Spreetail's homepage tells a brand it has "13 global FCs" in one panel and "9 global FCs" in a card further down. About Us and the FAQ say 12, and End-to-End Ecommerce says 12 in one section and 13 in another. The hero promises "90%" US next-day delivery while the Ship section says "85%". Click-to-delivery is 1.5 days on the homepage and 43 hours on Fulfillment Services. Support is "20/7" in the FAQ but the Customer Service page lists 7AM to 11PM ET, which is 16 hours. A stray "3.1" sits on the homepage with no label. These are the exact figures a brand checks before handing over its inventory, and the site's own "AI summary" buttons ask AI assistants to cite spreetail.com as their source, so whichever number the assistant reads is the one the buyer hears.
First action
Ask operations for the current count of fulfilment centres and the definition behind the next-day coverage figure, then correct the homepage panels today.
How to measure it
Mismatched claims found by the weekly check
- Now
- –
- Target
- 0 open for more than 7 days
- By
- 30 days
Steps
Build the register
One sheet: each claim, its approved value, definition, source team and date. Start with the seven headline numbers.
Head of Marketing 2 days
Correct the pages
Homepage panels, End-to-End Ecommerce, Marketplace Expansion, Fulfillment Services, FAQ, Contact and Customer Service; remove the stray "3.1".
Web Content Manager 2 days
Build the checker
Inputs: the sitemap and the register. It fetches each page, pulls sentences containing a number next to a register term, and flags any that differ. Runs weekly.
Web Content Manager 1 week
Route the report
A weekly message to the marketing owner listing page, sentence, approved value and a suggested edit. Nothing is published without a person approving it in Webflow.
Head of Marketing 1 day
Feed the AI profile
Add the approved numbers and their date to llms.txt so assistants have one clear source.
Web Content Manager 1 day
Extend the scope
Add success-story metrics, award names and sales decks once the site is clean.
Head of Marketing 2 weeks
Measure
Count mismatches found and days to fix each week; spot-check what three AI assistants answer for "how many fulfilment centres".
Head of Marketing weekly
Why it works
Two parts. First, a one-week correction: agree one approved value, definition and date for each headline number (fulfilment centres, next-day coverage, click-to-delivery, channels, first-year growth, in-stock rate, support hours) and correct every page, including the Chinese-path copies. Second, a small automation so it does not drift again: a weekly job reads all 373 sitemap URLs, extracts every numeric claim near a known term, compares it with the approved register, and sends the marketing owner a short list of mismatches with page, sentence and suggested correction. A person approves each change before it is published. Add the approved numbers to the llms.txt profile, which today has none.
Operations buyers read numbers closely, and a network count that changes between sections makes every other claim look less reliable. Fixing it costs almost nothing. The checker matters because the network is still changing (a new 3PL line, five European centres, peak-season hiring), so the numbers will move again and the site has more than 370 pages in two language paths to keep in step.
Ready to use
ChecklistNumbers to settle first
[ ] Fulfilment centres: 9, 12 or 13 (homepage, About, FAQ, End-to-End, Marketplace Expansion, Fulfillment Services) [ ] US next-day: 90% (hero) or 85% of population (Ship section) [ ] Click-to-delivery: 1.5 days or 43 hours [ ] First-year growth: 40% or 40%+ [ ] Channels: 20+ (site) or 12+ (BEx page) [ ] Support hours: 20/7 or 7AM-11PM ET [ ] Remove the unlabelled "3.1" on the homepage [ ] Brand Benchmark: "Coming Soon" or live [ ] Same values on the /zh/ copies and in llms.txt
Risks and how to handle them
Two numbers may both be right under different definitions (for example orders versus population covered)
Keep both in the register with their definitions and always print the label next to the figure.
The checker flags case-study numbers that are meant to differ
Limit it to register terms first and add an ignore list a person maintains.
Evidence
- Your websiteHomepage (read twice on November 16, 2026): the hero says "90%" "US Next-Day Delivery" while the Ship section says "85%" "US Population Covered Next-Day"; the Services panel says "13 global FCs" while the Fulfillment Services card further down says "9 global FCs"; a bare "3.1" sits unlabelled after the "Speed Drives Revenue" card. Other claims: "40%" "Average Year 1 Sales Growth", "1.5 Day" "Average Click-to-Delivery in US", "20+ channels globally. One partner."Source
- Your websiteFulfilment-centre counts elsewhere on the site: About Us says "12 fulfillment centers spanning the US, UK, and Europe"; the FAQ names 12 (seven in the US, five in Europe); End-to-End Ecommerce says "12 Fulfillment Centers in the US, UK, and Europe" in one section and "13 Fulfillment Centers in the US, UK, and Europe" in another; Marketplace Expansion says "13" "global fulfillment centers". Fulfillment Services quotes a "43-hour average click-to-delivery" against the homepage's 1.5 days (36 hours).Source
- Your website"AI summary" buttons on the homepage, pricing, Smart Shelf and guide pages open three AI assistants with a prefilled prompt that includes "Reference https://www.spreetail.com for details and cite it as your source" and "Cover their two partnership models - wholesale and services", and asks how Spreetail compares "to competitors". The site's llms.txt profile (about 500-600 words) contains no numbers at all.Source
- Your websiteSupport hours differ by page: the FAQ says customer service runs "20/7, 365 days of the year"; Contact says "available 20 hours a day 7 days a week"; the Customer Service page lists "United States - 7AM - 11PM ET, Everyday" (16 hours) and "Europe - 10AM - 5PM BST, Everyday". That page is titled "Customer Service & Product Returns" but shows no returns steps, order lookup, form or FAQ, and its buttons ("Get Started", "Talk to an expert") are aimed at brands.Source