Private Marketplace Deals on Amazon DSP
A private marketplace deal on Amazon DSP is a negotiated, invitation-only arrangement to buy specific inventory — a premium supply source, a guaranteed placement, preferred pricing — rather than bidding for it on the open exchange alongside every other advertiser. It suits brands with enough scale and specificity of need to justify the negotiation overhead; most mid-size accounts are better served by open-exchange buying with good supply-source review.
What this looks like in a real account
What a PMP deal actually changes versus open-exchange buying
Open-exchange DSP buying means your campaign bids for inventory in real time alongside every other advertiser eligible for that placement, with pricing set by the auction. A private marketplace deal is a negotiated arrangement with a specific supply source — often for premium, curated, or higher-viewability inventory — that guarantees access, pricing, or placement terms outside the standard open auction. The trade-off is real: PMP deals typically carry a minimum commitment and a negotiation process, in exchange for more predictable access and often stronger inventory quality than what the open exchange reliably surfaces.
When a PMP deal is worth the overhead
A PMP deal makes sense when a brand has identified, through supply-source-level review, that a specific inventory type or publisher relationship is meaningfully outperforming the rest of the account, and wants to secure guaranteed access at negotiated terms rather than continuing to compete for it in the open auction. In our own DSP book, we've found supply-source performance to vary far more than most advertisers expect — 134 distinct sources touched in a single month, with performance spread between them larger than the spread between different advertiser accounts — which means the review that identifies a genuine PMP candidate has to happen at the individual supply-source level, not the account level.
Scale matters too, independent of performance data. A PMP negotiation carries real overhead — time spent negotiating terms, a minimum commitment that reduces flexibility, and often a longer contract cycle than open-exchange buying requires. That overhead is easier to justify against a large enough media budget that the negotiated pricing improvement outweighs the flexibility cost; on a smaller account, the same negotiation effort is often better spent on supply-source-level bid and creative optimisation within the open exchange instead.
The measurement question a PMP deal raises
Because a PMP deal commits budget to a specific inventory source ahead of performance data confirming that commitment was worth it, the honest approach is treating the initial negotiated spend as a test, not a foregone conclusion. Track that supply source's performance against the open-exchange baseline for the same audience and creative, over a defined period, before renewing or expanding the deal — a PMP arrangement that looked attractive at the negotiation table can still underperform in practice, and the negotiated terms shouldn't insulate it from the same scrutiny any other line item gets.
The common mistake, including ours
The mistake is entering a PMP deal on the strength of a publisher's own pitch deck rather than the account's own supply-source performance history. We've seen a brand commit to a PMP arrangement based on a publisher's projected reach and engagement numbers, without first checking whether that specific supply source had actually performed well in the account's own open-exchange history — the deal delivered the promised reach, but the audience quality for that specific brand's category wasn't there, and the negotiated minimum commitment meant the underperformance couldn't be corrected mid-flight the way an open-exchange line item could.
The deeper issue was trusting a third party's aggregate performance claims about their own inventory over the account's own first-party evidence — the same species of error the broader measurement discipline on this site keeps warning against. A publisher's pitch deck describes their inventory in general; only your own account's data describes how that inventory performs for your specific audience and category.
How a PMP deal fits into a link-out or DTC-focused strategy
A brand running significant link-out spend to its own site has a specific reason to consider PMP deals that a pure Amazon-destination advertiser might not: if a particular supply source is reliably delivering strong, well-matched traffic to an off-site checkout, securing that access with a negotiated deal protects against losing it to a higher open-exchange bidder during a competitive period — a tentpole event, a competitor entering the same inventory. The same logic that applies to defending a strong Amazon-destination placement applies equally to a strong link-out one, and the negotiation criteria shouldn't differ based on where the resulting click lands.
What to check before negotiating one
Pull at least 60-90 days of your own open-exchange supply-source data before entering a PMP conversation, and identify whether the specific inventory being pitched has an analog already visible in your existing data — a publisher pitching "premium lifestyle inventory" is more evaluable if you can already see how similarly-categorised open-exchange inventory has performed for your account specifically. If no comparable data exists, treat the PMP deal as a genuine unknown and negotiate the smallest viable commitment to test it before scaling.
| Buying method | Commitment level | Best suited to |
|---|---|---|
| Open-exchange auction | None — bid per impression | Most accounts, especially without proven supply-source data yet |
| Private marketplace (PMP) deal | Negotiated minimum, often fixed term | Brands with proven, specific supply-source performance data to justify it |
Which one you should actually pick
Any advertiser with enough supply-source history can evaluate a PMP opportunity themselves using the framework above — the analysis needs a spreadsheet and existing reporting, not a specialised platform. reMKTR only recommends a PMP deal once an account's own open-exchange data has clearly identified the specific supply source worth securing, having seen a deal underperform when that groundwork was skipped, as part of the discipline behind Full Circle's $500M+ in managed Amazon spend across 100+ brands.
Shortlist on the job, not the feature grid. Pull your search-term report for the last 90 days and total the spend against terms that produced no orders — 33.6% on the account above. Then ask each vendor on your list what they would do about it in week one, and see who answers with a process rather than a screenshot.
Common questions
What's the main advantage of a PMP deal over open-exchange buying?
Guaranteed access to specific, often premium inventory at negotiated terms, rather than competing for it in the open auction — useful once you've identified a specific supply source worth securing.
Should a smaller or newer DSP account consider a PMP deal?
Usually not as a first move — open-exchange buying with good supply-source-level review is a lower-risk way to build the performance history that would justify a PMP negotiation later.
How do I know if a specific publisher's PMP pitch is worth pursuing?
Check whether comparable inventory already shows up favourably in your own open-exchange supply-source data — a pitch with no analog in your existing performance history is a genuine unknown worth testing small before committing.
Can a PMP deal be adjusted mid-flight if it underperforms?
Often less easily than an open-exchange line item, since PMP deals typically carry a negotiated minimum commitment — which is why testing the smallest viable commitment first matters more here than for standard buying.
We show the method before the number.
Claim the free auditRead next
- Skai Pricing: Five Published Tiers, Read CloselyPricing · skai pricing
- Acorn vs Podean: Two Amazon Agencies ComparedHead to head · acorn vs podean
- Pacvue Pricing: No Public Number — What to AskPricing · pacvue pricing
- Skai vs Pacvue: Contracts, Not Feature GridsHead to head · skai vs pacvue