google publicly documents that Quality Score is based on Expected CTR, Ad Relevance, and Landing Page Experience. When your ad copy, keywords, and landing pages are tightly aligned by intent, Ad Relevance and Landing Page Experience scores improve. That lowers your cost-per-click over time.
1. Audit existing campaigns and tag each ad group by the intent stage it's actually serving.
2. Separate branded traffic into its own campaign with its own budget and bid strategy.
3. Split transactional and informational keywords into distinct campaigns so budget can be allocated independently.
4. Match landing pages to the intent of each campaign, not just the product or service category.
Don't let a single campaign carry all traffic types just because it's easier to manage. That convenience costs you control. If your agency can't show you a campaign map that reflects funnel stages, ask them to build one. If they push back, that tells you something.
Google significantly expanded Broad Match behavior starting in 2021 and has continued expanding it since. Broad match now triggers on semantically related terms, synonyms, and implied intent. That flexibility is useful when managed well. When it's not managed, it bleeds budget into queries that have nothing to do with your business.
A shared negative keyword library, applied at the account level and reviewed weekly through search term reports, is one of the highest-ROI maintenance tasks in any Google Ads account. It's also one of the most commonly neglected. Agencies that set campaigns up and check in monthly are almost certainly allowing irrelevant spend to accumulate between reviews.
The goal is a living document: negative keywords organized by category (competitor names, irrelevant modifiers, geographic exclusions, job-seeker terms, etc.) and applied consistently across campaigns. New negatives get added as new irrelevant queries appear. Old negatives get audited to make sure they're not accidentally blocking valuable traffic.
1. Pull the search terms report weekly and flag any query that generated a click without converting or that clearly doesn't match your customer profile.
2. Add confirmed irrelevant terms to a shared negative keyword list applied at the account level.
3. Organize negatives by category so you can audit them without rebuilding from scratch.
4. Review the negative list quarterly to catch any terms that may have become relevant or that are blocking good traffic.
Ask your agency to show you the search terms report from the last 30 days and the negative keyword list. If the list hasn't been updated recently, or if the search terms report shows obvious irrelevant queries that are still running, the account isn't being actively managed.
Smart Bidding is not a set-it-and-forget-it improvement. tCPA and tROAS bidding require enough conversion data to learn from. Applying automated bidding to a campaign with 10 conversions a month doesn't produce smarter optimization, it produces optimization against noise. The algorithm makes confident decisions based on a sample size that's too small to be statistically meaningful.
Google's own documentation recommends a minimum of 30 to 50 conversions per month at the campaign level before switching to tCPA bidding. For tROAS, the recommendation is at least 50 conversions in the past 30 days. These thresholds exist because the algorithm needs signal volume to identify patterns. Below those thresholds, manual CPC gives you more control and produces more predictable results.
The right approach is staged: start with manual CPC while building conversion history, transition to Enhanced CPC or Maximize Conversions as volume grows, and move to tCPA or tROAS only once the campaign consistently meets Google's recommended thresholds.
1. Check current monthly conversion volume at the campaign level, not the account level.
2. If any campaign is running tCPA or tROAS with fewer than 30 conversions per month, flag it for review.
3. Set a documented threshold for when each campaign will transition to automated bidding.
4. Monitor performance for at least two to four weeks after any bidding strategy change before drawing conclusions.
Agencies sometimes switch to Smart Bidding prematurely because it reduces the manual work required. That's a convenience decision, not a performance decision. If your agency can't tell you the conversion volume threshold they use before switching bid strategies, push for a specific answer.
Running two RSAs in an ad group is not ad testing. Testing requires a hypothesis, a controlled environment, a defined evaluation period, and a documented process for acting on what you learn. Without that structure, you're just running ads and hoping one outperforms the other.
Since Google retired Expanded Text Ads in June 2022, RSAs are the standard search ad format. Google's Combinations report inside Google Ads shows asset-level performance ratings: Learning, Low, Good, and Best. These ratings are usable data, not just labels. A real testing system uses them as inputs: underperforming assets get replaced with new hypotheses, and the reasoning behind each change gets documented so learnings accumulate over time.
The hypothesis matters. Instead of "let's try a different headline," the approach should be "we believe leading with price will outperform leading with outcome for this intent stage." That specificity makes the result teachable, not just a data point.
1. Document the current best-performing asset combinations using the Combinations report.
2. Identify the lowest-rated assets and form a specific hypothesis for what to replace them with and why.
3. Make one change at a time so you can attribute performance shifts to a specific variable.
4. Set a review cadence (typically monthly) and record what changed, what the hypothesis was, and what the result showed.
Ask your agency for the last three copy tests they ran on your account. If they can't tell you the hypothesis behind each test and what they learned, you don't have a testing system. You have an account that occasionally gets new ads written.
Most agencies optimize to the click. They lower CPCs, improve CTR, and report on impression share. None of that tells you what's happening after the click. If your landing pages are converting at different rates across campaigns, and your agency doesn't know that, they're making bidding and budget decisions with incomplete information.
GA4 replaced Universal Analytics on July 1, 2023. GA4-linked Google Ads accounts allow conversion import and audience sharing. That integration gives you page-level data: conversion rates by landing page, session duration, and engagement metrics that indicate whether paid traffic is landing in the right place. Agencies that use this data feed it back into campaign decisions: pages with low conversion rates get flagged for CRO work before more budget is pushed to them, and pages that convert well become the benchmark for the rest.
1. Confirm GA4 is properly linked to your Google Ads account and that conversion data is flowing correctly.
2. Pull landing page performance data segmented by campaign to identify which pages are converting paid traffic and which aren't.
3. Flag any campaign where significant budget is directed to a landing page with a notably lower conversion rate than the account average.
4. Bring landing page performance into the monthly review conversation, not just platform metrics.
If your agency's reporting doesn't include landing page conversion rates alongside campaign performance, ask for it. The best agencies treat the click as the beginning of the data story, not the end of it.
Duplicate conversion tags, miscategorized actions, and missing events are common in inherited accounts. When you optimize against flawed conversion data, you get flawed results with confidence. An account can show strong conversion volume while actually tracking form submissions, page views, and phone clicks all as separate "conversions", inflating the numbers and misleading the algorithm.
Google's own guidance distinguishes between Primary conversions, which are used for bidding, and Secondary conversions, which are observation-only. That distinction matters enormously for Smart Bidding accuracy. If a micro-conversion like a page view is categorized as Primary, the algorithm optimizes for page views, not leads or sales.
Every new account engagement should start with a tracking audit before a single bid adjustment is made. That means reviewing every conversion action in the account, confirming tag implementation via Google Tag Manager or direct tag, and verifying that what's being reported as a conversion actually represents a business outcome.
1. Pull the full list of conversion actions in the account and categorize each as Primary or Secondary based on its actual business value.
2. Check for duplicate tags by reviewing Google Tag Manager and the conversion action settings side by side.
3. Verify that phone calls, form submissions, and purchase events are firing correctly using Tag Assistant or the real-time view in GA4.
4. Recategorize any micro-conversions currently set as Primary to Secondary so they don't influence Smart Bidding decisions.
If your agency inherited your account and didn't perform a tracking audit in the first two weeks, ask them to do one now. Optimizing against bad data doesn't produce bad results slowly, it produces them confidently, which is worse.
CTR, Impression Share, and Quality Score are diagnostic tools. They tell you something about account health, but they don't tell you whether the account is making your business money. Agencies that lead with these metrics in reporting are often doing so because the business outcome numbers are harder to improve and harder to explain.
A well-run agency reports on cost-per-lead, cost-per-acquisition, and revenue contribution. Those numbers connect directly to what the business cares about. Platform metrics belong in the report as supporting context, not as the headline.
Reporting cadence matters too. Weekly performance pulses should be brief: what moved, why, and what's being done about it. Monthly strategic reviews should be deeper: trend analysis, budget pacing, and decisions about what to test or change in the coming period. Mixing those two into a single monthly report produces something that's too detailed to act on quickly and not detailed enough to drive strategy.
1. Identify the two or three business metrics that actually matter to your organization: cost-per-lead, cost-per-acquisition, ROAS, or revenue.
2. Ask your agency to lead every report with those metrics before anything else is discussed.
3. Establish a weekly check-in format (email or dashboard) that surfaces performance changes within 48 hours of them happening.
4. Use monthly reviews for strategic decisions, not status updates.
If your monthly report leads with CTR and Impression Share, ask your agency to rebuild the report around cost-per-lead and cost-per-acquisition. If they resist or say those metrics are hard to track, that's a tracking problem, not a reporting preference.
Run through this list against your current account or agency relationship. A good agency should be able to show you exactly how they're handling each of these, not in theory, but in your actual account. Intent-based campaign structure, active negative keyword management, data-appropriate bidding, systematic copy testing, landing page integration, clean conversion tracking, and outcome-based reporting. These aren't advanced tactics. They're the baseline.
If the answers are vague or the reporting doesn't reflect real business outcomes, that's a structural problem, not a communication one. Vague answers usually mean vague management.
Triad Media Lab runs senior-level paid media management across Google Ads, Meta, Microsoft, LinkedIn, Amazon, Local Service Ads, and more. No account handoffs, no black-box reporting, no long-term lock-ins. If you want to see what active, transparent management looks like, learn more about our services.
