Flywheel Digital reviews — the honest answer is that a client review corpus does not exist

Who Flywheel is now, which is not who it was

Flywheel Digital grew up as the digital commerce arm of Ascential. Omnicom agreed to acquire it in October 2023 and announced the closing on 2 January 2024; Omnicom's own release did not disclose a price, while contemporaneous press reports put the value at roughly $835 million. Duncan Painter, previously chief executive of Ascential, leads Flywheel as a practice area inside Omnicom.

What it sells today, per its own site: a Commerce Cloud software suite spanning retail media, retail operations and market intelligence, alongside managed services covering media execution, retail excellence, content and creative, and consulting. Named capabilities include paid search, programmatic display and video, Amazon Marketing Cloud work, brand protection and supply chain. The scale claims on the site are substantial — more than 4,500 clients, over 400 digital marketplace partnerships, partnerships with nine of the ten largest global marketplaces, over $10B of media spend data processed annually, and more than 50 of the top 100 global CPG brands. No pricing is published; the call to action is a contact form.

A definitional note, because this page uses the acronym repeatedly. DSP means demand-side platform — programmatic buying of display, video and audio. Amazon's Delivery Service Partner programme is a parcel-delivery franchise and has nothing to do with advertising. This page is written by Dr. DSP, an Amazon demand-side platform product from Full Circle, a full-service Amazon management company with more than $500M in managed revenue across 100+ brands. We compete for some of the same accounts, which is a reason to be careful here rather than a reason to be quiet.

Why the reviews you want are not there

Type this query and the first page fills with employer-review sites and a thin software listing or two. There is a reason, and it is not sinister.

• Enterprise services do not generate review volume. A self-serve tool with fifty thousand subscribers produces a review corpus automatically. A managed service sold through procurement to a few thousand large brands does not — the buyers are contractually cautious and rarely post publicly.
• The buyer is a committee. Star ratings encode one person's experience. Enterprise commerce engagements involve brand, media, supply chain and finance, and their views frequently diverge.
• Scale hides variance. An organisation serving thousands of clients across dozens of markets delivers a range of experiences, and a single aggregate score would flatten exactly the information you need.

We will say plainly what we are not going to do here: grade a competitor on its staff satisfaction scores. Employer reviews measure what it is like to work somewhere, which is not what you are buying, and any organisation at that headcount will have a mixed board — ours included, and every agency in this category too. Using it as a proxy for the quality of work on your account would be unfair and, more to the point, useless to you.

There is one legitimate signal buried in that theme, and it is worth extracting rather than inferring: continuity on your account. Do not guess at it from a public rating. Ask directly, in the pitch, and ask for numbers.

What Flywheel is genuinely strong at

We would recommend them for certain briefs without hesitation, and we should be specific enough that a current client recognises the description.

• Global multi-retailer reach. Very few organisations can execute across hundreds of marketplaces in many countries with local retailer relationships in place. If your problem is thirty markets rather than one, that footprint is the product and it is genuinely hard to replicate.
• Retailer relationships and joint business planning. The commercial conversation with a major retailer — terms, promotions, content standards, media commitments — is a different discipline from bidding, and a group at this scale has people who do it every day.
• Data breadth. Processing media spend and product data at their claimed volume creates benchmarking most advertisers cannot buy any other way.
• Sitting inside a holding company. If your brand and retail media budgets need to be coordinated with the rest of your media, having commerce inside the same group as your brand agencies is a structural advantage rather than a coordination problem.
• Clean-room capability. They name Amazon Marketing Cloud work explicitly, which puts them ahead of the many partners who still measure display on platform reports alone.

If your brief matches two or more of those, they belong on your shortlist and we would say so on a call.

How to evaluate a partner this size without a review corpus

Replace the missing reviews with reference calls you design. Ask for three referees with a brief that fits your shape — same category, comparable spend, similar market count — and run each call to a script. These are the questions that produce information rather than testimonials:

• Who has been on your account for the last eighteen months? Names and roles, and how many have changed. Continuity predicts outcomes more reliably than anything else you can ask.
• What is the seniority of the person who touches the account daily? Pitch teams and delivery teams are frequently different people. Ask to meet the second group.
• What went wrong, and what happened next? Every long engagement has an incident. The recovery tells you more than the highlight reel.
• How are decisions documented? Ask whether they receive a change log — every change made, with the reason and the expected effect — and ask to see a redacted page of one.
• What did they stop doing? Partners who only ever add tactics are not really optimising. A good answer names something switched off and why.
• Who owns the platform seats and the data? Establish that your Amazon DSP seat, your clean-room instance and your campaign history are yours and portable, in writing, before you sign anything.
• What is the notice period? Ask the referee what leaving would look like, not just the vendor.

Seven questions, three calls. That is a better evidence base than any star rating would have given you.

The question to put to them, and to us

Whatever partner you choose, there is one thing worth being stubborn about, because it is where display budgets quietly go to die.

Last-click attribution cannot prove incrementality. It identifies the touch nearest the purchase, which is a different question from whether the purchase needed the advertising. Display is the format where that gap is widest: retargeting reliably posts the strongest reported return in any account, because it puts impressions in front of people already on their way back. Optimise hard toward reported return and the machine will buy more of exactly that, producing an improving chart and a flat business.

What settles it is a design, not a dashboard. Withhold display from a matched set of regions or ASINs, run it for a defined window with a stated success condition, and reconcile in Amazon Marketing Cloud, where DSP impressions and sponsored-ads events sit in one clean room instead of separately claiming the same order. Amazon makes AMC available free to eligible advertisers, so the constraint is analyst time and willingness rather than budget.

Flywheel names AMC capability on its site, so this should be a comfortable conversation. Ask for the design in the proposal: what is withheld, from where, for how long, and what result would count as a failure. Ask us the same thing. A partner who cannot write that down is asking you to take display on faith.

Who they suit better than we do, and what we offer instead

Choose Flywheel over anything we sell when the brief is global and multi-retailer: many countries, many marketplaces, retailer negotiations alongside media, and a need for commerce to sit inside the same holding company as the rest of your marketing. That is a real advantage and we would be the wrong answer to it.

Choose us when Amazon is effectively the business, when you want the same senior attention on a mid-sized account that a very large group reserves for very large accounts, and when the open question is whether display is adding sales rather than reporting them. Dr. DSP is Amazon display run as a managed product. Every proposal arrives with its evidence, its measurement plan and its rollback trigger. You set the autonomy level — everything waiting on your approval, routine work automatic with larger moves queued, or fully autonomous inside agreed guardrails — and you can change it whenever you want. Inventory risk, pricing, new products, new creative and the decision to stop spending on display always come to a human at every level. There is no published price: a demo, the first 30 days free, priced on the call, month to month, with Orbit included at no extra cost.

In place of reviews we can offer a reference book, scoped honestly. Across 30 advertisers in July 2026 — not the whole book — a direct API pull showed 6.04x return on ad spend, 78.4 million impressions at a $4.00 CPM, a blended $1.42 cost per click, $5.49 cost per acquisition across 57,137 attributed purchases and 20.1% new-to-brand. One month, Amazon-attributed, no guarantee attached to any of it. We never publish a DSP client's name, which is a constraint we accept as the price of being trusted with the accounts.

Two redirects worth making. If the account's real problem is sponsored search rather than display, Dr. PPC is the right product and it publishes its price. If you want a named agency team and a written scope of work rather than a product subscription, reMKTR sells the same media buying agency-side.