Google Ads Agency Pre-Sale Audit: Evaluate Account Health Before Signing a Client

1. Make the pre-sale audit a scoping decision, not a free optimization project

A Google Ads client acquisition audit is useful when it answers a small set of commercial and operational questions: Is the account measurable? Does the current setup support the client’s stated objective? What could prevent effective management? How much remediation is likely? What evidence is missing? The audit should help you decide whether to take the engagement, what work to include, and what dependencies the client must resolve. It is not a promise that a particular CPA, ROAS, or revenue increase will follow a takeover.

Set expectations before requesting access. Explain that a pre-sale review is a diagnostic, that platform data may not represent the complete business picture, and that recommendations will be limited by the access and records provided. Avoid pitching fixes as guaranteed savings or forecasting performance from a few days of account data. An inherited account may have tracking defects, offline sales, delayed lead qualification, inventory constraints, seasonality, or changes outside Google Ads that make a platform-only conclusion incomplete.

Agree on the audit boundary before opening the account

Define the account, products or services, business locations, date range, and decision the audit is intended to support. Confirm whether the client expects a high-level sales review, a technical tracking assessment, or a deeper campaign-by-campaign diagnosis. A bounded review protects both sides: the prospect receives useful findings, while the agency avoids spending days rebuilding campaign plans before a contract exists.

• Request read-only access where possible, and confirm the correct Google Ads customer ID, manager-account relationship, time zone, currency, and billing profile.
• Ask which outcomes matter to the business: qualified leads, booked appointments, completed purchases, contribution margin, store visits, or another agreed result.
• Record the current agency or internal team’s responsibilities, planned promotions, sales capacity, fulfillment limits, and any recent account changes.
• Agree on the review period. Use at least 90 days for a baseline when volume permits; extend it to 6–12 months if seasonality, low conversion volume, or long sales cycles make a short window misleading.
• Clarify what is out of scope, such as a full landing-page review, analytics implementation, CRM reconciliation, feed rebuild, or extensive creative production.

2. Gather evidence consistently and protect the audit trail

A repeatable Google Ads account audit checklist for agencies starts with evidence collection. Look at the account settings and reporting alongside business records, analytics, and the prospect’s explanation of how customers become valuable. Keep screenshots or dated exports of important settings and results so that findings can be traced to what you actually reviewed. Record the report date, time zone, attribution setting, conversion actions included, and any filters applied.

Build a minimum evidence pack

Before interpreting performance, collect enough context to distinguish an account problem from a reporting or business problem. Google Ads reports can show what the platform attributes to advertising; they do not automatically establish that a lead was qualified, a sale was incremental, or the business could fulfill additional demand.

• Account configuration: time zone, currency, auto-tagging, linked analytics properties, conversion settings, location options, language, and network settings.
• Performance by campaign and conversion action: cost, impressions, clicks, conversions, conversion value, CPA, ROAS, impression share, and budget-limited status where applicable.
• Change history: budget, bidding, targeting, conversion, campaign, and asset changes. Note whether a recent change makes the selected reporting period a poor baseline.
• Business context: product margins or allowable acquisition cost, lead-to-sale rates, average sales cycle, seasonal calendar, target geography, inventory, and sales capacity.
• Measurement context: analytics and tag setup, consent behavior, CRM or point-of-sale records, call tracking, offline conversion imports, and the process used to qualify leads.

Check that the reporting period is complete. Recent conversions may still be attributed back to earlier clicks, so the newest days can look artificially weak, especially for high-consideration products or sales-assisted leads. Ask for the account’s typical conversion delay and compare the time from click to conversion, or the equivalent available reporting, across several months. If the majority of conversions arrive within 14 days, a 14-day recent-data exclusion may be a reasonable starting point for directional comparison; if the sales cycle is longer, use the observed lag instead. Do not apply one universal lag window to every account.

3. Validate conversion measurement before judging efficiency

Measurement is the foundation of the audit. Automated bidding learns from the conversion actions and values supplied to it. If those actions include duplicate events, low-intent page views, unqualified calls, or incomplete revenue values, the bidding system can optimize toward the wrong outcome even when campaign settings look polished. First identify which actions are primary for bidding and which are secondary observations. Then determine whether each action represents meaningful business value.

Trace the conversion from ad interaction to business outcome

For each primary conversion, document its source, trigger, attribution approach, counting method, and downstream business meaning. A purchase action should be compared with order records and should pass an appropriate transaction value. A lead action should be checked against CRM stages or another reliable qualification process. For phone leads, confirm call duration thresholds, duplicate handling, business-hour coverage, and whether calls are answered. For offline conversions, ask how identifiers are captured, how records are matched, and how frequently qualified outcomes are uploaded.

• Check for multiple tags or imported events that may record the same purchase or lead more than once.
• Confirm that primary actions are genuinely valuable and that low-intent events are not included in a blended conversion goal.
• Compare Google Ads totals with analytics, CRM, ecommerce, or call-tracking records over matching dates and definitions.
• Investigate large or changing gaps. A persistent discrepancy above roughly 10% deserves explanation; a gap above 20% is a material risk signal, not an automatic verdict, because systems may use different attribution, consent, and counting rules.
• Check whether conversion values represent revenue, gross margin, lead value, or a static proxy. Do not describe revenue ROAS as profit ROAS unless the values actually reflect profit.
• Verify consent and tag behavior across relevant regions and devices with the client’s technical or privacy owner. Do not infer legal compliance from a dashboard alone.

Separate measurement confidence from campaign efficiency in the audit. If tracking is unverified, say that reported CPA is a platform-reported figure and may not equal cost per qualified lead or sale. If a conversion lag window is not mature, mark recent CPA and ROAS as provisional. For a target CPA or target ROAS proposal, establish the business’s allowable acquisition cost or return threshold first, then assess whether the historical data is sufficiently reliable and representative to use as a starting point.

4. Inspect campaign structure, targeting, and demand coverage

A well-organized account is not simply one with many campaigns or tightly themed ad groups. The structure should support meaningful budget control, clear reporting, appropriate bidding signals, and the client’s real business priorities. When you audit Google Ads account before taking over, inspect whether campaign boundaries reflect products, services, locations, margins, or operational constraints—and whether that separation has enough conversion volume to support the selected bidding approach.

Review settings and query coverage before recommending a rebuild

For Search, review search terms, keyword and match-type coverage, negative keyword governance, brand versus nonbrand reporting, geographic settings, ad relevance, landing-page alignment, and the distribution of spend across themes. Look for recurring irrelevant queries, brand demand being reported as generic growth, geographic leakage, and high-cost themes with no qualified outcomes. Compare search-term behavior with actual sales or lead quality when the prospect can provide it. Do not equate a large negative-keyword list with good query control; an old list can also block valuable demand.

For Performance Max, review asset groups, listing groups where relevant, conversion goals, product coverage, feed quality, audience signals, brand controls, and the account’s ability to distinguish incremental demand from existing brand demand. Asset groups should have a clear product, service, or audience rationale and usable creative coverage; splitting them arbitrarily can dilute data without improving control. A PMax campaign can participate in the same auctions as other campaign types, so look for overlap and interpret channel reporting carefully rather than assuming every reported conversion is incremental.

• Check whether the account has a clear separation between branded and nonbranded demand when the business needs that distinction for reporting or budget decisions.
• Review campaign budgets against business priorities. A campaign with high impression share lost to budget may be constrained, but increasing its budget is not automatically profitable.
• Inspect location targeting options and exclusions against the actual service area, delivery area, or store footprint.
• Check whether search partners, display expansion, or other network settings match the campaign’s objective and the client’s tolerance for placement-level visibility.
• Review ad assets and landing pages for policy status, message consistency, mobile usability, offer availability, and a clear next action.
• For ecommerce, verify that product titles, identifiers, availability, pricing, shipping, and category coverage are current enough to support the intended campaign strategy.

Use the PMax Cannibalization Checker when the account has overlapping Performance Max and Search coverage that merits a more focused diagnostic. It can help frame the investigation, but it should not replace review of the account’s query, conversion, brand, and business context. Likewise, use Lost IS Calculator to make impression-share loss easier to interpret; lost share is a diagnostic signal, not proof that the account should spend more.

5. Evaluate spend patterns, unit economics, and the right audit depth

Assess performance against the client’s economics, not a generic platform benchmark. Establish the maximum affordable CPA from contribution margin, close rate, repeat purchase behavior, and retention where available. For lead generation, translate the allowable cost per sale into an allowable cost per qualified lead: multiply the allowable acquisition cost by the qualified-lead-to-sale rate, then adjust for costs outside media if the client expects the agency to manage against a full-funnel target. For ecommerce, compare the target ROAS with margin and fulfillment economics; revenue ROAS alone can conceal unprofitable orders.

Separate blended performance from marginal opportunity

Blended CPA and ROAS describe the account’s combined reported outcome. They do not show what the next dollar of spend will produce. Review performance by campaign, product or service, geography, device, and relevant customer type where data volume permits. Look for budget-limited campaigns that meet the client’s threshold, campaigns consuming spend without qualified outcomes, and segments where apparent efficiency depends on brand demand or returning customers. Evaluate impression share lost to budget and lost to rank separately: budget loss indicates a budget constraint, while rank loss may point to relevance, bid, competition, or quality factors.

Use pacing as an operational check rather than a performance forecast. A simple projected monthly spend is month-to-date spend divided by elapsed days, multiplied by total days in the month. Compare that projection with the approved monthly cap, seasonality, and planned promotions. At the campaign level, compare actual spend with the intended budget share, then account for Google’s daily spending behavior and any budget changes. Flag a projected overspend when the forecast exceeds the client’s approved cap; do not assume that a smooth daily pace is necessary if demand varies by day or promotion.

These tiers are planning guides, not rigid rules. A $5,000 account can be technically complex if it spans many locations or relies on offline conversion imports. A $200,000 account can have a simple campaign structure but high governance risk. Use the Google Ads Waste Calculator to frame a discussion about potentially unproductive spend, then validate each candidate against search terms, conversion quality, and business context before describing it as recoverable waste.

6. Score findings by business impact, confidence, and work required

An audit becomes useful to a sales and delivery team when findings are prioritized consistently. Give each issue an impact score, confidence level, urgency, owner, and estimated effort. Impact asks how much the issue could affect business outcomes or account control. Confidence reflects the quality of supporting evidence. Urgency considers spend exposure, policy risk, data loss, and proximity to a launch or promotion. Effort includes analysis, client dependencies, implementation, testing, and reporting—not just the number of clicks needed to change a setting.

Use a transparent triage method

A practical approach is to classify findings as critical, high, medium, or low. Critical findings block responsible performance management, such as a primary conversion action that clearly records the wrong event or a billing or access constraint that prevents operation. High findings can materially distort efficiency or expose spend, such as broad irrelevant query coverage at meaningful cost. Medium findings affect reporting quality or optimization options but do not prevent an initial engagement. Low findings are improvements with limited near-term impact. Revisit the ranking with the prospect because their margin, sales capacity, and strategic priorities can change what matters most.

Estimate work in deliverables rather than vague labels like account cleanup. For example, specify conversion-action mapping, feed diagnostics, campaign restructuring, negative-keyword governance, landing-page recommendations, or a monthly pacing and performance review. Identify dependencies that belong to the client, such as CRM field access, margin data, legal approval, website releases, or creative assets. If the work cannot be estimated yet, state the discovery task and the condition that will trigger a revised scope.

7. Turn the audit into a defensible agency sales recommendation

The Google Ads agency sales process should convert findings into a decision the prospect can understand. Start with the client’s business objective, summarize what the evidence supports, explain the principal risks, and show how the proposed engagement addresses them in sequence. Keep the distinction clear between observed facts, likely causes, recommendations, and hypotheses that require testing. This makes the proposal more credible and gives the delivery team a usable baseline after the account is transferred.

Structure the readout around decisions

• State whether the account is ready for ongoing optimization, needs measurement repair first, or requires a broader technical discovery phase.
• Show the current baseline with the reporting window, conversion lag, included actions, and any known data limitations.
• Present the three to five highest-priority findings with evidence, business consequence, confidence, proposed owner, and estimated timing.
• Separate agency work from client dependencies and media budget decisions.
• Define what success will mean during the first 30, 60, and 90 days, including operational milestones when performance outcomes need more time to mature.
• Document what the agency will not guarantee, especially outcomes dependent on demand, competition, sales follow-up, inventory, pricing, or unverified tracking.

Use a go, conditional go, or no-go recommendation. A go is appropriate when access, measurement, economics, and operational ownership are sufficient for the proposed service. A conditional go means the contract should include named prerequisites, such as tracking validation, a CRM data feed, additional discovery, or a client-approved budget. A no-go is reasonable when the prospect expects guaranteed results from unreliable data, refuses necessary access, has no capacity to handle leads, or wants an agency to accept responsibility for decisions it cannot control.

For ongoing change execution, PPC Tuner positions itself as a Gemini 3.8 AI human-in-the-loop alternative: it can support analysis and stage mutate operations for human review and approval rather than treating automated recommendations as self-authorizing changes. Reviews and approvals occur inside PPC Tuner’s secure web application workspace. The agency remains responsible for checking the evidence, confirming the client’s objectives, and approving changes that fit the agreed scope.

8. Preserve the audit baseline through takeover and the first 90 days

Once the client signs, convert the pre-sale findings into a controlled transition plan. Confirm account ownership and access, billing arrangements, user permissions, linked properties, conversion definitions, and the client’s change-approval process. Record the last known settings and performance baseline before major edits. This is particularly important when another provider is still making changes or when automated bidding has recently been reset, restructured, or moved to a new goal.

Sequence changes so the team can learn what worked

During the first two weeks, prioritize access, tracking checks, budget controls, business-goal confirmation, and urgent policy or delivery issues. Make only necessary containment changes immediately, and document why they cannot wait. In weeks three and four, complete agreed measurement work, stabilize reporting, and propose a prioritized optimization plan. During days 30–60, run controlled tests where data and business conditions support them. During days 60–90, assess mature outcomes against the original baseline, account for conversion lag, and revise targets using observed lead quality, sales, or order economics.

Avoid changing tracking, bidding strategy, campaign structure, targets, and budgets all at once unless a serious issue requires it. When feasible, isolate important changes, note the hypothesis and success criterion in advance, and allow enough time for conversion lag and bidding adaptation before calling a test. If the client changes prices, promotions, landing pages, CRM stages, or fulfillment capacity during the same period, include those events in the interpretation instead of attributing every result to the ad account.

• Carry forward the audit’s confidence labels, unresolved questions, and client dependencies.
• Keep a change log with date, reason, expected effect, approval, and follow-up measurement.
• Review pacing against the approved budget and explain any projection that approaches or exceeds the cap.
• Report platform conversions separately from qualified leads, closed sales, revenue, and margin whenever the available data supports those distinctions.
• Re-baseline only when there is a documented reason, such as corrected tracking, a business-model change, or a materially different conversion goal.