Product Portfolio Optimisation for FMCG Margin Recovery

Product Portfolio Optimisation for FMCG Margin Recovery

The $5M Coffee Brand That Was Quietly Going Broke

A founder I worked with last year ran an Australian challenger coffee brand doing $5.4M in revenue across grocery and DTC. Topline was up 22% year on year. The board deck looked great. Six quarters earlier the portfolio held 41 SKUs. By the time I met him it carried 87, spread across whole bean, ground, capsules, ready-to-drink, cold brew concentrates, brewing accessories, and three different gift packs.

He was proud of that growth. He was also bleeding contribution margin per shelf-foot per week at a rate that would have put the brand in trouble inside twelve months, and his revenue dashboard was the reason he could not see it.

The pattern is not unusual. McKinsey complexity research puts the cost of product complexity to U.S. food and beverage manufacturers alone at roughly $50 billion in gross profit each year, and the firm finds that a structured complexity program typically lifts gross margins three to six points while trimming SKU count by a quarter. The cost is not exotic. It is the same handful of line items, repeated across most $1M to $10M physical product brands: trade spend on slow movers, slotting fees retailers do not refund, freight drag from low-cube SKUs, 3PL pick rates that punish small orders, and the management attention you sink into items the consumer does not actually need.

The coffee founder's dashboard ranked SKUs by gross revenue. The top 12 SKUs delivered 73% of revenue. Looked healthy. When we loaded the actual cost-to-serve onto every line item, those 12 SKUs delivered 91% of contribution margin, and the bottom 41 SKUs collectively delivered roughly 6% of revenue at deeply negative gross margin. Every retailer "exclusive" he had agreed to over six quarters was a margin tax he was paying for the privilege of growth he could not bank.

This article is about what we did next, the framework I have used across multiple FMCG portfolios since, and the two-quarter window in which a brand can move from revenue pride to margin discipline without losing distribution. Most of the operators I have walked through this exercise resist the first phase because it forces them to look at numbers their finance team has been avoiding. Once they see the bottom quartile rendered in true contribution margin per shelf-foot per week, the kill list writes itself.

Why the Math Doesn't Work: The Tail You Cannot See

A revenue-sorted SKU dashboard is the single most expensive piece of misinformation in a growing FMCG portfolio. It hides the negative-margin tail because revenue is the wrong denominator for a shelf-led business.

The consumer does not buy a SKU on its own. They buy it from a finite block of shelf real estate that is sold to you by the retailer at a known cost. Every centimetre your slow movers occupy is a centimetre your hero SKUs cannot use. Every dollar of trade spend you commit to keeping a tail SKU on shelf is a dollar that does not flow through to the items that actually carry your brand. Bain on simplification reports that brands which take a strategic approach to portfolio reduction grow sales 2 to 5 percentage points faster than peers and lift margins by 100 to 400 basis points, because the tail is not just unprofitable, it is actively suppressing the head.

The hidden cost stack on a tail SKU has at least seven entries that almost never appear in a P&L review at this revenue band. Trade spend allocated to the SKU. Slotting fees the retailer charged at listing and never refunded. Freight drag from low-cube or low-velocity items shipped on full pallets. 3PL per-pick fees on small grocery picks and DTC orders. Spoilage and short-coding write-offs on slow turners. The complexity tax on demand forecasting, where every additional SKU pulls forecast accuracy down. And the attention tax on your operations team, who spend disproportionate hours managing items that contribute almost nothing to the result.

Once you stack those costs back onto the SKU, the rerank changes everything. Circana SKU rationalization finds that for most growing consumer goods brands, the bottom quartile of SKUs typically generates under 10% of true profit while occupying a disproportionate share of distribution, slotting, and trade dollars. Harnessing simplicity puts the upside more concretely: a structured simplification program lifts net revenue 1 to 4 percentage points, margins 3 to 6 percentage points, and asset productivity 10 to 25%, with about a 25% SKU cut as the typical lever.

The denominator that actually matters is contribution margin per shelf-foot per week, weighted by velocity. That is the only metric that captures the three things FMCG operators are really trading off: profit per unit, units sold per period, and the finite shelf real estate the unit needs to sell from. A flat 80/20 cut by revenue gets you nowhere close, because some 20% revenue items are top-quartile shelf earners and some 80% revenue items are quietly destroying value. The contrarian spine of every FMCG portfolio review I run is shelf-foot-velocity-weighted contribution margin, not a Pareto chart drawn off the wrong axis.

The Portfolio Contribution Engine Blueprint

I call this The Portfolio Contribution Engine. It is a four-component framework that turns a revenue-sorted dashboard into a margin-led portfolio decision system. I have run a version of it across coffee, supplements, sauce, and personal care brands between $3M and $25M in revenue, and the bottom-quartile finding holds up almost every time. The split is not always 30 to 50% of SKUs delivering under 10% of profit. Sometimes it is 25%. Sometimes it is closer to 60%. The shape of the tail varies. The fact that it exists, and that the revenue dashboard hides it, does not.

The four components of The Portfolio Contribution Engine are sequenced because each one feeds the next.

The first component is the Cost-to-Serve Load. You take the last four quarters of trade spend, slotting fees, freight, 3PL pick fees, returns, and spoilage, and you allocate them to the SKU level using the cleanest rolling P&L your finance function can build. SKU profitability framework is a practitioner-grade template for exactly this exercise, with trade spend, freight, and 3PL overhead loaded into a CFO-style rolling SKU model. Most $5M operators build this in Google Sheets first, then graduate to a tool like Cin7, NetSuite, or a Jirav model once the numbers are stable.

The second component is the Shelf-Foot Velocity Index. You pull POS or scan velocity from each retailer and DTC channel, divide by the linear shelf-foot the SKU occupies (or the equivalent for online: category page real estate, hero rotation slots), and produce a weekly velocity figure per SKU per channel. Multiply that by the loaded contribution margin per unit and you have your true denominator: dollars of margin per shelf-foot per week.

The third component is the Kill, Keep, Fix Matrix. Each SKU lands in one of four quadrants based on contribution margin per shelf-foot per week and strategic role. Kill quadrant: negative margin with no strategic role. Fix quadrant: positive margin but bottom-quartile velocity, where price-pack, packaging, or placement work has a real upside. Keep quadrant: top-quartile margin and velocity. Halo quadrant: lower direct margin but draws shoppers into the category or unlocks retailer commitments. Mastering portfolio complexity gives a methodology for the consumer-first rebuild that complements this matrix when you need to reset assortment from scratch.

The fourth component is the Retailer Negotiation Script. Most founders treat SKU cuts as a confession to the buyer. Done well, the cut list becomes a negotiation lever: cleaner shelf, higher productivity per facing, better forecast accuracy, in exchange for shelf space on hero SKUs, improved terms, or co-promo dollars. The kill list is leverage, not surrender.

Execution: Day 0 to Day 90

The Portfolio Contribution Engine runs in four phases over a 90-day window. A two-person finance and ops team can run it without external consultants. A founder running solo can do it in roughly three days of focused work plus weekly check-ins.

Day 0 to Day 15: Cost-to-Serve Load. Pull the last four quarters of trade spend by SKU, slotting fees by retailer, freight cost by SKU using actual shipped weight and cube, 3PL pick fees by line, returns and spoilage write-offs, and any private label cannibalisation cost. Allocate them to the SKU level. Build the rolling SKU P&L in Sheets if you do not already have one in your finance stack. Owner: finance lead or fractional CFO. KPI: every SKU has a fully loaded contribution margin per unit by Day 15.

Day 16 to Day 45: Shelf-Foot Velocity Ranking. Pull POS data from each grocery account (Coles, Woolworths, IGA, independents), DTC velocity from Shopify, and Amazon velocity if relevant. For grocery, use IRI or Circana scan if available, otherwise retailer-supplied weekly movement. Compute units per shelf-foot per week per channel, multiply by loaded contribution margin per unit, and produce one ranked list per channel and a blended-weighted list across the portfolio. Tag each SKU with its quartile and its contribution share. Owner: ops or category lead with finance support. KPI: full ranked list across channels by Day 45.

Day 46 to Day 75: Kill, Keep, Fix Decisions. Run the four-quadrant matrix. The kill list is items with negative loaded contribution margin and no halo or strategic role. Bain reducing complexity is the right reference for the value-versus-premium distinction at this point: in a value-led category the tail is pure cost and should be cut hard, while in a premium-led category a longer tail can fund choice and novelty if it is genuinely earning its keep. The fix list is items with positive but bottom-quartile margin per shelf-foot, where you will run packaging, price-pack, or merchandising work over the next two quarters. The keep list is the top quartile and any halo SKUs that draw shoppers and unlock retailer commitments. Clorox SKU cuts is the canonical operator example of a formal kill process boosting both margin and sales, and it remains the cleanest reference for a brand owner who needs to convince a sceptical co-founder or board. Owner: cross-functional kill committee (founder, finance, ops, sales). KPI: every SKU labelled by Day 75.

Day 76 to Day 90: Retailer Communication and Trade Reset. For each retailer affected, build a one-page deck that frames the cut list as a service: cleaner shelf, higher productivity per facing, fewer out-of-stocks, better forecast accuracy. Trade the kill list for shelf space on the keep list, improved payment terms, gondola end commitments, or co-promo dollars. Reset trade spend agreements so the new portfolio gets the budget the old one was wasting. Owner: founder or sales lead. KPI: every grocery account renegotiated by Day 90.

A single landmine to avoid: founders often resist cutting SKUs because a particular retailer "asked for them." Those SKUs are not protected. They are negotiation chips. If a buyer demanded an exclusive that is now bleeding margin, the cut conversation becomes the renegotiation conversation, and the price of keeping the SKU should be paid back in shelf space, terms, or co-promo. If the buyer will not pay, the SKU should not stay.

From Revenue Pride to Margin Discipline

Two quarters after we ran The Portfolio Contribution Engine on the coffee brand, the portfolio had moved from 87 SKUs to 52. Contribution margin per shelf-foot per week was up 6.4 points across the grocery footprint and 8.1 points on DTC, where 3PL pick fees had been silently eating the smallest pack sizes. The kill list released roughly $186,000 in working capital that had been sitting in dead inventory and obsolete packaging. The founder used the cut list to negotiate two new gondola ends in Coles and a payment-terms shift from 30 days to 14 days with one independent grocery group. None of that was visible from the revenue dashboard. All of it was visible the moment we changed the denominator.

The mindset shift is the part that lasts. Revenue rank is vanity. Contribution margin per shelf-foot per week is the only sane denominator for a shelf-led business, and it is a metric that cannot be cheated by adding more SKUs to chase topline. Once a founder runs the rerank once, the urge to say yes to every retailer "exclusive" gets replaced with a question they did not ask before: will this new SKU earn its centimetre of shelf, this week, after fully loaded cost-to-serve.

If you are running a $1M to $10M FMCG brand and you have not loaded full cost-to-serve onto your SKU list in the last four quarters, your dashboard is lying to you in the same direction it lied to the coffee founder. The fix is not exotic. It is a four-week finance build, a four-week velocity ranking, four weeks of kill, keep, fix decisions, and four weeks of retailer renegotiation. The brands that run it twice a year do not have a tail problem. They have a portfolio that earns its shelf. That is the real test.

Pull your top 50 SKUs this week. Sort them by contribution margin per shelf-foot per week, with full cost-to-serve loaded. Look at where the bottom quartile sits. The number you find is the margin you are leaving on the floor every quarter you keep the tail.