Product data intelligence
Every search and recommendation tool answers, because answering is all they can do. Pulse measures whether an answer is supported at all, and when it is not, tells you what is missing and what it would cost to record it. One category export in, a ranked repair list back in 48 hours. Not a PIM, and it replaces nothing.
94 of 216 products a shopper still cannot tell apart
347 fields are blank across this shelf. A shopper cannot filter on any of them, and neither can your search.
| Name | SKU | 1Screen size80 missing | 2Memory60 missing | 4Processor162 missing | 3Storage55 missing |
|---|---|---|---|---|---|
| LG 16Z90TS-G | A1B2C3D4 | 16.0" | Missing | Missing | 512GB |
| Spectre x360 | E5F6G7H8 | Missing | 16GB | Missing | Missing |
| ThinkPad X1 Carbon Gen 9 | I9J0K1L2 | Missing | 32GB | Missing | 1TB |
| Razer Blade | M3N4O5P6 | 15.6" | Missing | Missing | Missing |
| MacBook Pro | Q7R8S9T0 | Missing | 18GB | M3 | 512GB |
| MSI GS66 | Y5Z6A7B8 | Missing | Missing | Missing | Missing |
Every blank cell is a question your catalogue cannot answer.
readiness 94 / 216What this looks like
Consumer goods
A parent is buying washing powder for a baby's skin.
Ecommerce and electronics
A designer is buying a laptop and needs the screen big enough to work on.
The second example is a shelf we measured. The first is the same failure in a category where nobody has measured it yet.
The blind spot
Whatever you sell, one attribute decides it. Usually it was never a field.
Not our numbers
Where the budget goes
Both plans cost a quarter. Only one of them changes what a shopper can choose.
Fix the emptiest field
Fix the deciding field
From the shelf we measured. Screen size came top in all 12 runs; positions two to four moved, and we would rather say so.
One category export. A ranked list of what to fix, in 48 hours.
Request a Decision AuditNo code. No tag. No customer data.
How it works
A CSV or JSON export. Product name, description, category, price, and whatever attributes you already record. No code, no tag, no access to your systems.
Within 48 hours: the ranked list of what to record first, the specific products missing each field, where each missing fact actually lives, and your decision-readiness number.
You may not need us for the fix at all. When it is done we recompute on the repaired catalogue and show you exactly what moved. No traffic, no test, no waiting.
Why Pulse
Customised
Scored against your category's buying criteria, not a generic completeness rule.
48 hours
One category export in, a ranked repair list back, two days later.
Real time
The optional runtime ranks on the session in front of you, with a held-back control arm.
Your data
No customer records, no order history, no personal data of any kind is needed to run the audit.
The runtime infers from what happens in the current visit. It does not need an identity to work.
Your catalogue is never pooled and never sold, and nothing crosses a tenant boundary. If we ever pool aggregate intent across merchants it will be opt-in, it will be counts rather than records, and we will ask you first.
Including the catalogue export itself, at any point, on request.
The audit runs on the first row alone. The rest is not collected, so there is nothing to leak or delete.
Pick the shelf where you suspect shoppers struggle most. You get the ranked list, the products affected, and half an hour with whoever owns merchandising to walk through it.
No obligation at any step. And if your catalogue turns out to be fine, we will tell you that too.