The Power Law in DTC Product Portfolios: Why 6 of 38 Products Drive 94.5% of Revenue

Key Takeaways

Every DTC operator with more than a handful of SKUs knows the feeling: a few products carry the business while the rest sit in the catalog consuming attention, inventory dollars, and fulfillment bandwidth.
The power law in product revenue is well-documented.
The 23 micro-tests averaging $56 each are not waste.
If you are running 20+ SKUs and allocating ad spend proportionally across the catalog, you are subsidizing losers with winner revenue.

The Setup

Every DTC operator with more than a handful of SKUs knows the feeling: a few products carry the business while the rest sit in the catalog consuming attention, inventory dollars, and fulfillment bandwidth. The instinct is to fix the underperformers -- run more ads, adjust pricing, redesign the landing page. The data says something different.

Product revenue in DTC does not follow a normal distribution. It follows a power law. A small number of products generate nearly all the revenue. The rest contribute noise. The operators who understand this do not try to make every product a winner. They build systems that find winners fast, scale them aggressively, and kill the rest before they become a drag on operations.

This is not a branding problem or a marketing problem. It is a portfolio management problem. And the data is unambiguous about how it works.

What the Data Shows

The power law in product revenue is well-documented. A 2023 analysis by Profitero found that across 30+ DTC brands, the top 10-15% of SKUs typically drive 70-80% of total revenue. Harvard Business Review research on product portfolio management (2024) showed that the Pareto principle (80/20 rule) understates the concentration in most DTC catalogs -- the actual split is closer to 90/10 in high-SKU-count portfolios. BCG's 2024 DTC portfolio study found that brands with disciplined SKU rationalization grew 2.1x faster than those that maintained bloated catalogs.

The Stealth Labz portfolio data shows a concentration even steeper than the industry benchmarks.

38 products tested over 28 months. $1,075,946 in gross revenue. The distribution:

Top 6 products (15.8% of SKUs): 94.5% of net revenue (~$888K)
Mid-tier 9 products (23.7% of SKUs): 5.0% of net revenue (~$47K)
Bottom 23 products (60.5% of SKUs): 0.5% of net revenue (~$4.3K)

The top 6 by net revenue:

Product	Net Revenue	Peak Month
PRD-01	$509,821	Feb 2024 ($173K)
PRD-02	$128,794	Feb 2024 ($93K)
PRD-03	$100,909	Sep 2025 ($59K)
PRD-04	$71,792	Feb 2024 ($72K)
PRD-06	$48,280	Dec 2024-Aug 2025
PRD-05	$28,574	Feb 2025 ($11.5K)

That is a 94.5/15.8 split -- far steeper than the standard 80/20 Pareto distribution. In practical terms: 6 products made the business. The other 32 were either learning opportunities (the mid-tier) or fast, cheap experiments (the micro-tests).

How It Works

The 23 micro-tests averaging $56 each are not waste. They are the testing layer. Each ran on shared infrastructure -- same Konnektive CRM, same payment processing, same affiliate tracking, same fulfillment pipeline. The marginal cost of testing a new product was near zero.

The portfolio management discipline has three rules:

Rule 1: Test cheap, kill fast. A micro-test that generates $56 and gets killed in 30 days costs almost nothing in infrastructure. It costs a configuration, a landing page, and a small traffic test. If it does not convert, it dies. No sunk cost fallacy, no "maybe it just needs better creative."

Rule 2: Scale what converts, not what excites. PRD-03 launched in August 2025 and generated $100,909 net in five months. It was not the most interesting product in the portfolio. It was the one that converted. The operator scaled traffic to it because the unit economics worked, not because of brand enthusiasm.

Rule 3: The infrastructure serves the portfolio, not any single product. When PRD-01 declined from $173K/month to zero, the infrastructure did not die with it. The same stack that processed PRD-01 at peak went on to process PRD-03, PRD-08, and every other product in the portfolio. The products are disposable. The infrastructure compounds.

The mid-tier products ($1K-$25K each) serve a different role. Products like PRD-07 ($14,419), PRD-08 ($11,487), and PRD-09 ($7,317) represent recent launches still building traction. Some will break into Tier 1. Most will not. The portfolio discipline gives them a defined window to prove traction before they get killed or scaled.

What This Means for DTC Operators

If you are running 20+ SKUs and allocating ad spend proportionally across the catalog, you are subsidizing losers with winner revenue. The power law says your top 2-4 products are generating 80-95% of your actual contribution margin. Every dollar of CPL spent on a bottom-quartile product is a dollar not spent scaling a winner.

The operational implication: build your stack so product testing is a configuration exercise, not a rebuild. If launching a new SKU requires a new Shopify store, new tracking setup, and new affiliate agreements, you will test 3-5 products per year. If launching a new SKU requires configuring an existing CRM, routing existing traffic, and measuring through existing attribution, you will test 10-15+ per year. The latter operator finds more winners because they run more experiments.

Stop trying to fix the bottom 60% of your catalog. Find the next product that belongs in the top 15%. Your infrastructure should make that search cheap and fast. Your ROAS on the winners will more than cover the cost of the experiments that did not work.

References

Profitero (2023). "DTC SKU Revenue Analysis." Revenue concentration across 30+ DTC brand catalogs.
Harvard Business Review (2024). "Product Portfolio Management." Pareto distribution in high-SKU portfolios.
BCG (2024). "DTC Portfolio Study." SKU rationalization and growth velocity.
Keating, M.G. (2026). "Case Study: The Product Launch Engine." Stealth Labz. Read case study
Keating, M.G. (2026). "The Compounding Execution Method: Complete Technical Documentation." Stealth Labz. Browse papers