For online stores
Your catalog is the research corpus. We write from it, not around it.
Most store blogs are written by someone who has never seen the catalog. Vupie reads your products, your attributes and your reviews before it plans anything, so a buying guide names the variants you actually stock and links to the collection that sells them.
Nothing is recommended that you cannot sell
Every product mention resolves against the live catalog before publish
Draft
“For everyday use we would start with the linen wide-fit shirt, and pair it with the oak serving board or the small ceramic vase.”
Linen wide-fit shirt
Matched, 6 variants in stock
Oak serving board, 40 cm
Matched, linked to the collection
Ceramic vase, small
Discontinued, recommendation rewritten
Example of the resolution pass, not a real catalog. The same check runs again on the refresh pass, so a live article never keeps pointing at a product you have retired.
Written from the catalog
Every variant checked.
Every link resolved.
Then it publishes.
A buying guide that recommends something you discontinued costs you more than no guide at all. Product links are matched against the live catalog before publish, and rechecked on every refresh pass.
Business analysis
What we read before we plan anything.
The 90 day plan is not a keyword list bought from a tool. It is built from what is already connected to your account, which in e-commerce means these four things first.
Your product catalog
Titles, attributes, variants, price bands and what is in stock. A guide that recommends a variant you discontinued is worse than no guide, so the plan is built against the live catalog and rechecked before publish.
Queries you already almost rank for
Search Console positions 8 to 25 on category and comparison terms. Those are the ones a well-sourced article can move, and they get scheduled first.
Your review corpus
Reviews are a free record of what buyers ask before they buy, and of the objection that loses the sale. Both become article briefs.
Pages with traffic and no revenue
Analytics landing pages that pull sessions but no add-to-cart usually answer the question and then abandon the reader. Those get an internal-link and intent pass instead of a new article.
Where the plan aims
The four clusters that carry
e-commerce traffic.
Priority goes to queries you already sit close to, because those are the ones a well-sourced article can actually move.
Category buying guides
"best [category] for [use case]", "what to look for in [product]"
Comparisons
"[product] vs [product]", "[material] or [material]"
Fit, sizing and compatibility
"does [part] fit [model]", "[size] chart", "is [x] compatible with [y]"
Care and after-purchase
"how to clean [material]", "[product] not working", "how long does [product] last"
The output
What actually lands on your site.
Buying guide for a collection you want to grow
Every recommendation points at a variant you stock, and the internal links go to the collection page, not the homepage.
Comparison between two things you sell
Both sides sourced. Where the comparison is against a competitor product, the claims come from that vendor's own published spec, or the line is cut.
Compatibility and sizing reference
Built from your attributes, so it stays right when the catalog changes and gets a refresh pass when it does not.
Care and troubleshooting piece
The article that keeps the return from happening. Sourced to manufacturer documentation, and to your own returns data when you let us publish it.
Two or three a week
The questions only you can answer.
Research retrieves what is public. It cannot retrieve what only exists inside your business, and guessing it is exactly how AI content gets caught. So before writing, Vupie sends the two or three questions it cannot answer to you by SMS or Slack.
Answering takes a minute, and the answer becomes the part of the article nobody else can publish. In e-commerce, they usually look like this.
Is the spring restock on the linen range confirmed, or should the guide avoid it?
Your return rate on the wide-fit line is unusually low. May we publish the number?
Which of the two suppliers behind this range may we name?
Answered
The model may cite only from the retrieved source list plus what you answer here. Anything else is refused.
The gate
What we decline in e-commerce.
Worth reading before you sign up rather than after. A refused brief frees the slot and you see the reason, but if most of your topics sit on this list then Vupie is the wrong tool for you and we would rather say so now.
Supplements, skincare claims and anything health-adjacent
Health topics are blocked on the default pipeline. If half your catalog is supplements, say so before you sign up, because we will decline those briefs rather than write around them.
Invented social proof
No review counts, ratings or "shoppers love" lines that cannot be read out of your own data.
Competitor rankings we cannot source
A "best of" that ranks other stores needs a published, retrievable basis for the ranking. Without one, the piece is refused rather than padded with opinion.
Publishing
The usual setup here.
Shopify posts to your store blog with theme styling and product links intact. WooCommerce stores go through the WordPress integration.
Shopify
Native APIPosts to your store blog through a custom app, with theme styling and product links intact.
How it connectsWordPress
Native APIPublishes through the REST API with categories, tags, author and featured image set for you.
How it connectsBigCommerce
Native APIStore-blog publishing with meta fields and URLs handled.
How it connectsThirteen destinations in total, including a webhook and a hosted blog for sites with no usable CMS. See every integration.
Questions
What e-commerce teams ask.
That is the failure mode we plan against. Commercial category terms stay with the category page, articles take the informational half of the cluster and link down into the category. Where an article and a category page do start competing, the article is the one that gets adjusted.
No. Product structured data stays exactly as your platform emits it. Vupie adds Article schema on the articles it publishes and nothing else.
Articles are checked against the live catalog before publish. For pieces already live, a discontinued variant is picked up on the refresh pass and the recommendation is rewritten rather than left pointing at a dead PDP.
Other industries
Connect the site. See the plan
before anything publishes.
The business analysis runs first and produces a rolling 90 day calendar you can see and change. Nothing goes live until you have looked at it.