Turn Your Product Feed Into a Profit Engine

Your product feed is lying to your ad platform

You have 10,000 SKUs. Every single one lands in Google Merchant Center or Meta Commerce Manager looking identical. Same weight. Same priority. Same bid eligibility. The algorithm cannot tell the difference between a $200 jacket earning you 62% margin and a $9 phone case you lose money on after returns.

I bet 2,000 of those SKUs are eating your ad budget right now.

The flat feed problem

Three things go wrong when every SKU enters the auction with equal standing:

• All SKUs get equal budget. Google Shopping and Meta Advantage+ distribute impressions based on click probability, not profit. A $9 phone case that gets high CTR will eat through budget faster than a $200 jacket with lower CTR but 60% margin.
• High-return products waste money twice. You pay for the acquisition click. Then you eat the return shipping, restocking, and refund processing. A 35% return rate on a low-margin product means you’re paying to lose money.
• No repeat-purchase intelligence. The algorithm does not know that buyers of Product A come back 3x in the next year. It optimizes for first-order revenue, missing the 2.8x LTV multiplier hiding in your subscription and replenishment SKUs.

Five signals your feed is missing

1. Gross margin per SKU

Your ecommerce platform knows what you paid for every product. CustomerLabs pulls COGS (cost of goods sold) from Shopify, WooCommerce, Magento, BigCommerce, or a CSV upload. It calculates gross margin per SKU and writes the tier label into your feed daily. Products at 55% margin and products at 8% margin stop looking the same to the algorithm.

2. Return rate (last 90 days)

A $100 sale with a 38% return rate is a $62 sale before you subtract return shipping and restocking. CustomerLabs calculates rolling 90-day return rates per SKU from your order management system. Anything above 25% gets flagged and excluded from prospecting campaigns. It can still appear in retargeting — the customer already knows what they are buying.

3. Repeat purchase probability

Some products are gateway drugs. A $28 protein powder that gets reordered every 5 weeks is worth more than a $150 kitchen gadget bought once. CustomerLabs calculates repeat-purchase rates from your order history and assigns a multiplier. The algorithm finally sees lifetime value, not just first-order revenue.

4. Category-level LTV multiplier

Individual SKU data can be noisy for products with low order volume. CustomerLabs rolls up LTV data to the category level. If customers who first buy from your “Running Shoes” category have a 2.4x 12-month LTV compared to “Accessories” buyers, every SKU in that category gets a multiplier that feeds into bid strategy.

5. Inventory velocity

Sitting inventory costs money. CustomerLabs flags slow-movers (below 2 units per week) and fast-movers (above 50 per week). Slow-movers get bid boosts to clear stock. Fast-movers with high margin get aggressive bidding because you have the supply to fulfill the demand.

How enriched feeds change bidding

The left side is your feed today. Five products, five identical budget allocations. The right side is the same feed enriched with margin, LTV, and return data. The algorithm now knows the jacket is your best product, the protein powder has hidden LTV, and the phone case should get zero prospecting spend.

Product set strategy for Meta Advantage+

Tier A product sets: margin above 40%, return rate under 10%

These are your best SKUs. They make money on the first order, customers keep them, and many come back for more. In Meta Commerce Manager, create a product set filtered by the custom properties CustomerLabs populates. Give this set 60-70% of your Advantage+ catalog budget. Set no ROAS cap — let the algorithm bid aggressively.

Tier B product sets: margin 20-40%, standard bidding

Solid products that contribute to revenue without being stars. Give them 25-30% of budget. Use standard ROAS targets. These SKUs keep the catalog diverse for the algorithm while protecting your margin floor.

Excluded: return rate above 25% or margin under 10%

Pull these from prospecting entirely. If someone searches for the product or visits the product page, retargeting can serve them. But do not spend prospecting budget acquiring new customers through products that lose you money after returns and fulfillment.

Google Shopping custom labels

CustomerLabs pushes enrichment data into Merchant Center custom labels automatically. No manual CSV uploads. No spreadsheet mapping.

The setup

Week 1: Connect COGS and return data

CustomerLabs connects to your ecommerce platform and pulls cost-of-goods-sold for every SKU. It also pulls return data from your order management system or returns provider. Every SKU gets a margin calculation and a 90-day return rate by the end of the week.

Week 2: Build margin tiers and product sets

Margin tiers are assigned. Repeat-purchase rates are calculated from order history. Custom labels deploy to Google Merchant Center. Product set rules go live in Meta Commerce Manager. You review the tier assignments and adjust any manual overrides.

Week 3: Launch tiered campaigns

Separate asset groups and product sets go live. Exclusion rules activate for sub-floor margin and high-return SKUs. Labels update daily as COGS, return rates, and order data change. The algorithm starts relearning within 48 hours.

Rahul Menon, VP of Growth, Lumiverde Home: “We had 4,200 SKUs and no idea which ones were actually profitable after returns. CustomerLabs tagged every product with margin tiers in 6 days. We cut 380 SKUs from prospecting and ROAS jumped 31% in the first month — same budget, same creative.”

Results from enriched feeds

Sahil Kanojiya, Head of Mediabuying, Smars Jewellery: “After switching our campaign objective to new customer purchase using CustomerLabs signal engineering, we saw a 200% increase in new customers.”

The SKUs you should kill

The hardest part of feed optimization is not boosting your winners. It is cutting your losers. Every SKU below your margin floor that stays in prospecting steals budget from a product that would have made you money.

Run this audit quarterly: pull every SKU with margin under 15% and return rate over 20%. Count how many there are. Multiply by average CPC. That is the budget you are burning every month on products that will never be profitable through paid acquisition.

Sometimes the best signal is “do not bid on this at all.”