eCommerce Product Categorization: 10 Scientific Approaches

Given how much shopping happens online these days, we’ve almost (well, almost) forgotten how we’d need directions and specifications to guide us through the aisles at a supermarket. 

The best ones had wayfinders shoppers could just follow to come across products they wanted. 

These “wayfinders” in a typical eCommerce store are categories that nest multiple products. 

So clearly, the better the categorization, the easier shoppers find it to discover products they’d really want to buy. 

This piece is going to cover what’s relevant for you to come up with effective product categorization.

You'll also find info on: eCommerce Categorization: How a Category Tree Can Be The Answer

10 Powerful Ways to Optimize Your eCommerce Product Categories

Now that you have a sense of what it takes to come up with a well-structured eCommerce product taxonomy, let’s see what you can do it optimize your chosen categories better:

1. Uncover missing or confusing categories

A part of streamlining your current product categorization is to uncover categories with confusing terminology, which shoppers are perhaps eager to buy from but don’t end up finding:

👉 Study zero-search results 

It’s possible that shoppers are looking to find products under “formal shirts” but your products are hidden behind “office wear”. 

✅ You will have to amp up your synonym matching capabilities and rename your categories based on key phrases customers are most likely to use. 

👉 Look at categories competitors are using differently

For example, are they using certain seasonal categories more prominently to target specific customer intent? Or do they have more specific subcategories like “wide fit shoes” under “footwear” or “flaky skin” under “skincare?” 

✅ What you need is a gap analysis matrix that can help you map the differences between how you and your competitors are categorizing eCommerce products.

To give you an example, we went looking for similarities and differences in how skincare brands FaceTheory and Keys Soul Care go about their eCommerce product categorization. 

What we found FaceTheory does is make their site more navigable by using attributes like “concern” and “ingredient” that narrow down a shopper’s search intent:

On the other hand, by introducing “Featured” as a separate attribute alongside regular categories, Keys offers a more comprehensive view of what their website features: 

👉 Combine customer feedback & heatmap testing to reveal patterns

Typically, missing or confusing categories will cause drop-offs from different points across the customer journey. 

✅ Check for clicks & drop-offs across heatmaps and position exit intent surveys with questions like “Did you find what you were looking for?” by adding a number of probably multi-choice options under “No”. 

2. Include specific terms for category naming SEO

Search engines love or de-prioritize eCommerce websites based on how effective the internal information structure is and by judging if shoppers will find what they’re looking for. 

To avoid being ranked low, you’re better off using more specific keywords for category and subcategory names, also matching customer intent in the process. 

👉 Leverage the most relevant keywords

When we say “relevant”, we essentially mean those keywords that best match customer intent across various levels of the conversion funnel. 

✅ Enter top competitor domains into a keyword tool and look at transactional keywords—this will tell you how their product searches take place.

👉 Analyze click paths to find out how customers navigate

The intent behind this is to decode your shoppers’ mental models so that you can definitively know why they click on certain categories and what search terms they prefer. 

✅ Check your analytics to identify which categories serve as common entry points and which categories shoppers skip frequently. Similarly, your click path analysis can improve if you find out how shoppers transition between categories. 

👉 Check customer feedback & reviews for natural language descriptors

While your heatmaps and A/B tests may reveal technical patterns, direct feedback and review sections will tell you what customers are enjoying or not able to resolve. 

✅ Find patterns in the adjectives shoppers use to describe a product (for example, is “lightweight laptop case” a common feature or “non-greasy foundation” one?)—also watch out for negative reviews that mention multiple categories, as these may likely indicate category confusion.

3. Run competitor filtering analysis for each category

To make your eCommerce product categories come alive beyond the primary navigation, you need a filtering system that shows product matches closest to shopper preference. It’s anyone’s guess why this has a big positive impact on conversion rates. 

👉 Check filter count & placement

This can potentially tell you if applying the filters is easy for shoppers and how UX-friendly their visit will be. 

✅ Count across filter types including multi-choice, binary, range selectors etc.—alongside seeing which filters appear most frequently in competitor sites, find out how many you realistically need based on your product range. Also check which filters appear above the fold and how they’re grouped. 

👉 Identify both common & unique filters

If you notice certain filters across, say, three of your top competitors, then that’s a reasonable guess those filters are contributing to at the least engagement, and at best, conversions. Similarly, it’s worthwhile to look at filters that are uncommon but seem to be highly relevant for that category. 

✅ If you find certain filters present more than 90% of the instances, then they’re critical to engagement and conversion success—secondary filters would appear in about 50% to 90% of the instances while niche filters would appear less than 50%. 

👉 Take a look at filter naming patterns

This is a step to identifying not just patterns across your competitor eCommerce sites but also what seems to be working as an industry standard. 

✅ Pay special attention to clarity, word count and structure, to see which filters seem intuitively more usable than the others (for example, a filter like “resolution” technically makes sense for a store that features TVs, but in real-time shoppers may prefer a simpler term like “screen sharpness” to buy.)

When we went and compared how competing hair care brands Olaplex and Kitsch use filters, we saw that they have products that treat almost the same issues, but that the naming of their filters were only similar in “feel”—while the former sticks to the “nature” of the product (what the product does) in their filters, the latter seem to prefer “type” (what the product is) as a naming convention:

4. Analyze your product returns

Unclear categorization and misclassification can often lead to shoppers buying products that don’t match their concerns or preferences. This is why going back to your data sets to check patterns and gaps in your store’s product returns can be key to an improved eCommerce categories list. 

👉 Identify categories & attributes that register the highest returns

Assess return rates by category and attribute—poor segmentation can often be the reason why shoppers purchase products that don’t match their intent. 

✅ Check if certain materials / colors / ingredients are prone to more returns—also check if certain categories are experiencing unusually high returns. 

👉 Audit frequently stated return “reasons”

It may help to classify these reasons based on product type, category and attribute. 

✅ Look for patterns across the reasons like “inaccurate product description” or “color mismatch”—alongside, consider showing more narrowed down attributes to match customer intent more closely (for example, introducing “true to size” may be important in a category if you’ve been getting “runs small” as a reason for returns.)

👉 Compare returns data with customer reviews

This will help you go an extra layer into understanding customer objections and how many of them have got to do with your product taxonomy. 

✅  Watch out for objections like “doesn’t work for tooth sensitivity”, which may in turn mean subcategorization of “sensitive” under “tooth care” could be missing and needs to be included.

Further Reading: 21 Proven Ways to "Prevent" eCommerce Returns (+ Smart Handling ideas)

5. Go deeper into questions customers ask

Your shoppers will ask questions when they don’t find answers right away in your store, either by way of FAQ, customer support or available product info. But this is in fact a great way to see if your product categorization is doing what it needs to do: resolve customer objections and help conversions. 

👉 Audit customer questions about the nature of a product

Are they frequently asking about fit, benefits or compatibility? For example, are you getting queries like “is this water-resistant?” around several of your eye makeup products?

✅ To resolve this, find out if a question being asked has a high volume (for example, 15 times in a month) or if there are more clicks to view the answer to a question (it means that question is important)—in context of the earlier example, higher question & engagement volumes may mean you’ll have to introduce “water-resistant” as a separate attribute within a specific category.

👉 Identify search queries that don’t return matching results

High numbers of searches that return generate generic results, which are then not engaged with often indicate a mismatch between recommendation and intent. 

✅ This may require you to add those search terms (or a part of them) to relevant products and even create a separate category / attribute if the data reveals such a pattern (for example, if you get many searches for “eco friendly winter jackets” and can only match that with “winter jacket” suggestions, it may mean that some of your products need a separate category under “Sustainable Fashion”.)

👉 Review Q&A sections across product pages

These can reveal fresh opportunities for product categorization tweaks as you may come across areas that haven’t yet been touched across reviews or chat queries. 

✅ Looking specifically at pre-purchase questions may help you identify patterns better—for example, a question like “is this chair good for extended use?” and several +1s as replies to that may reveal several people needing the same answer. In this case, you may need to tag the product under “extended outdoor seating” for better discoverability. 

6. Identify high intent

The better you’re able to map the mental shopping models of those who visit your site, the more you’ll be able to sharpen the product categories on your eCommerce store. 

This will in turn serve the interests of high intent shoppers more while upping your store conversions as well. 

👉 Study search results that have led to conversions

While viewing the data, you may want to focus on specific metrics more than others - the more granular you get, the more you’ll be able to see consistent patterns around which search terms are used more and which products get more views. 

✅ Check conversion rate per search query, product views & add to cart directly from searches and which search queries have brought in higher AOV.

👉 Study failed searches

Especially keep a lookout on queries that result in low product views, queries that lead to multiple searches but no buys and queries that typically led to cart abandonment or bounce offs.

✅ Focus on failed searches that have to do with product variations, complementary products or even similar products from different brands—this could help you improve the product scope of your eCommerce store. 

👉 Identify patterns across search-to-purchase paths

The idea is to identify patterns that will reveal higher buying intent, like searches that show up exact product names and searches that often come with multiple filter refinements. 

✅ Map the entire customer journey from search to checkout to get a fuller understanding of product page views, clicks on offers and searches between product views—if there are repeated searches, it may indicate shoppers are experiencing friction along the way. 

eCommerce mammoth Amazon tests user intent by A/B testing dynamic filters across products based on how well they’re selling—for example, they introduce price filters for bestselling products just to assess which price bracket shoppers are buying most from:

Further Reading: Build high-converting category pages (13 ideas + great examples)

7. Make it easier to discover products

Product discovery is one of the biggest yet understated reasons for eCommerce conversions to repeatedly happen. And it’s got everything to do with how you balance depth and breadth in your product categorization. 

👉 Assess how accessible your products currently are

Ideally all your products should be only about 2-3 clicks away when shoppers are in your main navigation.

✅ Apart from a well-organized product navigation, introduce breadcrumbs across category and product pages, and ensure the most relevant & related products show up as recommendations on individual product pages.

👉 Identify if some categories have too many subcategories

This could mean there are duplicates across the board and some subcategories may not even logically belong to the categories they’ve been tagged under.

✅ If you’re creating subcategories out of every variant, you’ll have to look at turning them into filters instead—this will create less confusion for shoppers.

👉 Study categories with higher bounce rates

These could indicate some have too few products, irrelevant products or pricing that doesn’t match what shoppers are looking for (for example, more discounts to throw off competitor products).

✅ Check if the category names are unclear and highlight trending subcategories to call attention to those categories that have been experiencing higher bounces. 

👉 Check if complex categories are optimized enough

Ideally no category should contain multiple similar sounding subcategories, which in turn feature product duplicates, leading to friction and confusion- the same is the case with categories that contain too many filters. 

✅ Identify how shoppers find complex categories on your store—is it through the search function, the navigation or through recommendations? This can tell you which category path to optimize more over others.

❌ What to avoid:

- Orphaned categories (those with about 1 or 2 products)

- Overcrowded categories (with too many subcategories, and often confusing duplicates)

- Dead-end categories (those with a larger logical theme but with no further subcategories or products)

One eCommerce brand that maintains a hierarchy that’s unique to every large category depending upon complexity (for example, since “gifts” falls under a relatively complex category for the brand, they distinguish several attributes by which a shopper can buy whereas “merch” is less popular, so price is the only attribute they use to categorize this section) is Burt’s Bees:

Further Reading: 33 Scientific Ways To Improve eCommerce Product Discovery

8. Make space for additional categories

Before deciding on how many eCommerce categories you want to maintain or expand to, you’ll have to think about how you want to scale up your business. Without this forethought, you may end up with very few or way too many categories. 

Here are some questions you’ll be better off asking: 

👉 How much do I plan to increase my store’s inventory? 

Pay attention to your current product line, market demand and product fit of existing offerings. 

In a way, you have to track patterns of the bestselling categories because they likely indicate what sort of categories can work in the future. 

✅ It’s ideal to design for about 5 x of your current inventory size to be on the safe side. 

👉 Which of my categories are bound to take a backseat?

Like categories that are doing well, you’ll have to also focus on categories that aren’t or are likely to drop in demand in the future. This could be for several reasons including market changes, product innovations in the category or even bestselling categories winning a chunk of your sales. 

✅  Some underperforming categories actually need merging and not getting rid of. Merge them into more intuitive categories—this could even improve the visibility of the products that are currently under them. 

👉 What market expansion opportunities are likely to present themselves?

Apart from reassessing your TG based on emerging trends, you’ll also have to be on top of changes / innovations competitors are implementing and gaps that exist in product-market fit that haven’t yet been addressed by anybody. 

✅ Continue your customer segmentation efforts to predict emerging trends in the short term more precisely—for example, are buyers cutting their spending yet moving towards affordable indulgences?

👉 How do I stay on top of seasonal fluctuations?

To do this effectively, you’ll first have to firm up year-round categories. Next, you’ll have to see what temporary promotional categories you’ve run so far and if including a few more would be relevant to your marketing strategies. Finally, look back at seasonal category performance in the last years and zero in on patterns around what worked and what didn’t. 

✅  Plan better for your category transitions, whether it’s pre-season or post-season. Giving a room of 6 to 8 weeks is ideal. 

Further Reading: How to Boost Sales with Smart eCommerce Merchandising

9. Opt for clear & consistent naming

How you organize your categories, subcategories, filters and attributes can be a defining experience both for human visitors as well as search engine crawlers. 

Here are a few aspects that you’ll have to look into:

👉 What organizing principle do you want to use for your product categorization?

For example, if you consider Patagonia’s categorization naming convention, it goes like this:

Primary layer - the naming explores activities & use cases

Secondary layer - the naming explores genders & product types

Tertiary layer - the naming explores product types for the main categories and conditions for attribute-based distinctions

✅ Find out how the majority of shoppers search for products on your store—do they look into occasion, use case, technical specification or problem/solution? This will help you arrive at the organizing principle faster.

👉 Have you found a way to replace industry jargon with easy terminology?

If your eCommerce categories list doesn’t have its naming convention to match customer-speak, you’ll end up having many more bounce-offs and much less overall category engagement. 

✅  For example, if you sell gadgets, you’d much rather say “monitors & screens” than “display technology”. Or for that matter, if a fashion store sticks to “activewear” instead of “athleisure”. 

👉 Does the category naming in your store balance descriptiveness with conciseness? 

Without a balance of the two, you can end up with a naming convention that’s either too long or vague, or worse, too generic to incite curiosity. 

✅  Maintain a rule of three or lesser words for your naming convention to target slightly longer key-phrases that customers may naturally be using while not making it clunky. 

🚫 What to watch out for: 

- Super funky, creative names (these tend to be ambiguous at best and don’t tell shoppers what products they’ll find under which category)

- Highly descriptive category names (they’re usually longer than 5 words and disturb the design flow as far as mobile optimization is concerned)

- Inconsistency between category, attribute and filter names (for example, “Little Black Dress” could be a category—in which case, the filter and attributes should also be labeled as “black” and not “charcoal” so that tagging is straightforward)

10. Create intuitive shopping pathways

A shopping experience that is unintuitive will result in drop-offs no matter how extensive your eCommerce category and product list. The usual shopper-in-a-hurry is looking to compare quickly if they’re just beginning to engage with your brand or close a buy quickly if they trust you. 

This requires you to factor in logical pathways of categorization before you come up with a final structure:

👉 Create personas to match intent better

This will also help you come up with subcategories that are more nuanced and match different parts of the customer journey. 

✅  For example, if you don’t have a separate persona for Advent pre-orders, how will you introduce that as a separate seasonal category?

👉 Use parent-child logic to come up a categorization structure

Start with broad logic and narrow down one level at a time to avoid being too narrow or messing up the logic in the process. 

✅ “Shoes > Men’s Shoes > Running” is better than “Shoes > Men’s Shoes > Lightweight Shoes > Running”

👉 Leverage filters and tags to create cross-pathways

The idea is to create multiple pathways for shoppers who may have different nuanced preferences while wanting the same product largely. 

✅  For example, when you’re subcategorizing “Running Shoes”, you can make these discoverable through multiple tags including “Nike”, “Marathon”, “Minimalist” and even “Under $5000”. 

👉 Enable faceted search for potentially complex categories

This is especially true for categories that can potentially have a vast number of subcategories, attributes and filters, and if all were to be a part of the main menu, it would run more than 5-6 layers deep. 

✅  Men’s brand Bonobos has so many different kinds of T-shirts that they bring in facets like “tees”, “henleys” and “polos” for a narrow, close-to-preference search:

Further Reading: The Founder's Guide to Customer Journey Map (eCommerce)

eCommerce Categorization: How a Category Tree Can Be The Answer

In the end, the larger end goal for your eCommerce categories list is to help shoppers find products they’d like and in the process, improve your conversions as well as profit margins. 

How a Product Categorization Tree Helps

Just how climbing the right tree can fetch the fruit you want, creating a product category tree that flows logically and intuitively can affect three key areas for eCommerce business success:

✔ Product discovery: A well-structured category tree intuitively leads shoppers to products that match their intent, where each click is a possible movement towards a conversion

✔ Search engine indexing: An effective product categorization tree also offers search engines the most suitable pathways to index your store, leading to better internal linking, crawling and finally, ranking

✔ Inventory planning: By looking at your analytics in relation to the product categorization tree, you’ll be able to locate underperforming categories, those that need to be merged and those that are consistently performing

How to Build an eCommerce Product Categorization Tree

To build your store’s very own product taxonomy tree, follow these steps:

⛳ Perform in-depth research to identify patterns in how your current customers are searching for products and which searches are yielding product views, product page visits and add-to-cart actions. At this stage, it’s also necessary for you to make a record of product attributes that you can later develop as filters, facets and sorters. Alongside, also look at competitor websites to see category structure and hierarchy. 

⛳ Plan the structure carefully to keep things simple and UX-friendly. This is when you define the main categories so that you can nest products within the most logical subcategories. At this stage, you also have to set hierarchies and plan for scaling up your inventory. Creating a naming hierarchy based on a set pattern like Gender > Category > Subcategory > Attribute will also help plan for the next stage. 

⛳ Validate with real users to see if your eCommerce categories are high on navigability and findability. Verify if the navigation paths you’ve set are working or causing friction and if your categorization logic is intuitive enough for users. Also check if your naming conventions are enabling people to find what they’re looking for. 

⛳ Build the category framework by setting up category rules, establishing URL patterns and setting up the content & SEO structure as well. This is also the right stage to decide on which redirects are most critical and how the logical flow of breadcrumbs will read. 

⛳ Monitor, analyze and adjust based on the feedback you receive during testing and validation. Pay attention which navigation paths seem to carry the least friction and which return the maximum conversions. 

🚫 What to watch out for:

- Structural depth of 3 to 4 levels 

- Too much jargon in the naming

- Poor URL structure

- Lack of mobile optimization

- Hidden categories

- Inconsistent breadcrumb structure

Further Reading: 15 Critical Steps In eCommerce Competitor Analysis

Enhance product categorization with sharper UX

98% of visitors who visit an eCommerce site—drop off without buying anything.

Why: user experience issues that cause friction for visitors.

And this is the problem Convertcart solves.

We've helped 500+ eCommerce stores (in the US) improve user experience—and 2X their conversions.

How we can help you:

Our conversion experts can audit your site—identify UX issues, and suggest changes to improve conversions.