Product Taxonomy & Attributes
Scalable taxonomies, classifications and attributes informed by customer demand, assortment needs and business priorities.
Product Data Consulting for E-commerce
Better product data.Better commerce.
I help retailers build smarter product data foundationsthat improve discovery, simplify complexity, and enable scalable growth.
Full service
Strategic advisory, analysis, assessment and design for retail product data.
Search & Product Discovery Analysis
How customers search, browse and shop: demand patterns, language opportunities and gaps in product information.
Product Data Strategy & Governance
Data models, standards, governance frameworks and strategic direction to manage product information at scale.
How I work
Listen, Shape, and Embed stay connected from the first conversation through governed foundations, fed by demand, assortment, and context.
Listen
Search, navigation, and assortment signals that show how customers shop and what the business can support.
Shape
Taxonomy, attributes, and data models shaped by demand, grounded in the assortment, and built to scale.
Embed
Standards, governance, and operating recommendations that keep product data foundations adaptable.
Three inputs inform every decision.
The same demand, assortment, and context that feed Listen, Shape, and Embed. Relevant today. Scalable tomorrow.
Customer Demand
How customers search, browse and shop, from external demand signals to internal site search, navigation and filter behaviour.
Product Assortment
What the business sells today and plans to sell tomorrow: categories, variants, complexity and future expansion.
Business Context
Commercial, organizational and technical needs: priorities, systems, selling channels, regions and constraints.
Outcomes, not just deliverables.
Product data improvements connected to meaningful customer, operational and commercial value.
- Better Product Discovery
- Stronger Data Quality
- Reduced Complexity
- Scalable Foundations
What better product data is worth.
Where product data shows up on a retail P&L: manual rework behind the scenes, high-intent demand lost in search and filters, margin eroded by avoidable returns, and the first-year payback from fixing the foundation.
$50–150K
Yearly waste reworking the same product data
Grounded in Harvard Business Review research
HBR finds teams spend up to half their time hunting down and fixing bad data. Mid-market band when attributes lack an owner and seasonal feeds get rebuilt by hand.
$100–250K
Yearly demand at risk from failed search and thin filters
Sized from Google Cloud & The Harris Poll + Baymard Institute
Mid-market revenue band where zero-result searches and broken attribute filters bounce high-intent buyers before they reach a product page.
$35–80K
Year-one margin kept by cutting avoidable online returns
Sized from National Retail Federation benchmarks
With 19.3% of online orders returned industry-wide, accurate fit, sizing, and material specs protect mid-market margin before orders ship.
3–5×
First-year return on focused taxonomy and attribute work
Harvard Business Review: The Cost of Bad Data
Grounded in HBR's 1-10-100 data quality multiplier: fixing product data at the source compounds across lower manual rework, recovered search demand, and fewer spec-driven returns.
Catalogue agility at the source
A governed product schema means new seasonal lines, vendor feeds, and categories go live in days instead of burning merchandising hours on manual spreadsheet cleanup (Harvard Business Review).
Search that converts
Shoppers search with natural language, feature combinations, and intent. Structured attributes bridge customer queries directly to relevant products, eliminating zero-result drop-offs (Google Cloud & The Harris Poll).
Filtering that prevents abandonment
Baymard Institute benchmarks prove that optimizing attribute filters and category taxonomy reduces product list abandonment from 67% to 90% down to 17% to 33% (Baymard Institute).
Fewer returns, protected margin
Accurate dimensions, clear variant mapping, and verified material specs address the primary controllable driver of online retail returns before orders ship (National Retail Federation).
Research figures synthesize usability benchmarks, retail return studies, and merchant surveys. They reflect directional retail dynamics, not a single promise for every catalogue.
The numbers behind the cost.
Follow a shopper through an unstructured catalogue: search misses the query, filters fail to narrow the list, conflicting specs stall the checkout, and inaccurate details send the order back.
76%
U.S. shoppers who hit a failed retail site search
Google Cloud & The Harris Poll
Step 1 (Search): When shopper queries do not map to structured attributes, about 8 in 10 leave the retailer and buy elsewhere.
67–90%
Product lists abandoned when filters and categories fail
Step 2 (Browse): Shoppers cannot narrow the assortment. After taxonomy and filter fixes, abandonment drops to 17% to 33%.
54%
Shoppers who abandon a purchase over conflicting specs
Salsify Consumer Research, 2025
Step 3 (Product Page): The buyer reaches the item, but contradictory dimensions, materials, or titles break trust at checkout.
19.3%
Online retail sales returned to the merchant
National Retail Federation (NRF), 2025 + Salsify Consumer Research
Step 4 (Post-Purchase): Nearly $850B in U.S. retail returns, with 71% of shoppers citing inaccurate or incomplete product specs as a return trigger.
What can be done
A structured product data foundation does not need a multi-year replatform. Auditing the attribute model, unifying category hierarchies, and setting clear governance for variant mapping removes discovery dead ends and protects operating margin.
Curious what weak product data is costing the catalogue? Independent advisory for e-commerce taxonomy, attribute mapping, and data strategy.
Book a callWhat's new
Latest notes on retail product taxonomy, discovery, and governance.
Why AI Discovery Fails Without Structured Product Data
Why conversational shopping assistants, semantic search engines, and AI merchandising tools hallucinate or drop high-intent products when the underlying taxonomy and attribute architecture are ungoverned.
Practical Governance for Enterprise Product Information
How retail and e-commerce leaders establish lean, pragmatic data stewardship that accelerates seasonal launches rather than bottlenecking them.
Closing the Gap Between Shopper Intent and Catalogue Data
Why internal merchandising vocabulary often fails customer queries, and how attribute enrichment turns zero-result searches into high-intent conversions.