5 minutes
A lot of CPG brands found out what tariffs cost them at month-end close, not before. If your duty costs are folded into COGS without a separate line, you can't tell whether a margin dip came from pricing, volume, or a rate change. Breaking that out is the first step, and the rest of the model follows from there.
TLDR:
Tariffs fold into COGS through landed cost, so your gross margin is wrong if duty isn't loaded per SKU.
Build a six-column exposure map (HTS code, origin, volume, pre-tariff COGS, rate, annual duty cost) before modeling anything.
Use a blended inventory rate per period: on-hand, in-transit, and open POs carry different duty costs simultaneously.
Run three scenarios (base, bear, bull) through your full realization waterfall per channel, not a single blended assumption.
Iris Finance can carry tariff duty as its own SKU-level cost field, separate from product COGS, and runs bull, base, and bear scenarios against a locked base case, with a dedicated analyst keeping duty inputs current so the model never needs a spreadsheet rebuild.
Why Tariff Exposure Varies Across Your CPG Portfolio
Not every SKU in your portfolio faces the same tariff. That sounds obvious, but most brands model it as though it doesn't matter.
The average statutory tariff rate stood at 11.2% as of the Budget Lab at Yale's November 9, 2026 update, and it is projected to reach 11.4% by the end of 2026 under current law. That average is built from thousands of HTS codes, each carrying its own rate, often varying sharply by product category and country of origin. A supplement brand importing raw ingredients from one country and finished goods from another can easily have SKUs spread across 0% and 25% duty rates within the same catalog.
Three variables drive this spread:
HTS classification: how your product is categorized under the Harmonized Tariff Schedule
Country of origin: where manufacturing or substantial transformation occurs
Product category: finished goods, ingredients, and packaging often fall under separate codes
A single blended assumption collapses all of that. If your average across-portfolio rate looks manageable, it may be masking a handful of high-volume SKUs sitting at rates that are quietly destroying margin.
How Tariffs Enter the COGS Line
Tariffs don't appear as a separate line in your P&L. They fold into COGS, quietly, through landed cost.
Ask most founders what a unit costs and they will quote the number on their co-manufacturer's invoice. That number is the starting point of your cost, not the end of it.
Landed cost is the true per-unit cost to get inventory from the factory floor to a sellable location. For any imported product, it stacks like this:
Manufactured or purchase cost (the invoice price most brands stop at)
Inbound freight: ocean, air, or ground from the supplier
Customs duties and tariffs: calculated by applying your HTS-assigned duty rate to the customs value of the goods
Customs broker fees and entry filing costs
Port fees: MPF, HMF, and terminal handling
Marine or cargo insurance
The duty itself is straightforward math: customs value multiplied by the applicable HTS rate. What makes it complicated is that customs value is not always the invoice price. It can include assists, royalties, or related-party adjustments depending on your import structure.
When any of those components are missing from your COGS figure, every downstream number is wrong, which is the blind spot Iris Finance closes for CPG brands by tracking duty as its own SKU-level cost field inside one unified model. Gross margin is overstated. Contribution margin is overstated. The pricing floor you think protects you doesn't.
Building a Tariff Exposure Map by SKU
Start with your full SKU list. For each one, you need four data points before any modeling is possible:
HTS code (8-10 digits for US imports) so you can pull the correct duty rate from the schedule
Country of origin, since rates vary dramatically by sourcing geography
Unit volume imported per period, which sets the scale of your exposure
Unit COGS pre-tariff (invoice price only, no duty loaded in yet)
Two more columns complete the picture: the applicable duty rate and the resulting annual duty cost per SKU (unit volume x unit COGS x duty rate).
That table (a SKU identifier plus the six data columns) is your tariff exposure map. The SKUs, HTS codes, origins, volumes, and rates below are illustrative, not real duty rates for any product or country:
SKU | HTS Code | Country of Origin | Units/Year | Unit COGS (pre-tariff) | Duty Rate | Annual Duty Cost |
|---|---|---|---|---|---|---|
SKU A | 2106.90 | China | 50,000 | $4.20 | 25% | $52,500 |
SKU B | 3304.99 | Vietnam | 30,000 | $6.80 | 10% | $20,400 |
SKU C | 1901.90 | Mexico | 20,000 | $3.50 | 0% | $0 |
This shows you exactly which SKUs are carrying the tariff burden, and it feeds directly into tracking GMROI benchmarks at the SKU level. In this illustrative example, SKU A absorbs $1.05 per unit in duty costs alone. If your current COGS for that SKU excludes that number, your gross margin on it is fiction. Every model that skips this map is building on a cost input that's already wrong.
Calculating the Inventory Blend Rate Per Period
Applying today's tariff rate to all your inventory is a modeling error that compounds across quarters.
On-hand stock was purchased and cleared customs at a prior duty rate. In-transit orders will clear at whatever rate applies at entry. Open purchase orders haven't landed yet. All three buckets carry different duty costs per unit, and your gross margin for any given period draws from all three simultaneously.
To model this accurately, you need a blended rate for each period based on your inventory composition:
Units on hand, with duty already embedded at the prior rate
Units in transit, taxed at the duty rate expected on their entry date
Units on open POs, taxed at the projected rate on their landing date
Weighted blend rate = sum of (units x unit duty cost) across all three buckets, divided by total units available for that period.
If 60% of your sellable inventory this quarter landed before a rate increase, only 40% carries the new rate. A model that applies the current rate to everything will overstate COGS and understate gross margin now, then flip the error in the next period once older inventory sells through. That is the kind of structural flaw that undermines a Series A financial model when investors dig in. Variance analysis becomes untrustworthy when the mistake changes direction every quarter.
The Three-Scenario Framework for Gross Margin Modeling
Tariff policy has changed quickly enough in the past two years that a single-point forecast is a guess with a spreadsheet attached.
The right structure is three scenarios built off your exposure map and blended rate calculation: a base case (rate holds at current level), a bear case (rate escalates to prior peak), and a bull case (rate reverts toward the pre-tariff baseline). Each scenario inputs a different duty rate into your landed cost formula, which flows into unit COGS calculation, then into gross margin at your current pricing and volume assumptions.
According to the KPMG 2026 Tariff Survey, which surveyed 300 U.S. C-suite leaders at companies with $1B+ in annual revenue in February and March 2026, 55% planned additional price increases of up to 15% over the following six months, even as sentiment improved after the Supreme Court's ruling invalidating IEEPA as a basis for tariffs. A range matters more than a midpoint right now.
One rule before you build variants: lock the base case first. Changing it after building scenarios means your variants no longer mean anything relative to each other. Build it, freeze it, then build the bear and bull cases as deviations from that fixed reference point.
Connecting Tariff Scenarios to Your Realization Waterfall
A tariff increase doesn't land in one place. It enters at COGS, then ripples downward through every margin layer below it.
The realization waterfall runs: gross sales, minus discounts and refunds, minus COGS (where tariff lives), equals delivered gross profit. Subtract merchant fees, ad spend, and variable fulfillment costs to arrive at contribution margin. A higher duty rate raises unit COGS and compresses delivered gross profit first. If you hold pricing flat, every order below that point inherits a tighter margin before channel fees enter the picture.
Where it gets channel-specific is below the gross profit line:
On DTC, merchant fees are relatively fixed as a percentage of revenue, so compression stays proportional.
On Amazon FBA, referral fees and fulfillment fees are layered on top, and those don't move when your COGS does. For illustration, a $12 landed cost SKU selling through FBA at a 15% referral fee plus $3.50 in fulfillment leaves a thinner cushion than the same SKU on Shopify.
On wholesale, if you're selling on a fixed invoice price with net terms, there's no room to recover the duty cost without renegotiating the account.
Run each tariff scenario through the full waterfall per channel, the same lens used in SKU and channel profitability analysis. That's where actual exposure becomes visible.
Pass-Through vs. Absorption: How to Model Each Response
Once your tariff exposure map and scenario waterfall are built, you face two choices: raise prices to recover the duty cost, or absorb it and accept lower gross margin. Both are defensible. Neither should be chosen without modeling the math first.
Take a hypothetical example (illustrative figures, not benchmarks). A unit with a $12 landed cost selling at $30 retail carries a 60% gross margin. A 15% tariff on that $12 adds $1.80 per unit, pushing landed cost to $13.80 and gross margin to 54%. Recovering the full $1.80 in dollars requires raising retail to $31.80, a 6% increase.
Whether that holds depends on price elasticity by channel:
DTC (owned site): elasticity tends to be lower, especially for replenishment categories. Loyal subscribers are less price-sensitive than new visitors.
Amazon: highly elastic. Even a 5% price increase can shift rank and suppress conversion enough that volume loss exceeds the margin gain.
Wholesale/retail: price increases require retailer approval and lead time. If you're on a fixed invoice price, absorption is the near-term reality until renegotiation.
Partial pass-through is a real option too. In the same example, recovering $1.00 of the $1.80 impact (a $31.00 retail price) moves gross margin to 55.5% with a smaller 3.3% retail increase. In the KPMG 2026 Tariff Survey, 78% of organizations reported higher COGS in their most recent fiscal quarter and 51% reported margin declines. The respondents were 300 executives at companies with $1B+ in annual revenue, so read it as a directional signal, not a CPG benchmark.
Model full pass-through, partial pass-through, and full absorption separately. The right answer varies by SKU, channel, and competitive context.
Adapting Your Pricing and Sourcing Strategy After Modeling
The model tells you where you stand. This section is about what to do next.
Four response postures are available. Which one is viable depends on what your gross margin baseline, competitive pricing ceiling, and retailer constraints actually allow.
Absorb: Accept margin compression and protect volume. Only sustainable if gross margin is high enough to tolerate it and you expect the rate environment to ease.
Pass through: Execute a price increase. Timing, channel sequencing, and magnitude matter. DTC first, wholesale with lead time, Amazon with competitive monitoring.
Restructure COGS: Review alternative sourcing countries with lower duty rates, domestic co-manufacturing, or component-level changes that shift HTS classification. This takes quarters, so model the interim exposure while the change is in progress.
Restructure product: Modify formulation, packaging, or bundle configuration to shift the HTS code or customs value. A packaging change that reclassifies a finished good from one chapter to another can meaningfully change the duty rate without changing what the customer buys.
If your gross margin after full absorption still clears your contribution margin floor across all channels, absorption may be the simplest call. If it does not, the model tells you exactly how much recovery is required and where the pricing ceiling sits relative to competitive benchmarks. Work from that number, not from a gut read on what the market will accept.
Integrating Tariff Assumptions into a Rolling Forecast
A one-time tariff model goes stale the moment rates move. The only way to keep it useful is to wire it into the forecast so updates happen automatically, without requiring a full rebuild.
The key structural change: break tariff duty cost out as its own visible line in the financial model instead of folding it into the base COGS figure. When duty cost is buried in COGS, month-over-month margin variance becomes unreadable. You cannot tell whether gross margin dropped because volume shifted, pricing eroded, or duty rates changed. Separating the line makes variance analysis reliable.
Here is how the integration works in practice:
Your rolling forecast holds a duty rate assumption per SKU or SKU group, part of building reliable SKU-level profitability data, updated each period to reflect current HTS classification rates.
That rate multiplies against your blended inventory cost base from the period calculation covered earlier.
The output feeds landed COGS as a sum of base cost plus tariff cost, with both components remaining visible.
When you reforecast monthly, you update one input: the duty rate assumption. Everything downstream recalculates. If the rate held flat, tariff variance is zero. If it changed, the variance isolates immediately without touching any other assumption.
One practical rule: treat the duty rate input the same way you treat your shipping rate or merchant fee assumption. Review it at every reforecast close, document the source and date, and flag it for the next 30 days if any pending policy decision could move it.
How Iris Finance Models Tariff Impact Across a CPG Portfolio
Running this analysis in a spreadsheet is slow and breaks every time a rate changes. Iris Finance is an AI-native FP&A solution for consumer brands: your sales channels, ERP, and ad data sit in one unified model, and your dedicated Iris analyst helps load SKU-level duty costs and keep them current. A rate change becomes a new scenario, not a rebuild.
Scenario Modeling in Iris is ML-based, drawing on Iris's cross-brand data, not a general-purpose AI model. A tariff scenario in Iris returns a realistic range based on what similar brands have actually achieved. Each scenario takes about 10 to 20 minutes to build.
The practical mechanics connect directly to the methodology this article has covered:
Tariff duty can be loaded into a miscellaneous SKU-level cost field, tracked separately from product COGS the same way freight per SKU is, and COGS can be time-bound by SKU so landed cost changes show up in the period they hit
Bull, base, and bear scenarios overlay against a locked base case, so rate changes don't contaminate your reference point
Data behind the Daily P&L refreshes every 1 to 2 hours, and Plan vs. Actual flags metrics as On Pace, Watch, or Off Pace, so you see tariff impact as it moves through inventory, not at month-end close
The 13-Week Cash Flow Forecast pulls in purchase orders automatically when the Inventory Module (currently in beta for complex businesses) is connected, so you see projected cash outflows before inventory lands
When actuals roll forward monthly, the model reforecasts in about 15 seconds after close. A tariff policy change that would otherwise require a full model rebuild becomes a new scenario built in 10 to 20 minutes, which is central to any strategy to improve gross margin for CPG brands.
Final Thoughts on Tariff Modeling and CPG Gross Margin
The work covered here includes building a SKU-level exposure map, blending inventory rate by period, and running scenarios through your margin waterfall. That's what separates a defensible forecast from a spreadsheet that just looks confident. Tariff policy moves fast enough right now that a model you built three months ago is probably already stale. Keeping duty cost as its own visible input is the simplest way to stay current without rebuilding from scratch every quarter. Connect with the Iris team if you want to see how that reforecast works in practice.
FAQ
How do I model the impact of a tariff increase on my CPG brand's gross margin?
Start by building a SKU-level exposure map with six columns: HTS code, country of origin, annual units imported, unit COGS pre-tariff, applicable duty rate, and annual duty cost. That table shows you exactly which SKUs are carrying the burden and by how much per unit. From there, run three scenarios (base case: rate holds; bear case: rate escalates to prior peak; bull case: rate reverts) through your full realization waterfall per channel, so compression shows up at delivered gross profit and contribution margin, beyond a blended COGS adjustment.
Should I absorb a tariff increase or pass it through to customers on Amazon vs. DTC?
The right answer varies by channel and your gross margin baseline, so model all three options before deciding. DTC tends to carry lower price elasticity, especially for subscription replenishment categories, making partial or full pass-through more viable. Amazon is highly elastic: a 5% price increase can suppress conversion enough that volume loss exceeds the margin gain, so absorption or partial recovery is often the more defensible call there. Wholesale requires retailer approval and lead time, which means absorption is frequently the near-term reality regardless of what the model says.
How do you calculate a blended inventory duty rate when on-hand stock, in-transit orders, and open purchase orders all carry different tariff costs?
Multiply units by unit duty cost across all three inventory buckets, sum the results, then divide by total units available for the period. On-hand stock carries the duty rate already paid at entry; in-transit orders carry the rate expected at their entry date; open POs carry the projected rate on their landing date. Skipping this step and applying today's rate to all inventory overstates COGS now and understates it in the next period once older stock sells through, which makes variance analysis unreliable in both directions.
What is a tariff exposure map and why does a CPG brand need one before building gross margin scenarios?
A tariff exposure map is a SKU-level table showing HTS code, country of origin, import volume, pre-tariff unit COGS, applicable duty rate, and resulting annual duty cost per SKU. CPG brands need it because a single blended portfolio rate hides which specific SKUs are carrying the heaviest burden. A supplement brand can easily have SKUs spread across 0% and 25% duty rates in the same catalog, and the average looks fine while a few high-volume items quietly destroy margin.
How do I keep a tariff model from going stale every time duty rates change?
Break tariff duty cost out as its own visible line in your financial model instead of folding it into the base COGS figure. When duty is buried in COGS, you cannot tell whether a gross margin drop came from a rate change, a pricing move, or a volume shift. With tariff cost separated, you update one input (the duty rate assumption per SKU group) at each reforecast close, and everything downstream recalculates automatically without touching any other assumption.
Related posts
Actionable data, proven strategies, and clear forecasts, empowering you to make smarter decisions and scale with confidence.





