19 Jul Your Product Taxonomy Sucks (And Here’s Why You Should Fix It)
By Jason Hein, Principal Strategist at B2X Partners
Congratulations! You’ve just launched your company’s first eCommerce website – after months of hard work selecting technologies, qualifying design partners, and reviewing wireframes the site is now online and the site is live – representing your business to your customers (both old and new). Why don’t you settle back for a few moments and enjoy the moment? Go get a cup of coffee with your boss, take your significant other out for a nice dinner (to remind them that you still exist and live outside the office), treat your team to a night out to celebrate the launch. Go ahead, I’ll wait – you should take some time to appreciate what you’ve done, it’s not a small thing.
Okay, you’re back? Feeling refreshed? Energized? Great! Because I have some news – your taxonomy sucks. That assessment may be wrong, but most eCommerce sites suck on launch day. It’s not because you or your team did a bad job – far from it, they did the best job they could given the resources they had. But most B2B eCommerce sites are built and launched on a lot of assumptions based on the same things that drive their offline sales strategies: opinions and experience. While those are important, offline opinions and experience are often less relevant in the online space. Successful tactics of an experienced VP of Sales for selling face-to-face are great, but they often aren’t relevant online, where users have different preferences and needs.
Why Your Taxonomy Sucks
So what’s wrong, you ask? Odds are, you jumped right into new technologies and site design – excited to start seeing what this new site would look like. Those are important, but many businesses ignore perhaps the most important part of their site: their product taxonomy. The product taxonomy is the way you group and organize the stuff you sell into a set of categories and is usually seen on the left-hand side of a homepage. In general, there are 5 things most companies get wrong with their launch taxonomy:
- Duplicate Categories
Taxonomies are designed to be authoritative – where each item you sell has “one and only one home” within the structure. If you have more than one category where an item can be classified, then you doing the online equivalent of a bad brick and mortar store manager and forcing your customers to walk to multiple parts of your store to see the full selection. You remember how annoying it is to spend an hour in the “Men’s” section looking at polo shirts, only to see one you like better in the “Sporting Goods” section on your way out? Yeah, customers don’t like that on a website either; make it easy for them, show them your full offering of a given product in one place where they can easily compare options and make an informed decision.
- Too Many Choices at Once
Most categories in a taxonomy are hierarchical, where each category has other more specific subcategories underneath it to help guide customers to their desired product. Taxonomies are also designed to be used by humans – and humans aren’t great at choosing from dozens of different options all at once. If your taxonomy features categories with 15 or more subcategories all at the same level then you are making it difficult for customers to chose the “right” subcategory. In those situations, users often bounce between similar subcategories under a parent, which slows discovery and reduces the odds a customer will actually get to a detail page.
- Missing Categories
Have you ever gone into a store looking for something specific, and then looked at the Store Directory near the entrance only to find that the thing you’re looking for isn’t listed? Even if that store does carry the product, if you can’t see it in the Directory then most customers will go somewhere else instead of wasting their time. The same logic holds online. If you don’t have a category in the taxonomy for each product type you carry, then it’s harder for customers (and search engines) to find them.
- “Junk Drawer” Categories
One way companies try to solve the “Missing Categories” issue is by creating generic, non-specific categories like “Miscellaneous” or “Other”. By doing so, they believe they are making it easier for customers who can’t find what they want by creating a centralized place where all the long-tail items can “live”. But what actually happens in these cases is selection in that category explodes. Because it is much easier to classify hard-to-classify products as “other,” these categories often grow exponentially until there is so much selection there (of so many diverse types) that the categories become unusable.
- Mis-Classified SKUs
Identifying categories and building the taxonomy is only the first part of a good taxonomy design – the other part is consistently assigning each product offered to the correct category, a process called “classification”. Categories with a lot of incorrectly-classified products not only confuse customers (“Why is a safety harness the top listing on the ‘Wire Harnesses’ category?”) but they damage your reputation as a trusted source of knowledge and information. If you don’t know where to classify product, then what else don’t you know about that product line? Can you be trusted as a source?
Why It Matters
So, you’ve discovered you have some problems with your taxonomy. Good news: you are not alone. Bad news – there’s not a quick fix. Product content is a program, not a project. This will be an evolution of improvement over time – as you refine your taxonomy, as you build your categories and as you optimize your SKUs. Working on getting it right, however, lets you create a fast, clean product discovery path that allows your business to grow and compete.
About Jason Hein
Jason Hein is an eCommerce expert and product strategy leader. He brings over 20 years industry-leading experience in industrial distribution, including at Amazon Business and McMaster-Carr. At B2X Partners, Jason works to expand content strategy and develop world-class processes for rich product data and optimized product merchandising.