Want to Fix Your Unit Economics? Do What Nestlé Did- Start Saying No to More SKUs

Tony Paul
Oct 31
8 min read

In 2021, Nestlé made a bold move that few companies of its size dare to make.

They didn’t launch a new product. They started deleting them.

Project TASTY — Nestlé’s global SKU rationalization program — was launched to simplify the company’s portfolio and improve unit economics.

Here’s what they discovered:

👉 34 % of Nestlé’s SKUs contributed just ~1 % of sales. 👉 Only 11 % of SKUs generated ~80 % of revenue.

The logic was clear but courageous: if one-third of your 100,000 SKUs drive only 1% of sales, it’s time to clean house.

By 2025, Nestlé reported over CHF 1.2 billion in savings — the result of trimming low-margin, low-rotation SKUs and focusing on the high performers.

The outcome?

✅ Higher on-shelf availability (up to 97%) ✅ Improved service levels ✅ Stronger margins despite inflationary pressure

As Nestlé CEO Mark Schneider explained, a leaner portfolio was “win-win-win”:

“Consumers get their popular items, retailers get fast-moving products, and we gain less complexity, more operational efficiency, and better on-shelf availability.”
That’s the power of data-driven simplification — cutting with clarity, not emotion.

Behind this transformation lies a data story — one built on analytics, digital tools, and tough decisions guided by facts, not feelings.

Because whether you’re Nestlé or a challenger brand, sometimes the smartest move is to delist strategically — not to shrink, but to simplify.

And simplification, done with data, compounds into clarity, speed, and profitability.

Project TASTY — Nestlé’s global SKU rationalization program

The Retail Overload Problem

Retailers today face a paradox: they have more data than ever, yet margins are thinner than ever. The biggest culprit? Over-assortment.

Every new product SKU adds cost — procurement, storage, marketing, and shelf space. Yet only a fraction truly drive revenue. At Datahut, we’ve seen this repeatedly across retail and e-commerce: product counts balloon, complexity rises, and profits quietly erode.

That’s where data-driven range rationalization comes in — the process of refining your assortment using internal and competitor data to strengthen your unit economics.

As Harvard Business Review notes, “getting product assortment right isn’t easy, yet it’s absolutely critical to retail success” — which is exactly why analytics and web-scraped competitor insights now sit at the heart of modern retail strategy.

Why Range Rationalization Matters

Most retailers assume adding more products equals more choice and happier customers. In reality, it often means higher costs, cluttered shelves, and confused shoppers.

Each extra SKU introduces:

Additional supply-chain handling
Increased working capital lock-up
Forecasting and replenishment complexity
Marketing dilution

When only 20 % of SKUs drive 80 % of profits, trimming the long tail becomes the smartest way to boost margins without raising prices.

A McKinsey case study on analytical assortment optimization found that a grocer cut SKUs by 36 % and lifted sales and margins by 1–2 %, showing how KPI-driven delisting directly improves unit economics.

The Profitability × Customer Commitment Lens

At Datahut, we encourage retailers to evaluate every SKU across two axes:

This framework clarifies which SKUs deserve your attention — and which don’t.

McKinsey’s Periscope assortment optimization brief calls this approach “identifying which categories are under- or over-represented to optimize the assortment and increase sales.”

How Competitor Data Strengthens Range Decisions

The smartest retailers don’t make assortment decisions in isolation. They combine internal sales data with competitor-scraped data to see what’s working across the market.

With retail data scraping and competitive assortment analysis, you can uncover insights such as:

1. SKU Density & Category Mapping

Web scraping lets you map how competitors distribute SKUs across categories — for instance, how many “premium organic snacks” a rival carries versus “everyday snacks.” This helps identify whether your range is over- or under-represented in profitable categories, echoing McKinsey’s advice that “category managers today should build strategies based on customers’ needs and willingness to pay”.

2. Price-to-Value Positioning

By collecting real-time competitor pricing, packaging, and size data, you can benchmark your price-value equation. However, as HBR warns in its guide to real-time pricing, “retailers that use simple heuristics like scraping lowest competitor prices miss significant opportunities.” The takeaway: intelligent pricing models must blend competitor data with demand signals, not chase the cheapest tag.

3. Discount & Promotion Patterns

Web-scraped data reveals how often competitors discount products and by how much. Platforms like BCG X’s Merch AI show how AI tools synthesize such internal + external data to create “customer-centric assortments and cost advantage,” often lifting margins 1–3 %.

4. Stock Availability & Demand Signals

Tracking competitor out-of-stock (OOS) events offers a goldmine of demand insights. If rivals repeatedly run out of a variant, it signals unmet demand — a theme echoed in Walmart Data Ventures’ Assortment Deep Dive, which stresses building “customer-centric assortments by gaining insights into shopper purchasing behaviors.”

5. New Product Velocity

Competitor monitoring shows how fast others launch or retire SKUs — critical for innovation pacing. McKinsey’s “How to Predict Your Competitor’s Next Move” advises focusing on one or two competitors’ pricing and product portfolios to anticipate actions instead of spreading efforts too thin.

When stitched together, these insights help retailers improve unit economics by reducing waste, focusing on profitable categories, and aligning pricing with real market behavior.

Web Scraping: The Data-First Advantage

Modern web scraping for retail isn’t about collecting random data — it’s about structured, compliant, and high-fidelity extraction that powers decision-making.

At Datahut, our infrastructure enables:

Automated capture of competitor product, pricing, and availability data across thousands of SKUs daily
Extraction of category-level intelligence such as product titles, images, variants, and metadata
Integration of clean, structured datasets into BI tools, pricing engines, or inventory models
Continuous monitoring of market shifts for pricing, promotions, and stock changes

Deloitte describes similar AI agents in retail as “digital workers that ingest sales trends, customer browsing behavior, and competitors’ movements in real time”, showing how automated data feeds enable dynamic assortment tuning. Likewise, NRF reports that AI assistants can become “experts in product assortment” with deep semantic understanding of every product, reinforcing the power of automated intelligence in merchandising.

Unit Economics: The Data-First Path

Optimizing unit economics isn’t just about cost cuts — it’s about improving contribution margins through smarter assortment.

McKinsey’s Online Grocery Fulfillment study notes that “the challenge boils down to unit economics” — a $100 basket can swing from +$4 in-store profit to –$13 online once logistics costs hit. That’s why assortment discipline and competitor benchmarking matter.

Datahut’s retail data extraction and pricing-intelligence solutions help brands:

Collect real-time competitor pricing and product data
Analyze SKU-level profitability
Identify assortment gaps and redundant SKUs
Feed structured data directly into pricing or inventory models

Bain & Company similarly points out that even Amazon’s cashier-less prototype “is likely to face challenging unit economics” (Bain Insight), underscoring that innovation still hinges on profitability per unit sold. Meanwhile, IBM’s 2024 NRF report shows AI tools now “optimize store-level assortments” by analyzing each store’s mix to maximize sales — the future of unit-economic optimization in action.

AI and Algorithmic Retailing

Retail is entering the age of Algo Retailing, where algorithms balance customer need and profitability at every SKU. As MIT Sloan Management Review explains, AI techniques use large data sources “to decide on store-level assortment and drive deep localization at scale” (MIT Sloan).

Kroger’s 84.51° case study illustrates this in practice: its ML-driven planogram optimizer predicts category sales for each layout, yielding about $18 million in extra annual sales through better assortment and shelf allocation.

These examples prove that web-scraped competitor data, when fused with AI and machine learning, can turn retail complexity into measurable profit.

The Courage to Simplify

The hardest part of assortment optimization isn’t the analysis — it’s the courage to act. Retailers often know which products underperform but hesitate to delist them. Simplification isn’t shrinkage — it’s strategy.

As McKinsey emphasizes in its category-management research, success means “building strategies based on customer needs and willingness to pay” — not on what suppliers push. Or as Walmart Data Ventures puts it, the goal is to “build a customer-centric assortment by gaining insights into shopper purchasing behaviors.”

With clean, compliant, and competitive web data guiding decisions, retailers can finally move from intuition-based merchandising to intelligent assortment design.

Final Thought

In modern retail, success isn’t defined by how much you sell — but by how efficiently you sell it.

Using competitor data scraping, AI-driven assortment analytics, and data-first decisioning, you can build a product range that’s lean, profitable, and aligned with customer demand.

If you’re ready to uncover what to keep, cut, or create —Datahut can help you get there.

FAQs

FAQ 1: What is Nestlé’s Project TASTY?

Project TASTY is Nestlé’s global cost-efficiency and portfolio simplification initiative launched in 2021. Its goal is to reduce operational complexity, streamline the company’s vast product assortment (over 100,000 SKUs), and drive structural savings—all without compromising product quality or taste (hence the name TASTY). The program focuses on eliminating underperforming SKUs and optimizing recipes and packaging, helping Nestlé improve service levels, cut waste, and reinvest in core brands and innovation.

FAQ 2: How many SKUs did Project TASTY target, and why?

Nestlé revealed that about one-third of its SKUs accounted for only 1% of total sales—a classic long-tail problem. Project TASTY was designed to “cut the tail” by removing thousands of low-rotation, low-margin, or duplicative SKUs that were consuming disproportionate resources. The rationale was to free up supply chain capacity, improve in-stock rates for high-performing products, and simplify manufacturing, logistics, and marketing operations globally.

FAQ 3: What results has Project TASTY delivered?

Since its rollout in 2021, Project TASTY has generated over CHF 2 billion in cumulative savings (as of early 2025), helping Nestlé mitigate severe cost inflation and protect profit margins. SKU reductions and recipe/packaging optimizations boosted supply chain efficiency and improved product availability—raising shelf availability of top-selling items from ~95% toward a 99% target. Nestlé also saw organic growth improve after cutting low-value SKUs, with executives calling the initiative “a significant boost” to their growth model.

FAQ 4: What is the Profitability × Customer Commitment matrix, and how does it help retailers optimize their assortments?

The Profitability × Customer Commitment matrix is a two-axis model that helps retailers classify SKUs based on their financial contribution and customer relevance. Each SKU is evaluated across:

Profitability: How much margin it generates, considering real costs.
Customer Commitment: How essential it is to target shoppers (e.g., loyalty, frequency of purchase, substitution risk).

This framework creates four clear quadrants — star performers, margin drainers, overlooked gems, and dead weight — enabling data-driven decisions to invest, optimize, promote, or delist SKUs. It transforms assortment planning from intuition-driven to margin-optimized.

FAQ 5: How does competitor data improve assortment and pricing decisions?

Retailers that rely only on internal data risk missing market context. By scraping structured competitor data — including SKUs, pricing, promotions, stock levels, and launch velocity — retailers can:

Benchmark assortment density by category.
Identify pricing gaps or opportunities.
Detect emerging trends (e.g., frequent OOS events signaling unmet demand).
Monitor innovation cycles and promotional aggression.

This external intelligence, when fused with internal performance data, helps build customer-centric, high-margin assortments aligned with real-time market dynamics.

FAQ 6: Why is simplifying the assortment critical to improving unit economics?

Over-assorted portfolios often include low-velocity SKUs that drain working capital, complicate operations, and dilute brand focus. Simplifying the assortment — based on both profitability and customer value — improves unit economics by:

Reducing warehousing and supply-chain complexity.
Increasing shelf availability for top-performing SKUs.
Lowering per-unit costs through higher production frequency.
Enhancing pricing power by focusing on high-demand products.

This isn’t about cutting for the sake of cutting — it’s about designing a leaner, smarter range that delivers better margins and stronger customer relevance.