Pocket Price Waterfall: The Marn-Rosiello View of Where Your Money Actually Goes

The company G2N cascade gives you one pocket price. Useful, but blunt. Run the same cascade customer by customer and you get something more diagnostic: a band of pocket prices, all sitting under the same list price, all reaching the P&L through different paths.

This is the original framing in the 1992 Harvard Business Review article Managing Price, Gaining Profit by Michael V. Marn and Robert L. Rosiello, which introduced the pocket price waterfall to a general business audience. Their core insight: the company-level waterfall hides where the money actually goes. The customer-level waterfall shows it.

Why same list does not mean same pocket

Every account in your top 20 looks at the same list price on the same rate card. They negotiate against the same on-invoice structure. The on-invoice line of the cascade is therefore relatively uniform across accounts (channel discount plus customer status plus efficiency terms, all roughly comparable).

Below the invoice line, the picture changes. Performance rebates depend on what each customer signed for in the JBP. Promotional funding depends on what they ran in the calendar. Consumer promo depends on who took the multibuy money and who passed. By the time you reach pocket, two accounts that started identical have travelled different routes.

The pocket price band

Plot every customer's pocket price on a single horizontal axis and you get a distribution. The shape matters more than the average. A tight band means a disciplined trade-terms operation. A wide band means accumulated complexity and probably accumulated leakage. The shape itself becomes the negotiation tool, not just the numbers.

A 20-customer band on a EUR 36 list

Where the spread comes from

Walk the cascade twice (once for the worst customer, once for the best) and the gap shows up in three places.

• Performance rebates: the worst customer hit lower thresholds in the JBP and earned the higher rebate band.
• Promotional spend: the worst customer ran more promo weeks, deeper discounts, more display, more feature.
• Consumer promo participation: the worst customer took the multibuy money and the loyalty offer; the best customer ran neither.

The on-invoice line for the two accounts is usually within 50 to 100 basis points of each other. The off-invoice line is where the EUR 7.20 spread is built. That is the diagnostic. The leakage is structurally below the invoice, which is exactly the layer the commercial team has the least visibility into.

Three numbers describe a pocket price band. Compute them and you have the diagnostic on a single line.

Per-customer pocket price

For each customer C, the bridge is the same as the company cascade with customer-specific rates:

Pocket Price (C) = List × (1 − On-Invoice rate (C) − Performance rate (C) − Promo rate (C) − Consumer rate (C))

The on-invoice rate is usually constant across the channel. The three off-invoice rates are where customer specificity lives.

Band spread

Band Spread (EUR) = Max Pocket Price − Min Pocket Price

Band Spread (%) = (Max Pocket Price − Min Pocket Price) / Min Pocket Price × 100

The percentage form is the comparable metric across categories and currencies. A 22 percent band on snacks reads exactly the same as a 22 percent band on home care.

Pocket price realisation by customer

PPR (C) = Pocket Price (C) / List Price × 100

PPR Spread = Max PPR − Min PPR (in percentage points)

In the EUR 36 worked example, the band runs from PPR 67.2 percent (worst customer at EUR 24.20) to PPR 87.2 percent (best customer at EUR 31.40). A 20-point PPR gap on a single SKU is large. On most rate cards, a 10-point gap is already worth a separate negotiation.

Volume-weighted vs unweighted

The chart shows one circle per customer. The dollars do not. Plot each customer as a circle scaled by annual volume and the picture often inverts: the largest circles cluster on the left of the band. That is the Marn-Rosiello finding restated in modern words. The customers who get the best deals are usually the ones who buy the most, which means a disproportionate share of trade investment is funding the lowest-realisation accounts.

Volume-Weighted Pocket Price = Sum across customers of (Pocket Price (C) × Volume (C)) / Total Volume

If the volume-weighted pocket price sits well below the simple average, the band is being dragged down by big accounts. That is the trade investment pattern the bridge is meant to surface.

A mid-sized European coffee brand (EUR 240M gross sales, four major customers covering 70 percent of volume) had a board-level commitment to grow profit faster than revenue. The CFO asked for one chart: where does the trade money actually go.

The setup

One SKU, the 250g ground coffee jar, EUR 8.40 list. Four customers: a national grocer (G), a discounter (D), a wholesale partner (W), and a convenience chain (C). Each had its own contract negotiated over different cycles by different commercial leads.

What the company-average bridge said

The company-level G2N for the SKU read like a normal mid-market coffee P&L:

A 22 percent total trade rate. Comfortable for the category. The number that ended up on every dashboard.

What the customer band said

When the same cascade was rebuilt customer by customer, four pocket prices came out:

What the chart revealed

Three findings from a single afternoon's analysis.

First, the wholesale partner (W) was the lowest-realising customer. The team's working assumption had been that the discounter (D) was the cheapest deal in the portfolio. They had been pricing the wholesale relationship as if it were a structural channel, but a stack of historical performance terms had compounded over six contract cycles and had never been re-examined.

Second, the volume-weighting made it worse. The grocer (G) and the wholesale partner (W) together carried 64 percent of the SKU's volume. The two lowest-realising accounts were also the two largest by far. The volume-weighted pocket price was EUR 6.32, materially below the unweighted average of EUR 6.55. The trade investment was concentrated where realisation was thinnest.

Third, the discounter (D), often viewed by commercial as the "bad-margin customer", was actually the highest-realising. The on-invoice structural rate was high, but there was almost no off-invoice. The customer's pocket price was the cleanest in the portfolio.

What changed

The next contract cycle did three things, all explicitly priced.

The wholesale partner's performance rebate band was rebuilt against current targets, not historical ones. Pocket price recovered EUR 0.45 per jar, worth EUR 1.6M over the contract.

The grocer's national multibuy funding was renegotiated with a measurement clause: the brand pays only against verified incrementality. Pocket price recovered EUR 0.20 per jar in year one, with line-of-sight to more in year two as the measurement matured.

The discounter's structural rate was held constant. The team stopped trying to "fix the discounter margin" because the customer view showed there was nothing to fix. The energy went where the band said it should go.

The band is a diagnostic. It does not tell you what to do; it tells you which conversations to have. The practitioner's job is to translate the shape into actions, customer by customer.

Step 1: build the band

Run the G2N bridge for every customer in the top 20 (or top 80 percent of volume, whichever is fewer). Use the same SKU across all of them, so the list price is constant and the only variation is the deduction stack. The first run takes a couple of weeks. Subsequent runs are much faster once the template exists.

Step 2: classify the outliers

Mark the bottom three and top three of the band. For each, write one sentence on what put them there. The bottom-three sentences usually read like "growth rebate threshold paid out at full + heavy Q4 promo + national multibuy". The top-three sentences usually read like "structural only, no JBP performance term, low promo participation".

Step 3: cross-tab band position with strategic value

This is the move most teams skip. The bottom of the band should be your highest-strategic-value accounts (the ones where you genuinely want to invest). Often it is not. It is whichever accounts simply asked for the most money over the years. The band-versus-value cross-tab is what forces the question: are we paying for strategy, or are we paying for accumulation?

Step 4: write the action map

Each bottom-band customer gets one of three labels.

• Earned the price: high strategic value, growing, hard-to-replace shopper reach. Pocket price stays where it is, possibly with a small structural recoupment over two cycles.
• Negotiated into it: medium strategic value, but the pocket price reflects historical concessions, not current performance. Restructure at the next contract window with a clear payback rationale.
• Outgrown the deal: low strategic value, declining share, or a channel that has moved past its prime. Pocket price needs material recovery within one cycle, and if the customer walks, the volume is replaceable.

The labels become the contract calendar. The calendar becomes the recovery plan. The recovery plan becomes the year-on-year improvement in the band shape.

Step 5: track band tightness as a KPI

The metric to watch over time is band spread, not average pocket price. A team that pulls the bottom of the band up (without giving away the top) earns higher PPR every year without raising the headline rate. Tracking the spread quarterly is what makes that improvement visible.