CREATE
Platform / Schema & Metadata
own data.
Searchpanel writes the corrected JSON-LD, meta tags, and llms.txt entries for every product, filling each field from your own specifications rather than guessing, then re-runs the checks that flagged the problem before anything ships.
✓ Filled from your catalogue
✓ Re-validated
✓ Applied per template
Generated markup
Waine89 Headphones · PRODUCT-V1 template
READY TO APPROVE
FIELDS FILLED
gtin13
FILLED
ratingValue
FIXED
weight
FILLED
batteryLife
FILLED
material
NOT SUPPLIED
JSON-LD GENERATED
{ “@type”: “Product”,
“gtin13”: “0812345678905”,
“aggregateRating”: {
“ratingValue”: 4.6 },
“weight”: “249 g”,
“batteryLife”: “50 h”,
“material”: — not supplied
}
!
1 value exists nowhere in your data. It stays flagged rather than guessed.
1 block clears 96 pages
Gemini
✦
ChatGPT
✦

Copilot
✦
AI Overviews
✦

Claude
✦
WHAT GETS GENERATED
Everything a machine needs, written for you.
Product schema, review markup, meta and social tags, llms.txt entries, and the validation pass that has to clear before any of it ships.
“@type”: “Product”
“gtin13”: “0812345678905”
“brand”: “Tekzone”
Product schema
A complete Product block with the fields engines need to match a SKU to a real item, each value pulled from your catalogue.
Before
“4.6 stars”
After
4.6
reviewCount
1,284
Review markup
Ratings written as numbers in valid aggregateRating markup, so an engine reads a score instead of a sentence.
TITLE TAG
Waine89 Wireless Headphones, 50h Battery, Multipoint
META ✓
OG:IMAGE ✓
TWITTER CARD ✓
Meta and social tags
The title tag, meta description, and social cards, written from the same facts as the listing rather than duplicating it.
# Tekzone
- Waine89 Headphones
- Pulse Mini Earbuds
+ 14 products added
llms.txt entries
A machine-readable entry per product in your llms.txt file, so AI tools that read it see your full catalog.
Rich results test
PASS
Merchant feed spec
PASS
Engine parse
PASS
Validated before it ships
Every generated block is re-run against the same checks that flagged the problem, and a block that fails cannot be approved.
HOW A BLOCK IS WRITTEN
Tools to help you fill the fields and prove they pass.
Every value is drawn from your own catalogue, generated as the markup each surface expects, and re-run against the same checks that flagged it in the first place.
THE FILL
See every empty field filled from your own catalogue.
A value is taken from a specification sheet, a lab result, or a connected marketplace listing. If it exists nowhere, it is reported as missing rather than filled with a plausible guess.
✓
Each value traced to its source
✓
Nothing invented to complete a block
✓
Genuinely missing values stay flagged
gtin13
0812345678905
Amazon listing
weight
249 g
Lab report
batteryLife
50 h
Spec sheet v4
material
not supplied
MISSING
!
material exists nowhere in your data, so it stays flagged instead of being guessed.
One value · weight 249 g
3 OUTPUTS
JSON-LD
“weight”: “249 g”
META TAGS
Weighs just 249 g
LLMS.TXT
weight: 249 g
!
Change the value once and all three regenerate together.
THE OUTPUTS
See every surface written from the same facts.
The JSON-LD, the meta tags, and the llms.txt entry are generated together from one set of values, so a product cannot describe itself three different ways across three places.
✓
One source, three outputs
✓
No drift between surfaces
✓
Regenerated together on any change
THE RECHECK
See it pass the check that flagged it.
Generated markup is re-run against the same validation that raised the problem, and because most faults come from one template, a single approved block clears every product built on it.
✓
Re-run on the same checks
✓
Applied across the template
✓
A failing block cannot be approved
PRODUCT-V1 template · 96 pages
PASSING
gtin13 present
✗ failed
✓ passes
ratingValue is a number
✗ failed
✓ passes
availability as a URL
✗ failed
✓ passes
!
One approved block cleared all 96 pages built on PRODUCT-V1.
5×
average lift in AI visibility within the first 90 days.
+32%
higher conversion on AI referral traffic.
2,847
buyer prompts tracked per brand, every day.
CUSTOMER STORY

How Herbs & Kettles strengthened discovery against larger tea brands.
Herbs & Kettles had a strong farm-to-cup story, but that differentiation wasn’t translating into enough digital visibility. Searchpanel revealed where the brand was missing from AI-generated recommendations, which competitors were being surfaced instead, and which product signals were limiting discovery.
Dr. Poorvi Chordia
Co-Founder · Herbs & Kettles

AD DRIVEN REVENUE
+214.7%
on flat ad spend
TACOS EFFICIENCY
+36%
▲ 103.3% ROAS
SALES GROWTH
+114.3%
quarter on quarter
FAQ
What does Schema and Metadata generate?
Product JSON-LD, review markup, meta and social tags, and llms.txt entries. Finding what is broken is Technical and Schema; this writes the correction.
Where do the field values come from?
What if a required value exists nowhere?
Is the output validated?
Can it run across the whole catalogue?
Do I have to deploy it myself?
How is this different from a schema generator?
revenue.
Find out which of your products AI names, which it ignores, and what it would take to change that.
Run a free audit
takes 2 minutes
No card needed · Cancel anytime · Read only, always