The Blended Metrics Lie
Google Ads records revenue at the moment of purchase. The £150 order fires a conversion. Done. But that order's journey isn't over.
Three days later, customer service issues a £30 credit because shipping took too long. A week later, the customer returns one item for £45. Two weeks later, they price-match with a competitor for another £15 off.
The Revenue Reality
Google Ads shows £150 revenue. Finance received £60. Your ROAS is inflated by 150%. Every decision you make from this data is wrong.
Types of Post-Sale Discounting
Goodwill Credits
Customer service compensation for delays, quality issues, or general dissatisfaction. Often 10-30% of order value. Invisible to advertising platforms.
Partial Returns
Customer keeps some items, returns others. The original conversion remains at full value while actual revenue decreases.
Price Matching
Competitor offers a lower price post-purchase. You refund the difference to retain the customer. Original conversion value unchanged.
Damage Compensation
Shipping damage results in partial refunds rather than full returns. Customer keeps the item at reduced price.
Loyalty Adjustments
Post-purchase loyalty rewards, cashback, or points redemption. Reduces effective revenue but appears after conversion.
How Discounts Destroy Metrics
Let's trace a month's campaign performance with and without post-sale adjustment visibility:
What Google Ads Reports
Ad Spend
£20,000
Reported Revenue
£100,000
Reported ROAS
5.0x
Status
Scale aggressively
What Finance Sees
Gross Revenue
£100,000
Full Returns
-£12,000
Partial Returns
-£4,000
Goodwill Credits
-£3,500
Price Matches
-£1,500
Net Revenue
£79,000
Actual ROAS
3.95x
A 21% gap between reported and actual ROAS. At 5.0x reported, you might aggressively scale. At 3.95x actual, you might optimise before scaling. Different data, different decisions.
Training Algorithms on False Signals
The bigger problem: Smart Bidding learns from your conversion values. If you feed it gross revenue, it optimises toward customers who generate high gross transactions-not high net transactions.
Some customer segments consistently require more post-sale adjustments. First-time buyers, discount-acquired customers, certain product categories. Smart Bidding can't learn to avoid these if it never sees the adjustment data.
The result: your algorithm scales toward customers who look valuable at purchase but require significant post-sale handling. You're optimising for problems.
How to Measure True Impact
1. Track Adjustment Rate by Source
Link adjustment events back to original acquisition source via UTM parameters. Calculate adjustment rate (adjustments / orders) and adjustment value (total adjustments / total revenue) by channel.
2. Build Cohort Analysis
Track monthly acquisition cohorts through 90-day adjustment windows. Compare final net revenue to initial gross revenue. Identify which campaigns produce "leaky" cohorts.
3. Calculate Effective ROAS
Effective ROAS = (Gross Revenue - Returns - Adjustments) / Ad Spend. Report this alongside Google's ROAS to show the gap.
4. Segment by Adjustment Type
Different adjustment types have different causes. Returns indicate product fit issues. Goodwill credits indicate service issues. Price matches indicate competitive pressure. Each requires different solutions.
Mitigation Strategies
Import Adjusted Conversion Values
Use Google Ads conversion adjustments to update transaction values when adjustments occur. A £100 order that becomes £80 after credits should train Smart Bidding on £80.
Apply Discount Rate Buffers
If you can't import adjustments, increase ROAS targets by your average adjustment rate. 8% adjustment rate means 8% higher ROAS targets.
Segment High-Adjustment Products
Some products consistently need adjustments (complex items, size-variable apparel). Use custom labels to bid differently on these SKUs.
Fix Upstream Problems
High adjustment rates signal underlying issues. Poor product descriptions cause returns. Slow shipping causes goodwill credits. Address the cause, not just the measurement.
Post-sale adjustments are the gap between marketing's version of revenue and finance's version. Close that gap in your data, and your decisions improve automatically.
Frequently Asked Questions
How do post-sale discounts affect Google Ads reporting?
Post-sale discounts (refunds, goodwill credits, price adjustments) reduce actual revenue but don't update Google Ads conversion values. If you report £100 at purchase but later issue a £20 goodwill credit, Google still shows £100 revenue. This inflates your ROAS and trains Smart Bidding on phantom revenue.
Should I import adjusted conversion values after discounts?
Yes, if your post-sale discount rate is significant (over 5% of revenue). Use offline conversion adjustments to update transaction values when discounts occur. This trains Smart Bidding on actual received revenue rather than gross transaction value.
How do I account for variable discount rates by product?
Track discount rates by product category or SKU. Some products consistently require more post-sale adjustments (complex items, size-variable apparel). Apply higher ROAS targets to high-discount products, or use custom labels to segment campaigns and bid differently.
Related Reading
Stay informed
Get our latest insights on Google Ads strategy delivered to your inbox. No fluff, no spam, just honest thinking.
Unsubscribe anytime. No spam, ever.
Want to understand your true campaign profitability?
We'll help you build measurement frameworks that account for post-sale adjustments and show actual contribution margin.
Book a Discovery Call