Articles · Market notes · September 1, 2026 · Radin Ahmadi · 4 min read
The words under the photo
Google answers the length question in its own documentation. The answer is too boring to sell, which is why nobody quotes it.

Ask how long a product description should be and somebody will give you a number. Two hundred words. Three hundred. Five hundred for anything you want to rank. The number moves depending on who is selling the advice, which is the first sign that nobody actually has one.
Google answers it directly in its own documentation, and the answer is dull enough that it almost never gets quoted. The SEO starter guide states that “The length of the content alone doesn't matter for ranking purposes (there's no magical word count target, minimum or maximum, though you probably want to have at least one word).” The parenthesis is Google enjoying itself, and it is also the entire position. Length is not a lever.
Where Google does hand you numbers
The place the guidance gets specific is the product feed, because a feed has to be read by a machine. The description attribute accepts 1 to 5,000 characters, and Google recommends putting the most important details in the first 160 to 500 of them, since a shopper has to click to see any of the rest. It asks you to describe only the product, and to keep out price, sale information, shipping detail, company names, comparisons to rival products, promotional language and capital letters used for emphasis.
The title attribute is capped at 150 characters, and Google notes that users usually notice only the first 70 or fewer. It wants the distinguishing details of a variant in the title itself, so a red shirt says red in the title as well as in the colour field.
Put those two specifications side by side and a shape falls out. The first seventy characters of the title and the first two or three sentences of the description are what a person reads. Everything after that serves the buyer who has already decided to care, and the machine building an index.
What actually goes wrong
Descriptions on small stores rarely fail for being short. They fail because the one fact a buyer needed is missing, and it would still be missing at three times the word count.
Baymard tested this against the largest sites in the world and found that 10 percent of the top sixty ecommerce sites fail to hold a consistent standard of description quality across their catalogue. In the sessions behind that figure the abandonment is very specific. A participant hunting for ingredients said “I would have thought [the ingredients] would have been under the description.” Another, comparing two chairs, said “There's a lot of information here, on the chair…I'm starting to discount [the other chair], because there's very little information.”
That second quote is the one worth pinning above a desk. The shopper did not leave because the writing was bad. They left because a competitor's description made a decision possible and yours did not.
The questions that decide a purchase are boring and entirely knowable. What is it made of. How big is it. Will it fit the thing I already own. How long does it last. Who is it for. A description that answers those in plain sentences beats one three times longer that is mostly atmosphere.
Say it the way it gets searched
Google's guidance on wording is one sentence, and for a small store it is the whole of keyword research. “Think about the words that a user might search for to find a piece of your content.” Nobody types artisan slow-poured vessel. They type unscented soy candle. Both can live on the page, and only one of them ever gets matched to a query.
The same guide warns that “Excessively repeating the same words over and over (even in variations) is tiring for users, and keyword stuffing is against Google's spam policies.” Use the buyer's words once, accurately, in the place they belong. Seeding them is a different activity with a different outcome.
The description you wrote at launch
There is a second failure that costs more than the first. The description was true on the day it was written. Then the supplier changed, the size chart shifted, the scent got reformulated, the bundle stopped including the spare wick. Nobody walks back through forty listings to check, because there is no morning of the week when that is the job.
The shopper did not leave because the writing was bad. They left because somebody else's description made a decision possible.
Structured data makes the shape of that drift easy to see. In Google's merchant listing markup the required properties are name, image and an offer carrying a price, while description sits among the recommended ones alongside brand, colour, category and availability. Meeting the required fields is cheap and automatic. The description is the field with no validator watching it, which is exactly why it rots quietly.
Listings are one of the areas Vantage carries. Descriptions written in the words people actually search, checked against what a product currently is rather than what it was at launch, and structured so a crawler and an assistant can both read them without guessing. Each rewrite arrives as a draft you read and approve, and nothing reaches the store until you say so. We wrote about the search half of this in Google is half the story now.
The honest answer to the length question is that a description should run exactly as long as the facts a buyer needs, in the order they need them, and not one sentence further. That is unsatisfying if you came for a word count. It is also the only answer that survives contact with a real person deciding between your product and somebody else's.
Sources
Next
The ads math is rigged against small stores
