Lessons from Maddie Lightening: Fixing Misreported ROAS, Legacy Account Structures, and Using AI in PPC

The insights shared by Maddie Lightening, as highlighted in Anu Adegbola’s article on Search Engine Land, offer valuable guidance for advertisers managing Google Ads. As the article notes, “One of Maddie’s early mistakes involved misreporting performance due to account currency differences.” (Anu Adegbola, Search Engine Land).

Why precise data and account design matter

Maddie’s experience shows how currency mismatches in Google Ads accounts can significantly distort reported ROAS. When account currency settings do not align with the actual transaction currency, data misrepresents campaign effectiveness, leading to misguided budget decisions. This is particularly important because automated bidding systems depend on accurate conversion values to forecast and optimize performance. As Google’s documentation explains, “Google Ads predicts future conversions and associated values using your reported conversion values, which you report through conversion tracking.” (Google Ads Help).

Common problems and quick fixes

1. Misreported ROAS due to currency mismatches

Symptoms: Sudden unexplained changes in reported ROAS when comparing platform data to CRM figures; apparent underperformance or overperformance after moving between regions or billing accounts.

Fixes:

• Verify the currency setting in Google Ads and match it to your CRM or sales reporting currency.
• Cross-check conversion values by exporting platform conversions and CRM transactions for the same period; look for systematic scale differences.
• Use conversion value rules carefully and document any multipliers applied so reporting teams and automated bidding strategies use consistent values.

2. Legacy account structure that blocks automation

Symptoms: Thousands of tiny campaigns or ad groups, duplicated targeting, and poor signal consolidation that reduces automated bidding efficacy.

Fixes:

• Run an account audit focused on overlap, redundancy, and historical naming schemes.
• Consolidate campaigns where possible to increase signal density (more conversions per campaign improves Smart Bidding).
• Plan restructures during lower-demand windows; if delays are unavoidable, prepare rollback plans and staged rollouts to reduce risk.

3. AI and automated bidding: control without blocking performance

Symptoms: Spikes in CPCs, erratic spend, or bid behavior that exceeds acceptable thresholds after switching to automated strategies.

Fixes:

• Implement max CPC caps or bid limits within portfolio strategies to set safe boundaries for automated bidding.
• Monitor performance for 1–2 conversion cycles after changes; allow time for the algorithm to learn before judging results.
• Use conversion inclusion settings so Smart Bidding optimizes toward the right set of conversion actions.

How to restructure accounts without losing performance

Start with a hypothesis-driven audit: identify which campaigns are cannibalizing traffic, which ad groups have persistent low quality scores, and where conversion attribution is inconsistent. Create a migration plan that consolidates related campaigns, preserves historical learning where possible, and sets clear success metrics for each phase. When possible, use experiments or drafts to validate changes before a full-sweep migration.

AI prompts, human oversight, and testing discipline

AI delivers better outputs when given structured, detailed inputs. Provide explicit goals, target audiences, value ranges, and constraints in prompts or campaign settings. Treat AI-generated suggestions as draft ideas to be tested, not final decisions. Maddie’s guidance centers on curiosity and iteration—test, learn, and adjust.

Action checklist for advertisers

• Verify currency settings across Google Ads, analytics, and CRM systems; reconcile differences monthly.
• Audit account structure quarterly and prioritize consolidation where conversion signals are thin.
• Set max CPC or bid limits when moving to automated bidding; monitor and adjust after 1–2 conversion cycles.
• Use conversion value rules and include only the conversions you want Smart Bidding to optimize for.
• Design AI prompts with clear constraints and success metrics; run controlled experiments to validate outputs.

Conclusion

Maddie Lightening’s examples remind advertisers that technical details and account architecture matter as much as strategy. Fixing misreported ROAS, simplifying legacy structures, and using AI with guardrails will improve campaign performance and decision-making. For a deeper look at Maddie’s experience and quotes, see the original piece on Search Engine Land.

Read the original article: Maddie Lightening speaks on misreported ROAS, account structure chaos & AI mistakes — Search Engine Land (Anu Adegbola)