Last post we covered better content, here we talk about context. Not just that, but a real world example of how AI can help build richer experiences for customers and agents. Check out the Claude example below.
First, let’s define context. While content tells the customer what the product is, what it is capable of, its features, dimensions, materials, etc, every AI search has context that accompanies the search. If you only have product content, and not product context, you will miss out on being matched by AI search in the future: both off site and on site search. Results will be disappointing.
Context is all about the situation and use of the product. It’s the kind of information your customers ask questions about. You see the questions on Reddit all the time (here’s an example). Is this jacket good for skiing in Whistler? How would I wear this jacket for skiing? Is it an outer layer or inner? If I want to use it for hiking as well, will it be suitable? I have long arms, how long are the arms on this jacket? The questions can go on and on and are very specific to that person and situation. This is context.
How can a brand meet this question barrage? How do you ensure that you answer with brand authority as many questions for your customers? If you don’t, someone else will. Here are five key areas:
1) Customer Q&A: you may have a customer Q&A section. An example in the UK is Screwfix (like this product) where the Q&A section has questions on product content (weight) and context (can I wear them after ankle surgery? Can you put orthotics in them?). What a gold mine.
2) Reviews: reviews are chock-full of information to build richer content experiences. They give so much information on sizing, quality and many questions your product content doesn’t traditionally cover.
3) Blog / newsletter: the revenge of the blog. People never read your blog content. They rarely read long product descriptions either. Guess what does? AI does. It is a hoover for all the rich information you can give it. Rediscover what you thought was low value and hard work.
4) Product brainstorm: brainstorm 25 different questions customers could ask about a particular product or have asked before, that are not detailed in the description already: how could you answer them?
5) Category brainstorm: often, a product based brainstorm leads to category applicable content. In our jacket example: a skiing piece of content could make recommendations of situations, linking to jackets that could be outerwear, or layers and why: providing rich context to agents.
But Rob, this doesn’t scale! I have tens of thousands of SKUs and there’s no way I can handle this manually, even for our top 100 products: it would take an age. You’re right. Enter our superstar content curator, creator and worker.
The worked example
What can AI do to help you build your content and context? Rather than tell, I want to show. I took this product from Next (a famous UK retailer) as the subject.

The product we’ll get after from Next - a complex piece of furniture
In my opinion, this product is not ready for agentic prime time:
Short product description: the description is very light on real details
No dimensions: hidden in the image (probably from the manufacturer)
Lacking content: outside of images, very little is said about the product
Lacking context: where will it work? How does it feel? What is it good for?
Alright, enter Claude Cowork to assist me with making this so much better. If you don’t know, Claude Cowork enables AI to drive a web browser so it can look at the webpage from Next and get all the details. You may show this content to the consumer, or you may only use it for an agentic feed. Either way, let’s see what we can come with. Here’s my first shot prompt. Not the best, but let’s see how well it does:
“We're going to do a little project together. I have a theory that you can build much richer product content for ecommerce sites to enable AI and agents to better match a company's product to a user’s intent when they are shopping / discovering new products. I am convinced that using a tool like you, we can create nice rich content from current sources, quite quickly. To prove this through a simple test I want us to use the product on https://www.next.co.uk/style/su345928/508901. It has a very short product description written by the company. In my view it is lacking what AI would need in the future. Let's take this page, and the reviews from customers (there are 15, you might need to hit the 'load more' button to get them all) and make a richer description of this product. Feel free to also examine the images of the product and pick out any details you find to write a richer description. Don't make anything up that you can't see or read about the product. Do think about the different contexts in which the product could be used and what it is really good for.“
Not bad, right? We’ll refine it later, but let’s see how it did.

Claude Cowork doing its thing via the browser
Here’s what it did:
Found the description, found the title, understood the brief
Examined all the reviews, both critical and complimentary
Examined all of the images in detail, including the one with the dimensions
Built a new rich description in a few minutes which is better than my 30 minute attempt would be. While I was doing other work.
Full text of what it came up with is below. Read it. It did a fantastic job, in my opinion. There’s little hallucination there, it really extracted the content from reviews well. The images: spectacular. The dimensions: not so good, that needs work: I give it a pass because in the image there are three separate items and dimensions: not easy. Overall, refinement to the prompt would include leaning on positive reviews more than negative reviews, really checking measurements, and I would give it brand guidelines / voice. So much potential.
This is how you can use existing tools and existing content to create content and context at scale to feed AI search (on site and off site). Look forward to speaking with you again on the next edition of The Conversion.
Soft Velvet Rust Nomad Left Hand Corner Dining Table and Bench Set: AI-Enhanced Product Description
The Product in Brief
A statement corner dining set combining a curved velvet bench with a dark marble-effect oval table, designed for 6-8 people and engineered to transform how families gather, eat, and connect at home.
What This Product Actually Is
The Table: An oval dining table featuring a dark marble-effect surface with dramatic golden-brown natural veining. The curved edges and substantial pedestal base create a restaurant-quality centrepiece that anchors any dining space.
The Bench: A wraparound corner bench upholstered in soft velvet (this version in warm "rust" — a rich terracotta/burnt orange). The bench features a curved, bolstered backrest that follows the table's silhouette, creating an intimate, enclosed dining experience.
The Configuration: Available in Left Hand and Right Hand corner options to suit your room layout. The corner section connects with a straight bench extension, creating continuous seating around two sides of the table.
Who This Is Really For
Families wanting to change how they eat together. Owners consistently report this set has transformed their dining habits — spending more time at the table eating, talking, and lingering rather than rushing through meals. The bench seating creates a casual, approachable atmosphere that encourages people to stay.
Home entertainers seeking restaurant ambience. Multiple owners describe achieving a "restaurant feel" at home — the combination of booth-style seating and elegant table creates the kind of intimate-yet-social dining experience typically found in high-end restaurants.
Open-plan living spaces. Works equally well in kitchens, kitchen-diners, and lounges. The sculptural design serves as a room divider or focal point in open-plan layouts.
People who value design with longevity. Described repeatedly as a "statement piece to last a lifetime" — backed by Next's 15-year structural guarantee covering manufacturing faults and frame defects.
Practical Realities (From People Who Own It)
Assembly: Most owners found assembly straightforward — notably, one self-described "female pensioner" completed it alone without difficulty. All necessary tools are included with clear instructions. However, one owner described unpacking and assembly as "hard work" — individual experience varies, likely based on DIY confidence.
Fit & Posture: The seat depth is slightly shorter than traditional dining chairs. This promotes upright posture rather than lounging — better suited to eating and conversation than extended relaxed sitting.
Table Surface Care: The marble-effect finish is beautiful but requires care. One owner reported the surface shows visible marks when wiped clean, and struggled to find a cleaning solution that didn't leave streaks. This is worth considering for households with young children or those prioritising easy-clean surfaces.
Delivery: Arrives well-packaged in recyclable materials, delivered with care.
Technical Specifications
Dimension | Measurement |
|---|---|
Overall table width | 200cm |
Overall table depth | 160cm |
Table height | 75cm |
Corner bench depth (at curve) | 90.5cm |
Straight bench depth | 46.5cm |
Bench seat height | 40.5cm |
Backrest height | Various |
Materials:
Table: Dark marble-effect finish (not real marble)
Bench: Soft velvet upholstery (Rust colourway)
Frame: Engineered for 15-year structural warranty
Colour Options: Soft Velvet Rust, Olive Green, Natural Mink, Grey
Use Context Signals (For AI Matching)
Best matched to queries involving:
Corner dining sets, booth seating, banquette dining
Family dining furniture, gathering spaces
Open-plan kitchen-diner furniture
Statement dining furniture, design-led dining
Restaurant-style home dining
6-8 seater dining solutions
Marble-effect tables, velvet dining furniture
Contemporary/modern dining furniture
Consider carefully for:
Households prioritising easy-clean surfaces
Those wanting deep, lounging-style seating
Very small spaces (this is a substantial piece)
Price positioning: £1,850 — consistently described as "great value for money" and "worth the investment" by verified purchasers
Customer sentiment: 4.8/5 stars from 15 verified purchasers (14 × 5-star, 1 × 2-star)
What Makes This Different
This isn't just furniture — owners report it fundamentally changes household behaviour. The booth-style configuration creates a psychological "destination" that draws people to sit, stay, and talk. Unlike traditional table-and-chairs sets where people drift away after eating, the enclosed, comfortable bench encourages lingering.
The globally-inspired design aesthetic (the "Nomad" name reflects this) creates visual warmth that works across interior styles — from minimalist contemporary to maximalist eclectic. The rust velvet is particularly versatile, pairing with botanical prints, natural materials, and both warm and cool colour palettes.
