After the ad money is spent, the trouble is often that two sets of numbers do not match: marketing has the conversion numbers from the ad platform, sales has stage progression in the CRM, nobody can explain the middle, and next quarter's budget has to be set between those two sets of numbers.
What Cometly sells is connecting that stretch. Its About page says it is a marketing attribution platform for B2B SaaS that helps operators connect ad spend to the sales pipeline and revenue and prove ROI. About Us | Cometly[1] What it wants to connect is the money already spent: matching ad spend to the sales pipeline and revenue.
The concrete actions are: install Pixel and server-side tracking on the website, recording an entry each time a page loads; connect accounts from major ad platforms such as Meta and Google; then match the sales pipeline and closed-won outcomes in the CRM against those records to do multi-touch attribution. Pricing | Cometly[2] The trouble is not installing code; it is getting CRM events and ad records onto the same basis. The product takes over the reconciliation piece; how accurate the connection is still depends on the person wiring it up.
That is why the pricing page has a sentence rarely seen: no free trial is offered. The reason is also in the FAQ—attribution only becomes useful after the CRM and ad platforms are connected; to get started, Professional Onboarding assigns a dedicated expert to wire the tech stack correctly from day one. [2]
The entry point is Book demo or Get started. At the bottom of the About page, it says the team will do a personalized walkthrough and within 30 minutes let you see your own data running; the pricing page describes the 30-minute call as: sizing your pageview, walking through your tech stack, and giving a quote on the same call. [1] · Pricing | Cometly[2] The quote is written into the same call, and the sales sequence is to look at the data first, then discuss price.
An AI That Pins Answers to Structured Data
The Core plan has an item called Agent. What is worth looking at is not “it has AI,” but the boundary it draws for itself.
The FAQ on the Agent page is direct: it uses an LLM, but does not do free-text retrieval; instead it generates queries against the attribution layer on structured Cometly data, and answers reconcile with the dashboard; every answer cites the underlying metrics, time window, and filters, and can be opened to see the corresponding rows; results can be saved as dashboard components, scheduled summaries, and shared with the team. [3]
What this approach buys is another thing: a number that can go back to the dashboard, Stripe, and the CRM for another check. Attribution tools often fear not miscalculation but that nobody trusts them. The same layer is also open to Claude, GPT, and self-built agents, queried through the MCP server; MCP Access is included in all plans. [3] · Pricing | Cometly[2] The site says “answers reconcile with the dashboard.”
Charging by Pageview, With Overage That Does Not Lock Attribution
Cometly does not publicly disclose list prices for Core or Enterprise; pricing is tailored to the customer's current monthly pageview, with monthly or annual payment, and annual is marked as saving 20%. A pageview is defined as each load of a website page that is recorded by Comet Pixel or a server-side event; the official estimate is that one visit typically produces about 1.5 pageviews. When the quota is exceeded, Cometly says it does not limit attribution; instead, it notifies at 90% of the tier and automatically upgrades to the next tier at the start of the next billing period. [2]
The two tiers have clear functional differences: Core includes attribution, Agent, the Cometly API, and MCP; Enterprise connects attribution data into the company's own data warehouse and internal systems, plus SSO, a dedicated solutions engineer, and custom development. Pricing | Cometly[2] These features correspond to needs where data must return to the company's own systems and permissions must go through SSO; the site does not say who actually pays more for this. Both tiers include connections to major ad platforms and the server-side Conversion API; connections are included in the plan, and page loads are what is metered.
Image: Source TrustMRR[4], accessed 2026-11-10; original interface retained.
273 Subscriptions, and the $2 Billion It Has Handled
The third-party site TrustMRR records: MRR $196,525, 273 active subscriptions, historical total revenue $10,209,207, with data last updated on 2026-11-10; the same page notes that the Stripe API key has expired. Cometly revenue: $221,393 in the last 30 days | TrustMRR[4] The page title says $221,393 in the last 30 days, which is not the same measure as the MRR in the body, and the page does not explain the relationship between the two. The page does not state the full basis for refunds, bad debt, and revenue recognition.
MRR is not profit. Without cost data for cloud infrastructure, staff, sales, and onboarding services, whether the company makes money cannot be seen. The 273 subscriptions are also only a total; with no customer distribution, averaging says nothing about any single customer.
Another set of numbers comes from the site: the platform has tracked more than $2 billion in ad spend ($2B+). This is ad spend that has passed through it, not Cometly's own revenue. About Us | Cometly[1] Put the two numbers together and there is something interesting—it is responsible for matching huge ad budgets to revenue, while its own bill hangs on pageviews.
The People Who Ran Ads First Left the Problem in the Ads
The About page lists founders Grant Cooper and Matt Pattoli, with Max Minin as CTO. About Us | Cometly[1] In the episode summary for an October 2024 ClickFunnels Radio episode, Cooper talks about moving from affiliate marketing selling products to doing data and analytics, about advertising on Facebook and Google, and about why tracking matters. [5]
This path—from selling products and running ads to doing data and analytics—at least explains where the product starts: it does not start from “data platform,” it starts from “which ad brought in revenue.”
Peers, and the Mismatch Between Billing and Value
HockeyStack's homepage describes itself as a modern attribution platform covering first touch to closed-won, unifying marketing and sales data into one actionable source of truth. AI GTM - B2B Revenue Data Intelligence Platform | HockeyStack[6] Both place themselves in the position of an attribution platform; what overlaps is the narrative from first touch to closed-won; what can be put side by side is that positioning, while price and accuracy cannot be compared.
Back to billing. Charging by pageview, while the value sold is “how much revenue ad spend actually brought in”—these two things do not always point the same way. Imagine two customers: one spends a lot on ads each month, has very light landing pages, and does not get much traffic; the other has a small ad budget, many on-site pages, and runs in a content-heavy way, with high pageviews. The latter may pay more, while having less ad spend to explain. Pageviews are easy to measure and low-dispute, but what they meter is not the money this company claims to explain.
Attribution tools have never really sold data; they sell permission for “this number can be taken into a meeting.” No trial, pinning AI answers to structured data—these choices all push toward “trustworthy, checkable,” at the cost of giving up “easy to get started”; overage not locking attribution, and automatic upgrade after a 90% notification, look more like not letting usage stop attribution from running.