On this page
Merchandising and CX teams get excited about fit technology. Finance does not sign off on excitement — it signs off on a model with inputs, assumptions, and a payback date. If your business case for a size recommendation tool reads like a product pitch instead of a P&L line, it will stall in the CFO's inbox. Here is the model that gets approved.
Why Finance Rejects Most Fit-Tech Proposals
Most pitches for size recommendation tools lead with customer experience metrics — conversion lift, satisfaction scores, reduced support tickets. Finance cares about those only insofar as they translate into cash. A CFO wants three numbers: the return-rate delta, the contribution margin recovered, and the time to payback. Everything else is supporting detail.
The uncomfortable truth is that returns are one of the largest hidden cost centers in apparel retail. Reverse logistics — shipping, restocking, inspection, markdown, and write-off — routinely consumes a fifth to a third of apparel revenue, a range confirmed by reverse-logistics research from the National Retail Federation. Size and fit issues are consistently cited as the single largest driver of apparel returns, ahead of damage, changed-mind, or delivery problems. That is the lever a size recommender pulls, and it is the lever finance actually cares about.
Building the Core ROI Formula
The base model for proving the ROI of a size recommendation tool needs four inputs your finance team will already trust because they come from existing systems, not vendor claims.
The Four Required Inputs
- Baseline size-related return rate — pull this from your returns management system, filtered to reason codes like "too small," "too large," or "doesn't fit."
- Average order value (AOV) for the segment where the tool is deployed.
- Fully loaded cost per return — shipping both ways, restocking labor, inspection, and markdown/write-off probability.
- Contribution margin per unit, not gross revenue — finance will immediately discount any model that uses top-line revenue instead of margin.
The formula finance wants to see is simple by design:
Recovered contribution margin = (baseline return rate minus post-adoption return rate) times order volume times AOV times contribution margin percentage, minus the fully loaded cost of returns avoided is added back as pure savings, not margin — the two lines stay separate because finance audits them separately.
A Worked Example
| Input | Value |
|---|---|
| Monthly apparel orders | 40,000 |
| Baseline size-related return rate | 22% |
| Post-adoption return rate | 14% |
| AOV | $85 |
| Contribution margin | 42% |
| Fully loaded cost per return | $18 |
Returns avoided per month: 8% of 40,000 = 3,200 orders. Contribution margin recovered: 3,200 times $85 times 42% = roughly $114,240 per month. Return-processing cost avoided: 3,200 times $18 = $57,600 per month. Combined monthly benefit: approximately $171,840, before subtracting the tool's subscription and implementation cost.
Translating Return-Rate Delta Into Payback Period
Payback period is the single number that moves a proposal from "interesting" to "approved." Divide the total implementation and first-year subscription cost by the combined monthly benefit calculated above. In the worked example, a typical mid-market implementation cost of $25,000 to $40,000 yields a payback window of roughly seven to eight weeks — a timeframe most finance teams treat as effectively risk-free, especially against a 12-month contract term.
This is also the calculation that should anchor an internal business case for fit technology: lead with payback weeks, not feature lists. Finance approves speed to breakeven far more readily than it approves promised experience improvements.
Sensitivity Matters More Than Precision
Finance will stress-test your assumptions, so build the sensitivity into the model yourself rather than waiting to be asked. Show the payback period at conservative (return-rate improvement of 3 points), expected (8 points), and optimistic (13 points) scenarios. A model that still pays back within a quarter under the conservative case is far more persuasive than one that only works in the best-case scenario.
Connecting the Model to the True Cost of Returns
Finance teams often underestimate the true cost of returns because most P&L reporting buries it inside logistics and cost-of-goods-sold lines rather than surfacing it as its own category. Walking through the true cost of a return for fashion brands alongside your ROI model helps finance see the baseline they are actually comparing against — not just shipping cost, but restocking labor, inventory carrying cost during transit, and the markdown probability on returned inventory that cannot be resold at full price.
Once that baseline is visible, the recommender's ROI stops looking like a discretionary CX investment and starts looking like what it is: a direct reduction to a cost line finance already tracks.
What to Put on One Slide
CFOs skim. Reduce the entire model to a single slide with five rows: baseline return rate, projected return rate, monthly contribution margin recovered, implementation cost, and payback period in weeks. Everything else — methodology, data sources, sensitivity ranges — belongs in an appendix that finance can request but does not need to read to approve the line item.
Frequently Asked Questions
What data does finance need before approving a size recommendation tool?
Finance needs a baseline size-related return rate pulled from actual returns data, not estimates, along with average order value and contribution margin for the relevant product segment. They also want a fully loaded cost per return that includes shipping, labor, and markdown risk, since that number determines how large the avoided-cost line actually is.
How long does payback typically take for a size recommendation tool?
Most mid-market apparel retailers see payback within six to twelve weeks when the tool reduces size-related returns by six to ten percentage points. Payback period depends heavily on order volume and AOV, so a smaller catalog with lower AOV will take proportionally longer even at the same percentage-point improvement.
Should the ROI model use revenue or contribution margin?
Always use contribution margin, never gross revenue. Finance teams discount models built on revenue because revenue ignores the cost of goods, fulfillment, and the return itself, and a model that overstates benefit by using revenue will lose credibility the moment finance recalculates it.
How is size recommendation ROI different from general returns-reduction ROI?
Size recommendation ROI isolates the subset of returns driven by fit rather than all return reasons, which matters because fit-related returns are the only category a recommender directly influences. Blending in returns caused by damage, delivery delays, or changed minds inflates the apparent impact and makes the model harder to defend when finance checks it against reason-code data.