The 5 Places D2C Marketing Budgets Quietly Leak
Most D2C marketing budgets don't lose money in one dramatic event. They lose it in five ordinary, unremarkable places that never show up as a single line item - because each one looks fine from inside its own dashboard.
The platform-reported ROAS gap
Every ad platform has an incentive to take credit for revenue it didn't fully cause. We measured this directly on ₹3.91 crore of Meta spend: platform-attributed ROAS and blended true ROAS told different stories, with a gap between 2.76x and 0.71x depending on which number you trusted. If your weekly budget decisions run off platform dashboards alone, you're optimizing against a number that's structurally flattered.
Spend that keeps flowing into a stockout
An adset doesn't know when its product runs out. We documented a real case: ₹6.2 lakh spent over eight days on traffic sent to a SKU that had already gone out of stock on day three. Nobody was negligent - the ad team optimizes for ROAS, the supply team tracks purchase orders, and neither system talks to the other fast enough to catch the five days in between.
Marketplace spend that's really just organic demand in disguise
Marketplace ad spend often gets credit for sales that would have happened anyway through organic search and brand recall on that same marketplace. Without separating true incremental lift from spend that's riding existing demand, brands can scale marketplace ad budgets well past the point where they're adding anything.
The best-seller that isn't the money-maker
Rate of sale and contribution margin are not the same ranking. A SKU can move the most units and still be a mediocre use of marketing budget once its actual margin, return rate, and discounting are accounted for. Teams that scale spend by unit velocity alone are often scaling their least profitable products the hardest.
Re-answering a question the business already answered
We call this the decision memory leak. A new hire, a new agency, or just a few months of turnover, and a team re-runs a test that was already conclusively answered - burning real budget to relearn something that was sitting in a Slack thread nobody could find. We documented one case at ₹8 lakh for a single re-run test. Multiply that across a year of team changes and it's one of the larger, least visible leaks on this list.
What connects all five
None of these show up as a fraud, a mistake, or a bad campaign. Each one is a case where a decision was made with real but incomplete information, by a team doing its actual job well within its own system. The leak isn't in any single team's execution - it's in the absence of a system that connects spend, SKU, supply, margin, and decision history before the next budget call gets made.
That's specifically what we built Lift and Vantage to close: Lift for the true-return question, Vantage for the root-cause question, both working off the same connected data instead of five separate dashboards that each see one piece of the picture.