WhyCashCycleBeatsMargin:AWorkingCapitalFrameworkforGoogleShopping

Most Google Shopping accounts are not even optimised against margin. The default segmentation in the wild is brand or product type, structures inherited from the merchant feed rather than designed around any commercial variable. The accounts that have moved past that usually land on margin tiering next. Both are upgrades on running everything in one campaign, and both stop short of the variable that actually governs ecommerce health: cash cycle. This article sets out the framework.

A quick taxonomy of what is out there. Tier zero is the single-campaign account, one PMax or one Shopping campaign covering the whole catalogue with a blended ROAS target. Tier one segments by brand or product type, which is really just mirroring the feed taxonomy and tells the algorithm nothing about commercial priority. Tier two segments by contribution margin, which is a genuine commercial signal but a partial one. Tier three, where this article lives, segments by return on working capital, treating each SKU as a financial instrument with a holding cost.

Margin is a profit-per-unit number. It tells you how much you keep on each sale. What it does not tell you is how often that sale happens, how much capital is tied up waiting for it, and what that capital is doing in the meantime. For a finance director, those questions matter more than margin in isolation. For a Google Ads account that controls a meaningful share of inventory movement, they should matter just as much.

Below: why stock is a depreciating asset most ad accounts treat as static, how to calculate cash velocity properly, how to build a campaign architecture around velocity rather than contribution margin, and the monthly review process that keeps the structure honest.

Stock Is a Depreciating Asset

Inventory in an ecommerce business is not a stable store of value. Every day a unit sits in your warehouse, its real economic value declines. There are six mechanisms at work.

The compound effect is that the carrying cost of slow-moving stock is significantly higher than most operators model in their P&L. A unit sitting on the shelf for nine months has absorbed storage cost, capital cost, and seasonal decay long before any markdown has been applied. By the time it sells, realised contribution can be a fraction of what the original margin calculation suggested.

This is the gap that margin-only frameworks ignore. Margin is a snapshot at the moment of sale. It says nothing about the cost of getting to that moment.

What This Means for Google Shopping

If you accept that stock depreciates and time-to-sell is a material variable in true contribution, the implications for paid Shopping are direct.

A 12% margin product that sells every two weeks recycles capital twenty-six times a year. Each cycle releases the original cash plus a small profit, redeployable into more inventory, more advertising, or operating expenses. The capital is working.

A 45% margin product that sells once every nine months recycles capital 1.3 times a year. Each cycle releases the original cash and a larger profit, but the absolute cash generated per pound of working capital deployed is significantly lower. The capital is largely sitting still.

A margin-only Google Shopping framework will look at these two products and tell you to bid aggressively on the 45% margin product and exclude the 12% margin product. A cash-cycle framework will tell you the opposite.

This is not an edge case. It is the default situation in any inventory-based ecommerce business with a mix of bestsellers, mid-range stock, and slow-moving lines. The "high margin" products in most catalogues are not the cash generators. The volume-velocity products are. And the SKUs that need the most aggressive paid Shopping support are often those with the worst velocity, because that is where the working capital is trapped.

Defining the Metric: Return on Inventory Capital

To structure a paid Shopping account around cash cycle, you need a single metric that captures both margin and velocity. Margin alone will not do it. Inventory turnover alone will not do it. The combination, return on inventory capital, does.

This is structurally the same metric as Gross Margin Return on Inventory Investment (GMROI), well-established in retail finance but largely absent from paid media management. It answers the only question that matters for a working capital allocation decision: how much profit does this product generate per pound of working capital tied up in it, per year?

A margin-only ranking puts Product B at the top: 55% is the highest in the catalogue. A cash-cycle ranking puts Product C at the top, narrowly ahead of A. Product B, despite the headline margin, generates roughly a quarter of the working capital return that the other two do. Same catalogue, very different account structures, very different cash outcomes.

Replacing Margin Tiers With Velocity Tiers

The conventional margin-tiered structure assigns each product to a tier based on contribution margin. The cash-cycle alternative assigns each product to a tier based on a composite of margin and velocity, with stock age treated as a separate dimension.

The Maths of Clearing Aged Stock

A fashion brand with £200,000 of off-season stock. Original cost £80,000 (40% landed cost ratio). At full price the stock would have delivered roughly £80,000 of gross profit. It did not sell at full price. Three paths.

A margin-only framework counsels against Path B, because the paid Shopping is running at zero or negative contribution. A cash-cycle framework counsels for Path B, because the alternative is not an alternative: the capital trapped in that stock is otherwise lost, and paid Shopping is the most controllable lever available to release it.

The deliberate-loss decision is not a marketing decision. It is a working capital decision. The Google Ads account is the execution layer; the strategic call sits with finance and operations. Most agencies do not have these conversations because they are optimising for the wrong scoreboard.

Building the Operating Framework

A cash-cycle Shopping account is not just a different feed structure. It requires an operating framework that connects three things usually sitting in separate parts of the business: paid media, inventory data, and finance.

Core data inputs:

• Per-SKU cost data, landed where possible, inclusive of inbound logistics and duty.
• Per-SKU velocity on a rolling 13-week window; longer for slower lines.
• Per-SKU stock age, days since first receipt for current holding.
• Per-SKU return rate. A 30% return rate effectively cuts headline margin by the same amount.
• Per-category capital cost, a defensible weighted-average figure (15%, 18%) applied per SKU.

Mechanics: a Google Sheet or BigQuery table maintained by finance or ops, one row per SKU, refreshed monthly. The paid media team consumes it as a feed input, typically via a supplemental feed in Google Merchant Center, writing custom labels from velocity tier, stock age band and GMROI band.

Labels are applied mechanically by the supplemental feed. Campaign structure then segments by velocity tier with overlays for stock age, allowing aged stock inside Tier 1 or Tier 2 to receive distinct bid treatment.

Campaign Architecture in Standard Shopping

Five concurrent campaigns with distinct economic objectives, each receiving the bid strategy that matches its commercial role.

Applying the Framework to Performance Max

PMax constrains the depth of structural control. Listing-group exclusions and asset-group segmentation are the available tools; campaign-level bid strategy applies to everything in the campaign. The implication is that running all products in one PMax campaign with a blended ROAS target is the worst possible outcome for a cash-cycle account, because the algorithm optimises across products with fundamentally different working capital roles using a single signal.

1. PMax 1, High-Yield. Tier 1 only, filtered via listing group. Asset groups by category. Aggressive Target ROAS. Brand exclusions at campaign level.
2. PMax 2, Quality Margin Controlled. Tier 2. Tighter ROAS target. Separate campaign so the algorithm does not blend Tier 1 and Tier 2 economics under one learning signal.
3. PMax 3, Volume. Tier 3. Target CPA or low ROAS target with daily budget cap. Often the most volatile and the one requiring closest weekly monitoring.
4. PMax 4, Clearance. Tier 4 aged stock. Maximise Conversion Value during active windows. Budget set against working capital release. Closed between windows; reopened when aged-stock band justifies it.

Products outside the framework (new arrivals without velocity history, supply-constrained items, items with active disputes) are excluded at listing-group level until they have enough data to be tier-assigned.

Setting Targets Correctly

Two methods, used in combination.

Method 1, Break-even ROAS by tier × aggressiveness factor:

• Tier 1 (high yield): break-even × 1.4 to 1.8
• Tier 2 (quality margin): break-even × 1.1 to 1.3
• Tier 3 (volume): break-even × 1.0 to 1.1 (often as CPA target)
• Tier 4 (clearance): break-even × 0.7 to 1.0, deliberately at or below break-even during active clearance

Method 2, Target return on working capital deployed:

Define a minimum acceptable annualised return on capital tied up in each tier's inventory. Back-calculate the ROAS or CPA that achieves it at the tier's current velocity. This produces a target that explicitly reflects capital cost, not just unit economics.

Method 1 is the starting point. Method 2 is the refinement. Where they diverge significantly, the underlying tier definition is usually off. Targets are reviewed monthly. Quarterly is too slow for any business with meaningful inventory turnover.

The Monthly Review Process

A structured monthly review integrates inventory, paid media and finance data, answering three questions for each tier.

The review output is a short document (one page, sometimes two) capturing tier answers, recommended target adjustments and structural changes for next month. Shared with finance and ops, not held within the marketing team.

Integrating With Finance

The framework only works if finance is in the loop. The reason most paid media management does not operate at this level is not that the techniques are hard; it is that data flow between finance and marketing is broken. Cost data sits in one system, sales in another, inventory in a third, ad spend in a fourth, and nothing reconciles cleanly.

Minimum integration: monthly data exchange (per-SKU cost, return rates, stock age), quarterly review of the cost-of-capital assumption, annual review of tier thresholds.

For larger operations, a shared BigQuery or warehouse table with refresh logic eliminates the manual exchange. The technical work is a week. The organisational work, getting finance to treat the paid media account as a working capital lever rather than a cost line, is harder and longer.

The end state: a paid media account optimised against the same scoreboard finance uses for inventory health. Return on capital deployed, working capital recovery rate, aged-stock proportion. Google Ads spend becomes a working capital management tool, and the campaign structure reflects that.

What This Framework Changes in Practice

For an operator on a margin-only Shopping account today, the move changes four observable things.

1. The Tier 4 conversation reverses. Products previously excluded as low margin become high-priority clearance candidates, with deliberate paid investment during clearance windows.
2. The Tier 2 conversation tightens. Headline-margin products that were over-invested get pulled back to measured targets. This typically releases meaningful budget previously absorbed by slow-moving stock.
3. The Tier 3 conversation expands. Thin-margin, high-velocity products previously starved get more investment. Often the largest single source of incremental performance.
4. The reporting language changes. The founder or CFO conversation is no longer about blended ROAS hitting target. It is Return on Inventory Capital by tier, aged-stock clearance velocity, and working capital deployed in each band. Marketing starts speaking the same language as finance.

Where to Start

Don't rebuild everything at once.

1. Start with the data. Per-SKU cost, velocity, stock age and return rate flowing into a single table. Allow four to six weeks; it always takes longer than expected because the data lives in places nobody has had to reconcile before.
2. Build tier classification. Apply to the catalogue and review with finance. Are the cash cows in Tier 1? Are the aged-stock candidates in Tier 4? If the output looks wrong, input data is usually the issue.
3. Introduce the campaign structure. Tier 1 first, as a separate campaign alongside the existing flat structure. Four weeks to develop learning. Tier 4 next, because working capital recovery is the most immediately visible improvement. Tier 2 and Tier 3 last.
4. Stand up the monthly review from day one. The framework's value compounds with discipline. An account that is restructured but not maintained drifts back to a flat structure within two or three months.

The end-state account is more complex and requires a closer working relationship between marketing, finance and operations than most businesses currently have. The trade-off is an account that manages working capital with the same discipline the finance team applies to the rest of the balance sheet, and produces materially better cash outcomes as a result.

Margin tiering was the start of the conversation. Cash cycle is the conversation worth having.