Intelligence Node

Retail pricing used to move weekly. It now moves continuously, because competitors are running algorithms and shoppers can compare every listing in the time it takes to open a second tab.

The response is not to price faster by hand — it is to know what everyone else is charging, continuously, and for the right products. That second part is where these systems succeed or quietly fail.

Intelligence Node is built for that, at enterprise scale, and it puts a guarantee behind the matching accuracy that everything downstream depends on.

Intelligence Node: Prices At Machine Speed

• 99% product matching accuracy, SLA guaranteed — the number that decides whether any of the rest is usable.
• A ten-second refresh rate, described as the fastest in the industry.
• 1 billion+ products, 190,000+ brands, 1,400+ categories and 1.6 trillion+ historical price points.
• Four capabilities: price monitoring, digital shelf analytics, PDP content optimisation and assortment intelligence.
• An Omnicom company, with 100+ languages across six continents and a petabyte of data monthly.

Why Does Price Monitoring Need A Ten-Second Refresh?

Intelligence Node states a ten-second refresh rate, describing it as the fastest in the industry, alongside 34,000 pricing opportunities surfaced every minute.

The obvious question is whether that speed is real value or a specification arms race, and the answer depends entirely on your category.

In electronics, marketplaces and anything with algorithmic repricing, prices genuinely move within minutes and a daily snapshot describes a market that has already changed. There, latency is the product.

In categories where prices move seasonally — furniture, much of grocery, most apparel outside promotional windows — a ten-second refresh is measuring noise, and acting on it would produce churn rather than margin.

The honest framing is that refresh rate is a feature you should buy only if your category needs it. What everyone needs is accuracy, and that is a different specification entirely — which is why the matching guarantee is the more important number on the page.

Refresh rate is a feature you should buy only if your category needs it. Accuracy is the one everybody needs.

Matching Accuracy Is The Whole Problem

Intelligence Node states 99% product matching accuracy, SLA guaranteed, and it is the specification that determines whether anything else is usable.

Product matching means deciding that the item on a competitor's site is the same item you sell. That sounds trivial and is the hardest problem in the category, because retailers deliberately make it hard.

The same product appears under different names, in different pack sizes, with retailer-specific model numbers, bundled with accessories, in a different colourway, or as an exclusive variant that exists precisely so it cannot be compared.

A system that matches wrongly does not report an error. It reports a price — for a different product — and you reprice against something that was never comparable. Bad matches produce confident, actionable, wrong decisions.

That is why an SLA on matching is worth more than any dashboard feature, and why the question to ask a vendor is not how many products they cover but what happens contractually when a match is wrong.

Digital Shelf Analytics Explained

Beyond price, Intelligence Node covers share of search, listing accuracy and consistency, minimum advertised price monitoring, and PDP content auditing for both first-party and third-party sellers.

Digital shelf is the online equivalent of where your product sits in a physical aisle, and the analogy holds better than most retail metaphors. Being invisible in search results is the same as being on the bottom shelf at the back.

Share of search matters because it is a leading indicator. It moves before revenue does, which makes it one of the few retail metrics that gives you time to react rather than a post-mortem.

MAP monitoring is the enforcement half — checking that resellers are not advertising below the price you set — and it is a compliance function rather than an analytical one.

PDP content auditing across third-party sellers addresses a specific modern problem: your product listed by resellers with wrong descriptions, missing attributes and poor imagery, all of which reflects on your brand. Content generation tools such as Describely fix your own listings; this measures everyone else's.

• Ask what happens contractually when a product match is wrong, not how many are covered.
• Buy refresh speed only if your category actually reprices that fast.
• Treat share of search as a leading indicator and watch it weekly.
• Check whether coverage includes your specific competitors, not a global count.
• Decide who acts on a pricing alert before subscribing to one.

Reading Very Large Scale Numbers

The site publishes a striking set of figures: over a billion unique products, 190,000+ brands, 1,400+ categories, 1.6 trillion+ historical price points, a petabyte of data monthly, and 100+ languages across six continents.

These are impressive and they answer a question most buyers are not actually asking. Total coverage tells you the platform operates at scale; it does not tell you whether your competitors are in it.

Retail is intensely local. A company selling homeware in the Netherlands cares about six specific competitors, and a billion products worldwide is irrelevant if four of those six are not tracked.

The historical depth is the more interesting figure. 1.6 trillion price points means a long time series, which supports questions scale alone cannot answer — how a competitor behaves before a promotion, whether a rival always follows within a week, how elastic a category actually is.

So the useful reading is that scale figures establish credibility and the buying question remains specific: these categories, these competitors, these markets, refreshed at this frequency, matched at this accuracy.

It is also worth separating this from adjacent measurement problems that look similar. Knowing what competitors charge is not the same as knowing whether shoppers can find you, which is what visibility platforms such as Conductor measure, and neither tells you which marketing spend produced the sale — the question Cometly exists to answer.

Retailers frequently buy one of the three and expect answers from all of them. Pricing, discoverability and attribution are separate data problems with separate vendors, and conflating them is how a platform gets blamed for not answering a question it was never watching.

An Omnicom Company And What That Signals

The footer identifies Intelligence Node as an Omnicom company, which is a material fact for a buyer rather than a footnote.

The straightforward reading is stability. Ownership by a large marketing and communications group means resourcing, longevity and the kind of compliance posture enterprise procurement asks about.

It also situates the product strategically. Retail pricing and digital shelf data are valuable to advertising and commerce businesses, so the platform now sits inside a group with adjacent interests in retail media and brand measurement.

For most buyers that is neutral to positive. For a retailer that competes with other Omnicom clients, or that is careful about where its commercial data sits, it is a question worth asking directly during procurement.

The footer also carries a notice asking visitors to trust only communications from the company's own domain, and stating that it works with no third parties for sales or recruitment — a sensible warning to publish, and worth heeding if you are contacted by anyone claiming to represent them.

Evaluating A Retail Intelligence Platform

Retail data contracts are long, expensive and hard to exit, and the evaluation usually focuses on the wrong things — dashboards rather than data quality.

This sequence tests the parts that determine whether the platform is usable at all.

1. List your actual competitors and markets, and ask for coverage of exactly those.
2. Supply fifty of your hardest products — bundles, variants, retailer exclusives — and check the matches by hand.
3. Ask what the matching SLA obliges the vendor to do when accuracy is missed.
4. Confirm your category actually moves fast enough to need the refresh rate.
5. Name the person who acts on a pricing alert before the contract starts.
6. Establish how data leaves the platform if you do not renew.

Retail Data Terms Worth Knowing

The vocabulary these platforms use, and what each term decides.

undefined

undefined

undefined

undefined

undefined

undefined

Intelligence Node: Scale And Ownership