Posted on

Machine-written product descriptions across vast retail catalogs

A catalog with twenty products can be written by hand. A catalog with twenty thousand products creates a different incentive.

Every item needs a title, description, features, materials, compatibility notes, dimensions, care instructions, and search-friendly language. That is exactly the kind of repetitive work automated text generation can accelerate. The danger begins when the system stops rephrasing supplied facts and starts filling gaps with plausible ones.

Shopify’s own documentation for its AI product-description feature, Shopify Magic, makes the distinction unusually clear. Merchants can provide a title and a few keywords, then generate a complete description. But Shopify also warns that generated copy can introduce product benefits or facts that the merchant never supplied, including details borrowed from similar products. Its guidance says merchants remain responsible for the accuracy of what they publish and should review generated text closely. See Shopify’s documentation on automatically generating product descriptions.

That warning gets more important as the catalog grows.

Plausible specifications are still invented specifications

A language model is very good at knowing what a product description usually sounds like. If the item is a jacket, the copy may naturally mention weather resistance. If it is a cable, the model may invent compatibility language. If it is a kitchen tool, it may add claims about dishwasher safety or materials because those details are common in similar listings.

The prose can sound more complete than the underlying record.

That creates a subtle reversal. Instead of the description being a readable version of verified product data, the description becomes a source of new claims that somebody now has to investigate after the fact.

The Federal Trade Commission’s general advertising guidance is boring but useful here: advertisers are responsible for express and implied claims, and material claims need a reasonable basis. Automation does not transfer that responsibility to the model.

The source record has to remain authoritative

The safest workflow is simple. Structured product data comes first: manufacturer specifications, measured dimensions, tested compatibility, ingredients, materials, warranty terms, and other facts that can be checked. Generated copy can then reorganize those facts into readable prose.

When the generated description adds something not present in the source record, it should be treated as an unverified suggestion, not as a discovered fact.

This matters because scale changes the consequences. One invented sentence on one listing is a correction. One invented attribute propagated across ten thousand SKUs becomes a catalog-level data problem.

Machine-written product descriptions are not inherently deceptive. They are a publishing tool. The problem begins when the smoothness of the language hides the difference between information supplied by the merchant and information guessed by the machine.

A huge catalog can be automated. Responsibility cannot.

Posted on

Multiply: a personal-sharing community redirected toward commerce

Multiply launched in March 2004 as a deliberately smaller kind of social network. Where MySpace and Friendster sold themselves on meeting new people, Multiply was built for the circle you already had: friends, family, coworkers. Each member got a personal homepage on their own subdomain, and posting a photo album, video, or blog entry meant deciding who was allowed to see it. Privacy was the product, not an afterthought.

Sharing with the people you already knew

The service made relationships the core object. When you added a contact, you also typed what they were to you — friend, family, coworker, business associate — and every post could be limited to one of those groups. A wedding album went to family; a happy hour invitation to coworkers. Multiply’s own 2006 FAQ pitched the payoff plainly: “by bringing everyone you know, in one place, you can share and discuss things like pictures, announcements and events.” That framing carried the site to millions of members, with particular strength in the Philippines and Indonesia, where it became one of Southeast Asia’s largest networks.

The sellers showed up first

What ultimately changed Multiply began on the ground, in the Philippines. Members started using the site to run small sales — clothes, gadgets, crafts — even though buying and selling violated its terms. These sellers called themselves Pinoy Multipreneurs. Multiply’s co-founder and CEO Peter Pezaris later recalled that when the company asked Philippine members to register what they used the site for, roughly half said they were sellers. “That signaled our shift to e-commerce,” he said. Multiply began building shopfronts and a Marketplace in 2010, claiming more than 70,000 merchants and 20 million monthly unique visitors.

Commerce took over the network

Later that year the South African conglomerate Naspers bought a controlling stake, and eventually replaced Pezaris with its own management. Multiply moved its base from Florida to Jakarta. In August 2012 its new CEO, Stefan Magdalinski, announced the end of the original product: from December 1, photos, videos, blogs, and messaging would be removed so Multiply could concentrate on being a marketplace for the two markets its owner cared about. “[We] have concluded that other Internet sites who are committed to social networking services will do a better job serving you than we can,” he wrote. The social network stopped, and the marketplace became the whole company.

What the shift cost the original members

The people who had built the site came along only in theory. Their albums, journals, and message threads had an expiration date, and the announcement pointed them elsewhere instead of moving them anywhere. The relaunch sputtered: sellers reported vanished listings, unclear orders, and slow payouts in the months before the end. In April 2013 Multiply announced its closure: the marketplace shut on May 6, and it ceased business operations by May 31. Its Indonesian unit later filed for bankruptcy, and users in Indonesia and the Philippines were pointed at the classifieds portals Tokobagus and Sulit.com.ph, both owned by Naspers.

The members whose family albums and hobby blogs made Multiply worth joining never got a home in its commercial successor; the archive was not carried forward. The domain was later reused by an unrelated company. Multiply is not a case of a community quietly aging out of the web. It is a case of one being traded away — the redirect led to a checkout counter, not a den.