Intentwise pricing — and the unit mismatch that makes most comparisons meaningless

What is on the site today, stated without editorial

Checked on 20 August 2026. The pricing URL returned a 404. Nothing resembling a figure, a tier, a band or a minimum appears on the homepage or in the navigation, and the primary calls to action are Contact and Schedule a Demo. That is the whole finding, and it is a finding rather than an accusation — enterprise data platforms are routinely sold this way because the scope genuinely varies by how many accounts, marketplaces and data sources a buyer connects.

What the site does describe clearly is the product line, and that matters more than the missing number. The navigation names Intentwise Foundation as a data infrastructure and unification layer, an AI Gateway (MCP) that exposes the data to assistants such as Claude and Copilot, Intentwise Intelligence for analytics and visibility including Product 360 and an Amazon Marketing Cloud explorer, Intentwise Optimize for execution, and Intentwise DSP. The site describes coverage across 21 marketplaces and integrations spanning Amazon Ads, Walmart Connect, Google, Meta, Criteo, Instacart, Target Roundel and TikTok, and it addresses both brands and agencies.

Since the acronym collides and this page sits in programmatic territory: DSP here means demand-side platform, the software used to buy display, video and audio inventory programmatically. It is not Amazon's Delivery Service Partner programme, which is the delivery-van franchise business. If you arrived looking for route density and vehicle leases, this is the wrong page entirely.

This page is written by Dr. DSP, an Amazon DSP product from Full Circle, a full-service Amazon management company with more than $500M in managed revenue across 100+ brands. We do not publish a price either, which we will come back to rather than skate past.

Several starting prices are in circulation. None of them is on the vendor's site

Search this term and you will be handed confident figures. Different pages carry different starting points, different products attached to those starting points, and different assumptions about what is bundled. We are not going to reprint any of them, because reprinting a number we cannot see on the vendor's own page is how pages like this one go wrong.

There is a specific, non-cynical reason the circulating figures disagree here, and it is worth understanding because it applies to a whole cluster of vendors: the reviews and the price quotes predate the product. Intentwise today leads with a data layer, a model-context-protocol gateway and an observability stack. A summary written against an earlier, narrower positioning is not wrong so much as obsolete — it is quoting a price for a product with a different shape. The same pattern is visible across this category, where platforms have repositioned around data layers and agent interfaces while the review corpus kept describing the previous generation.

Two habits protect you, and both cost nothing:

• Date every figure you collect. A price without the date you read it is not evidence. If you cannot say which page and which day, do not put it in the spreadsheet.
• Check which billing tab you are reading, on any vendor including us. Monthly and annual columns sit side by side on most pricing pages, and comparing a per-month rate against an annual total manufactures a discrepancy that looks like bad faith and is actually a misread. We have made that exact error and had to pull a page over it.

Where a vendor publishes nothing at all, as here, the honest move is to write "quote only" in the cell and move on to the questions that make a quote comparable.

The comparison that breaks: price per seat against price per outcome

Most shortlists in this space contain two genuinely different purchases wearing similar language, and buyers compare their prices as though they were the same thing.

A data and analytics layer is priced by inputs: seats, connected accounts, marketplaces, data sources, refresh frequency, rows retained. You buy visibility. Someone on your side then has to decide what to do about what they can now see, and do it.

Managed execution is priced by scope: the media under management and the work attached to it. You buy decisions being made and actions being taken, and accountability for the result sits outside your building.

These are not competing products. Plenty of good operations buy both, and a data layer plus a capable in-house trader is a perfectly sound answer. But their prices are quoted in different units, so putting them in adjacent spreadsheet cells and picking the smaller number is a category error. It is the buying equivalent of comparing the price of a gym membership against the price of a personal trainer and concluding the gym is better value.

There is one conversion that does make them comparable, and it takes ten seconds. Divide any fixed quote by your monthly media budget. A fixed fee expressed as a percentage of the media it is supposed to improve is the only common unit in this market. Run it on every quote you receive, including ours. It will occasionally produce an uncomfortable answer — a modest-looking licence at a modest media budget can convert to a startling percentage — and the discomfort is the information.

Then add the line nobody quotes, which is the subject of the next section.

The cost of a data layer includes the people who read it

A dashboard has no opinions. Neither does a clean-room query result. The value of a data platform is entirely realised by a human deciding something because of it, and that human is a line item that never appears in a quote.

This is particularly sharp for Amazon Marketing Cloud work. AMC is a clean room: you write queries against event-level data and get aggregated answers back. It is genuinely powerful — it is where display and sponsored ads stop double-counting each other's orders, which is the single most useful thing an Amazon advertiser can establish. It is also a skill. Somebody has to know which question to ask, how to define the audience, how long a window to use and how to read the output without fooling themselves. Amazon makes AMC available at no cost to eligible advertisers, which means the licence was never the expensive part; the analyst was.

So build the real number before you compare vendors:

• Hours per week to keep the data layer configured, reconciled and trusted. Configuration drift is silent and constant.
• Hours per month to write, run and interpret the queries that change a decision — not the ones that fill a slide.
• Loaded cost of the person doing both, in the same currency as the licence.
• Time to first decision. How long between signature and the first budget move that happened because of the platform? If nobody can answer, that is the risk.
• What happens when they leave. Are the queries documented and portable, or in one person's head?

A platform that costs less and consumes two days a week of a senior analyst is frequently the more expensive option. The spreadsheet will never say so on its own, because the expensive half is on a different budget line.

If you are buying a data layer, portability is the product test

Here is a buying question this category has not caught up with. When the product was a dashboard, the evaluation was about the interface — could you find the view you wanted, could you schedule the report. When the product is a unified data layer with an agent interface on top of it, the interface stops being the thing you are buying. The rows are the thing you are buying. Which makes getting the rows out the sharpest test of whether the platform is a tool or a lock.

Put these to any vendor selling you a data layer, and put them in writing:

• Can we export the underlying tables, not just rendered reports, on demand and on exit — and in what format?
• Who owns the queries and models we build? If we leave, do they come with us?
• Does our Amazon Marketing Cloud instance remain ours? Clean-room instances are provisioned to the advertiser; make sure the arrangement reflects that.
• What is the retention period, and what happens to history if we downgrade rather than leave?
• If an assistant or agent queries this data, where does the query and the result go? A gateway is a new surface and deserves a straight answer about logging and retention.

And the three clauses that belong in every ad tech agreement regardless of vendor, ourselves included, because this market consolidates constantly: a change-of-control termination right, a price-protection period that survives an ownership change, and a data-portability commitment. Ask all three of everyone. The speed of the answer is itself a data point.

Who Intentwise suits better than we do, and what we do instead

The genuine strength is breadth of connection. A team running media across Amazon, Walmart Connect, Google, Meta, Criteo, Instacart, Target Roundel and TikTok has a real unification problem, and a platform whose whole job is to land all of that in one modelled place — and now to expose it to an assistant through a gateway — is solving a problem an Amazon specialist simply does not solve. If you are an agency reconciling dozens of accounts across marketplaces, or a brand whose retail media footprint is genuinely multi-retailer, that is the reason to buy and we are not a substitute for it. Buy the data layer.

What we do is narrower and differently accountable. Dr. DSP is Amazon DSP run as a managed product. The argument we make is about evidence rather than visibility: last-click reporting can tell you what was clicked, and it structurally cannot tell you what would have happened anyway — which is the entire question with display. So we run holdouts and matched controls, withholding display from a comparable set of ASINs or regions for an agreed window, and reconcile the result in Amazon Marketing Cloud so DSP and sponsored ads stop claiming the same orders. The measurement plan, the evidence behind a change and the rollback trigger are agreed before anything runs.

Commercially: no published price — a demo, the first 30 days free, and pricing set on the call against real media budget and scope, month to month, with Orbit included at no extra cost. On the narrow question of transparency, any vendor with a live rate card is ahead of us, and we would rather write that than argue around it. You set the autonomy level and can change it whenever you like; inventory risk, pricing, new products, new creative and stopping display spend always come to a human. Scoped reference, not a promise: across 30 advertisers in July 2026, the pull showed 6.04x return on ad spend, a $4.00 CPM and 20.1% new-to-brand.

One redirect worth making: if the gap you are trying to close is sponsored search performance rather than display measurement, Dr. PPC is the correct product and it publishes both its rate and its cap.