Ecommerce product optimisation (EPO) is a way to make more sales by improving every aspect of your store's digital experience using your products’ data.
This isn't an acquisition channel like SEO, and it's not a single process or piece of technology. It's an approach to making your store work better by extracting every piece of data you can about your products, and finding novel and innovative ways to increase revenue that aren’t possible without that data.
Here's everything you need to know about ecommerce product optimisation.
What is ecommerce product optimisation (EPO)?
In the ecommerce world, EPO stands for ecommerce product optimisation. Just like other digital initialisms you're familiar with - SEO, CRO, PPC, UX - it refers to a set of processes, tasks and workflows with a common end goal.
The goal of EPO is to optimise the products in your ecommerce store to get the maximum financial value from them.
It's all about product data. Specifically:
- What product data you have, and
- How you use it
As an ecommerce store operator, you use EPO to extract all the data you possibly can from your products and put it to use on your website to make more sales more efficiently.
Find out about our EPO service
What is ecommerce product data?
"Product data" could mean lots of different things. So let's clarify 100%: when we talk about "product data" in an EPO context, we're talking about the taxonomy of structured data that describes your products.
That's largely product attributes, stored as a definitive set of tags.
Doing EPO starts with turning your unstructured data - mostly text and images - into structured data.
Because your product descriptions and images hold a wealth of untapped data. Through AI-driven image recognition and text mining, you can extract masses of attributes that usually get overlooked. Then you match these newly discovered attributes with those from keyword research, ensuring a comprehensive tagging system across your entire catalogue.
You might be thinking you already have this stuff - and maybe you do. But the key thing is structure. You might have incredibly detailed product descriptions, but if that information only exists as free text, then it's not useable data.
EPO turns it all into data that you can forever use in a manageable, scalable way in your PIM, backend, CMS, wherever you need it.
We have our own toolset for EPO, and it starts with automated product tagging to find every attribute in an inventory. From there we can tag products directly in some CMSs.
How many attributes can there really be?
Sometimes the difference EPO makes is turning knowledge you already have into structured data. And sometimes it uncovers attributes you'd never even considered. The difference can look something like this:
Key takeaway: EPO is about turning unstructured data into structured data, which you can then use in almost limitless ways in your marketing and website development
How do you do EPO on an ecommerce website?
EPO has six fundamental steps:
- Product research: Get market intelligence on the demand for your products. This can come from keyword research
- Attribute identification: Find every relevant attribute for your products to know exactly what your customers are searching for
- Attribute extraction: Find every attribute in your product inventory (shape, size, colour, style, material, active ingredient, etc.)
- Attribute matching: Match the attributes extracted from your products to the ones customers are searching for
- Catalogue enrichment: Update your product catalogue with complete tags and a clear taxonomy
- Ecommerce optimisation: Unlock the possibilities - publish hundreds of new landing pages for SEO and PPC, create richer product descriptions, improve cross-sell, recommendations and visual merchandising, feed into CRM for personalised offers and retention. It's up to you
Points 1 to 5 are where you create your EPO database. Point 6 is where you'll get value from EPO. This is where you set your product data to work, and where it gets really interesting. We'll cover all the use cases in detail in future posts, but let's look at a few ideas to get you started.
What are some example use cases for EPO?
Conversion
- Make personalisation much more targeted and accurate on your site and every touchpoint of your customer journey
- Offer better targeting for cross-sell, upsell and recommendations
UX
- Implement richer filters across your product listings pages
- Improve the accuracy and usability of site search
PPC
- Create campaigns around higher-quality, more relevant landing pages by targeting intents more precisely with attribute-specific landing pages
- Optimise your feed: EPO gives you the richest possible range of product product_types for paid campaigns:
Home > Screws, nails and fixings > Roofing screws > Countersunk > Self-drilling > Multipack > 100 > 5.5mm x 100mm
SEO
- Page-scaling: find opportunities to make your products much more visible in organic search. It works by combining attributes to create new landing pages - think 'green velvet sofas' and 'wooden Montessori toys for 2 years olds.'
- Optimise product description pages with better information and richer tags
- Optimise product listings pages with more accurate copy
- Make internal linking more relevant
What are the benefits of doing EPO for your website?
EPO has benefits across the entire customer lifecycle: acquisition, conversion and retention. Some of the key benefits are:
- improved SEO performance (visibility, traffic, revenue from organic)
- improved PPC performance (better ad relevance, more product_types to optimise you feed, higher quality scores, better CTR and CVR)
- improved CRM performance through better, more accurate personalisation and recommendations
- new merchandising possibilities: identify gaps in your inventory where there's significant demand from existing customers that you could meet with new products
- better UX: improve filters, customer journeys, personalisation
- CRO: endless possibilities for creating pages and experiences that more closely meet customer intent, create new A/B tests
What's the difference between SEO, CRO and EPO?
EPO overlaps with SEO and CRO techniques, but it is a distinct approach with its own specific goals, which are all about revenue.
Most EPO activities benefit SEO, CRO, UX, or all three. But the difference is this: with SEO, you are specifically trying to improve your website's organic search performance. Sometimes it's a happy coincidence that an SEO activity also benefits your conversion rate.
In contrast, the goal of EPO is to optimise the products in your store to benefit both search visibility and conversion. Instead of thinking about separate disciplines that relate narrowly to either acquisition, conversion or retention, EPO allows you to be more efficient because it deliberately improves metrics at all points of the customer lifecycle.
Why is EPO important for your ecommerce business?
The role of product attributes in SEO
The way your products are categorised and tagged significantly impacts how customers discover them. If your inventory lacks detailed, relevant attributes, you're missing out on valuable opportunities to attract and convert customers through organic search.
EPO enriches your product catalogue with comprehensive, targeted attributes, unlocking new pathways for customers to find exactly what they’re looking for on your site.
Working with an EPO agency
EPO is a new concept in 2024, and currently Melt Digital is the only EPO agency in the world. But the concept of ecommerce product optimisation is gaining traction and recognition, so you can expect more agencies to offer it as more retailers recognise its benefits.
The most important thing to quiz an EPO agency about is their data engineering capabilities. Like most things in marketing and ecommerce, there are different processes and ways to do EPO. Everyone will have their own way, and how they get it done is less important than the quality of outputs. So always look for EPO case studies that show evidence of results with EPO.
FAQs about EPO
Isn't EPO a doping thing from sports?
If you're an athletics or cycling fan, you might know EPO as an illicit performance-enhancing substance. The ecommerce kind of EPO is also performance-enhancing. But the similarity ends there.
Let's clear this up: in biology and sports, EPO is short for erythropoietin, a hormone that helps the human body produce red blood cells. Dodgy athletes use it to cheat. That is not what we're talking about. We're talking about an approach to managing an ecom store that uses the store's products themselves to create better shopping experiences and improve all the key metrics ecommerce professionals care about: AOV, CAC, conversion rate, revenue, traffic, ROAS and even CLV.
How long does it take to see results with EPO?
EPO can have an instant impact or it can take six months before you see the exact results you want. It completely depends on how you use it and what your end goal is.
When you carry out the first stage of EPO and create a complete tagging model for your products, you'll see the an immediate benefit in that you simply have a load more useful product data. Of course, that doesn't necessarily equate to material gain on its own.
After that, it can be relatively simple to make UX improvements like richer filters. Depending on your site build and CMS, you could implement a new set of filters within a month of starting an EPO project, immedaitely boosting your engagement and conversion rates.
Use EPO for SEO growth, for example, and it's realistic to see significant improvements within six months. That's what happened with our client Danetti, who saw 52% revenue growth and a 500% increase in non-brand traffic to collections pages in six months.
What kind of businesses can benefit from EPO?
If you have an ecommerce website with more than 50 SKUs, you can benefit from EPO. It's applicable to all sectors.
EPO also gets more powerful with scale. If you have over 1,000 SKUs, you should be doing ecommerce product optimisation. If you're not, you may be taking risks for BAU with your data model, and you're likely missing out on opportunities for growth. It's especially powerful with products that have lots of different attribute types and variants.
If you have a store with just a few products, it's still a good idea to put a structured data model in place so you can make sure that a) you have structured data for optimising your existing products and b) you're futureproofed if you ever expand your catalogue. You might be able to apply EPO principles to your catalogue and just create the data manually.
If you have a handful of products and would like more data insight to inform your buying strategy, you could use EPO to help you identify new products to stock.
