Sales Decomposition: 14% Growth Hid a Household Penetration Decline

A $25M specialty food brand reports 14% year-over-year sales growth to its board. Revenue up. Gross margin up. The CEO calls it a breakout year. By Q3, velocity has stalled across three retailers, and the broker is fielding calls about shelf resets. The topline hid the diagnosis for two quarters: the brand raised prices twice, existing buyers spent more per trip, but fewer households bought the product each period. Growth was real. It was also erosion wearing a revenue costume. The brand is Cinderhaven Provisions (synthetic, built to demonstrate the analysis below) and this exact failure is seeded into its numbers because it is the failure that recurs at real brands running price-led growth.

The decomposition that catches this takes one equation. Sales = households buying x purchase frequency x spend per trip. Each term is a different lever, driven by a different strategy, fixed by a different investment. A brand that reports total sales without breaking them into these three components cannot say whether its growth is sustainable or whether it is burning its buyer base for short-term revenue.

Each lever maps to a different operational motion

The equation is from Byron Sharp's How Brands Grow, and it is the framework behind every serious brand health assessment in CPG. Each lever maps to a different operational motion:

Household penetration, the percentage of households that bought the brand at least once in a period. This is the trial metric. It moves with distribution (more doors), awareness (marketing, sampling, display), and assortment (new items that attract new buyers). When penetration grows, the brand is acquiring customers. When it declines, the distribution footprint is shrinking or the brand is losing relevance against competitive sets.

Purchase frequency, trips per buying household per period. This is the loyalty and usage-occasion metric. It moves with consumption rate, pack size, and competitive substitution. A brand with high penetration but low frequency has a trial-conversion problem: people try it once and do not come back. A demand plan built on shipments rather than consumption will miss this signal entirely because shipments measure what the brand pushed, not what shoppers pulled.

Spend per trip, which itself decomposes into units per trip and price per unit. This is the pricing and trade architecture metric. It moves with price increases, pack-size changes, and promotion depth. When spend per trip drives growth while the other two levers are flat or declining, the brand is extracting more from a shrinking buyer base. That is the pattern that looks like growth on a revenue report and looks like erosion on a buyer-flow chart.

The three contributions sum to the exact sales delta

Decomposition is not useful if it does not reconcile. The three-lever waterfall bridges period-A sales to period-B sales, attributing the exact dollar delta to each lever: how much of the change came from more (or fewer) households, how much from frequency shifts, how much from spend-per-trip changes. The three contributions sum to the total sales change. No residual, no rounding bucket, no "other."

This matters because management teams argue about which lever drove the quarter. Marketing claims penetration grew. Sales claims frequency is up. Finance points to the price increase. The waterfall ends the argument with arithmetic. It names the lever, sizes the contribution, and makes the rest of the conversation about what to do, not what happened.

The buyer-flow analysis sits underneath: new buyers, retained buyers, and lapsed buyers between periods. Penetration can hold steady while the composition underneath churns, new buyers replacing lapsed ones at equal rates, masking a retention problem that a flat penetration number would never surface. The flow chart catches what the percentage hides.

POS dashboards cannot show buyer-level behavior

Most CPG dashboards (Retail Link, SPINS, NielsenIQ scan data) report aggregate sales, not buyer-level behavior. They can show units sold and dollars per store. They cannot show how many distinct households are buying, how often, or how much each one spends. That requires household panel data: a record of individual household purchases over time.

For brands with access to NielsenIQ Homescan, IRI/Circana consumer panels, or retailer loyalty card data, the decomposition runs on real numbers. For brands without panel access, which is most brands under $50M, the methodology still applies. The inputs can be estimated from POS trends and distribution data, or the decomposition can run on synthetic data that mirrors the brand's profile to show what the analysis reveals and what it would cost to leave it unexamined.

Decompose runs the full three-lever decomposition, buyer-flow analysis, and waterfall bridge on Cinderhaven Provisions, a synthetic $25M specialty food brand with 50 SKUs across six retailers. The data is synthetic. The methodology is the same one applied to real panel data. The demo has a deliberate story seeded into the numbers: a stretch where sales grow on price increases while household penetration quietly declines. It is the pattern described in the opening paragraph, made visible.

Three penetration tools answer three different questions

Penetration means three different things in CPG, and each requires a different tool. Decompose handles the household penetration question: are more people buying, or are fewer people spending more? Door Math handles distribution penetration: of the stores that could carry the product, how many actually do? Spin Rate handles velocity: once the product is on the shelf, how fast is it moving relative to its distribution?

The three tools answer different questions and point to different strategies. A brand losing household penetration needs trial drivers. A brand losing distribution penetration needs a door-count recovery plan. A brand with low velocity relative to its distribution needs a merchandising or pricing intervention. Reporting total sales without distinguishing which lever is moving, or declining, means funding the wrong strategy with real money.

The decomposition names which lever is actually moving

Lailara runs the three-lever decomposition on client panel data or retailer loyalty data, with the waterfall bridge and buyer-flow analysis. The deliverable is a period-over-period attribution showing exactly how much of the sales change came from each lever, plus the buyer-flow chart that names where households are entering and leaving. Book a 30-minute scoping call to determine which data sources are available and what the analysis would show.