PPC optimization checklist for a 30-day account review

Complete a controlled conversion through the real landing page journey. Check mobile forms, confirmation pages, calls, and checkout paths that matter to the business. Confirm that the intended conversion action receives the event. A button click should not silently replace a completed submission as the bidding goal.

Check primary and secondary conversion actions. Make sure a native purchase event and an imported event do not both count the same order toward bidding. For lead generation, verify whether the counting setting reflects one lead per ad interaction. For purchases, confirm transaction identifiers, currency, and values. Repeated confirmation page loads should not create repeated orders.

• Compare completed forms with accepted leads in the business system.
• Check call duration and answered-call criteria before treating calls as qualified leads.
• Review offline conversion import failures and the delay between inquiry and qualification.
• Separate real revenue from assigned lead values in every export.
• Document a tracking change as a break in comparability.

Investigate a sudden difference rather than expecting platform totals and business records to match perfectly. Attribution, consent, reporting dates, and conversion lag explain some differences. If the gap appears exactly when a tag or form changes, repair measurement before reducing bids. A weekly account audit helps maintain the recurring checks between monthly reviews.

Calculate CPA as ad spend divided by conversions. Calculate qualified CPA using qualified conversions in the denominator. Calculate ROAS as attributed conversion value divided by ad spend. If spend is $1,200 and qualified leads total 12, qualified CPA is $100. That arithmetic says nothing about profitability until you know how those leads convert to sales.

For a simple lead model, allowable lead CPA equals allowable acquisition cost multiplied by lead-to-sale rate. If allowable acquisition cost is $600 and the close rate is 20%, allowable lead CPA is $120. Use a close rate from comparable, mature leads. Do not borrow a sales team's best month to justify a higher bid target.

For ecommerce, a simplified break-even revenue ROAS is 1 divided by contribution margin before advertising. A 40% margin gives 2.5. Include returns, payment fees, shipping subsidies, and other variable costs in that margin. A revenue target that ignores those costs can reward unprofitable sales.

Use two times target CPA without conversions as an investigation trigger, not an automatic verdict. With small samples, relevant terms can still have no conversions. Separate confirmed irrelevance from uncertain performance. Avoid adding the same dollar to several overlapping waste categories.

Sort search terms by cost and read the highest-cost queries first. Label each as relevant, irrelevant, or uncertain. A relevant query with poor results needs a different response from a job seeker clicking an emergency plumbing ad. Where query reporting is incomplete, state that your findings cover visible terms rather than all account spend.

Group repeated words and phrases into n-grams. Then inspect the full queries behind each group. A word associated with research can also occur in purchase intent. For example, blocking "free" can remove searches for a paid service offering a free consultation. Exclude a clearly unwanted query with an exact negative before expanding the exclusion's scope.

• Check expensive terms with no mature conversions.
• Review terms with CPA above the target and enough history to judge.
• Look for employment, training, unsupported services, and areas outside coverage.
• Review negative conflicts with priority products or services.
• Verify shared-list assignments before applying account-wide exclusions.

Use the industry negative keyword guide to test intent categories against your offer. The negative keyword autopilot handles n-gram mining, exact or phrase negatives, and shared lists. Keep ambiguous commercial terms in Approve mode while you establish their value.

Open the actual destination for each high-spend ad group on a phone. Confirm that the headline answers the search, the advertised service is available, and the primary action works. Inspect price expectations, service boundaries, shipping terms, and appointment availability. A query can be relevant while the offer is a poor fit.

Review disapprovals, broken final URLs, asset coverage, and mismatched claims before rewriting everything. Use click-through rate to diagnose relevance within comparable positions and query groups. A higher CTR does not prove better business performance if the new message attracts less qualified traffic.

• Keep the query, ad promise, landing headline, and conversion action consistent.
• Test form validation and confirmation on mobile and desktop.
• Check that call actions work during staffed hours.
• Review responsive search ad asset signals alongside campaign outcomes.
• Define the success metric and observation window before launching a test.

Change one major hypothesis at a time. If you change the offer, landing page, and bidding target together, attribution becomes difficult. For sparse campaigns, prioritize obvious friction fixes and record them as operational corrections. Do not declare a winning ad because it receives a few early conversions.

Review search impression share lost to budget alongside search impression share lost to rank. Budget loss describes estimated eligible impressions missed because budget is insufficient. Rank loss points toward Ad Rank constraints. These percentages are not measures of wasted spend, and they do not establish that additional traffic will convert profitably.

For a review trigger, investigate campaigns with more than 10% search impression share lost to budget and mature CPA within target. That is a chosen operating threshold, not an industry benchmark. Confirm lead quality, available capacity, and marginal performance before assigning more money. Do not move funding away from a reliable campaign merely because another campaign underspends.

Calculate remaining daily allowance as approved monthly budget minus month-to-date spend, divided by remaining days. If $3,000 remains with ten days left, the planning allowance is $300 per day. Actual daily delivery can vary, so monitor cumulative spend and platform budget behavior rather than treating the allowance as a guaranteed daily ceiling.

• Check bid strategy status and recent target changes.
• Keep shared budget and portfolio relationships visible.
• Inspect device, geography, and schedule differences without assuming every segment supports manual adjustments.
• Set a daily change cap and a clear rollback condition.
• Reassess after an appropriate conversion cycle rather than reacting every morning.

Use bid and budget rules to connect targets with pacing and daily change caps. With Smart Bidding, choose supported strategy-level controls instead of issuing conflicting manual keyword changes.

Treat the 30-day period as an implementation window, not permission to make every change immediately. Repair broken measurement and destinations first. Then address confirmed irrelevant traffic. Only after those checks should you evaluate strategy targets, creative tests, and reallocations. An account with a tracking fault may need more time before efficiency decisions become reliable.

Prioritize each action by avoidable cost, confidence, reversibility, and implementation effort. Fixing an invalid destination outranks testing a minor headline variation. An uncertain high-cost query may deserve a controlled test rather than an immediate permanent exclusion.

• Assign an owner and due date to every approved action.
• State the expected mechanism rather than promising a numerical uplift.
• Define the metric that could disprove the hypothesis.
• Record concurrent promotions and website changes.
• Schedule the next review after conversions have time to mature.

Close the month with a short explanation of what changed and what remains unresolved. The reporting guide shows how to turn that decision log into an executive update without burying the next action in platform detail.