Programmatic Advertising vs. Real-Time Bidding: Choosing the Right Deal Type

One word describes the category, the other describes one deal inside it

Programmatic advertising is the category: any ad transaction executed by software rather than negotiated by a person, at scale, across an automated pipeline of DSPs, SSPs, and exchanges. Real-time bidding is a specific mechanism inside that category — the open auction, where every individual impression is priced live against every other bidder connected to the exchange in that moment. Asking "programmatic vs. RTB" is really asking "is RTB the only way to buy programmatically," and the answer is no. It's the fastest-moving and most competitive lane, but it's one of at least three.

The three lanes, and what each one actually controls

RTB — the open exchange — gives a buyer scale and speed: any advertiser can bid on any available impression, prices float in real time, and there's no negotiation required to start. What it doesn't give a buyer is certainty about exactly which inventory they'll win or guaranteed volume.

A private marketplace (PMP) deal is still an auction, but an invite-only one — a publisher curates which advertisers can bid on a specific slice of premium inventory, sets its own floor price, and the buyer gets more transparency about where the ad actually ran, at a price that usually runs above the open exchange.

Programmatic guaranteed (PG) removes the auction entirely — a fixed volume of impressions at a fixed price, negotiated once between a specific buyer and a specific publisher, executed automatically rather than manually, but with no live bidding involved at all. It trades RTB's flexibility for the certainty of a traditional insertion order, delivered through programmatic plumbing instead of a fax.

Most media plans that run at any real scale end up using more than one lane at once, not choosing a single one permanently. It's common to run open-exchange RTB for broad prospecting and retargeting, where reach matters more than knowing exactly which site an impression landed on, while reserving PMP or PG deals for a smaller slice of the budget aimed at specific premium placements a brand wants guaranteed visibility on. Treating the choice as all-or-nothing is itself a planning mistake — the three lanes are complementary tools, not competing philosophies.

A worked example: the same $50,000 across three lanes

Say a brand has $50,000 to spend on display this month and can run it through any of the three. Through the open exchange, that budget might clear at a blended $4.00 CPM, buying roughly 12.5 million impressions across a broad mix of sites — high reach, uneven quality, and no guarantee which publishers actually ran the ads. Through a PMP, the same budget might clear at $7.00 CPM on curated, brand-safe inventory from named publishers — about 7.1 million impressions, fewer than the open exchange, but each one on inventory the buyer explicitly chose. Through a PG deal, the buyer might negotiate a fixed 6 million impressions on one specific premium property at a locked $8.33 CPM — the smallest reach of the three, the highest certainty, and zero real-time optimization once the deal is signed.

None of the three is the "correct" choice in the abstract. The right lane depends on whether the campaign's goal is reach, quality control, or guaranteed delivery on a specific placement — and a media plan that only ever runs open exchange is choosing reach every time, whether or not that's the actual goal.

The mistake: treating RTB volume as proof of efficiency

The most common mistake with open-exchange RTB is judging a campaign's health by how much volume it moved rather than where that volume actually went. In a 31-day pull across 27 advertisers in our own Amazon DSP book, 18.3% of live line items consumed 80% of total spend — 242 line items out of 1,325 accounted for four dollars in five. That's not a waste finding on its own; it's evidence that most of the reporting attention in a typical account review goes to an alphabetical list of line items rather than a spend-sorted one, which means the tail gets scrutinized while the concentration that actually decides the outcome gets skimmed.

The practical fix is to sort by spend before reviewing anything else, in RTB or any other lane — the line items worth a detailed look are usually a fifth of the list, not all of it.

How to pick, and what to check when the wrong lane is costing you

Start with the goal, not the mechanism. If the campaign needs raw reach and can tolerate some inventory-quality variance, open-exchange RTB is the right default. If brand safety or placement quality is the binding constraint, move the budget toward a PMP even at a CPM premium. If a specific publisher's audience is non-negotiable and the budget can absorb guaranteed delivery at a fixed price, PG is worth the loss of flexibility. If a campaign already running RTB isn't performing, check the deal type before touching the creative or targeting — a brand-sensitive campaign forced into the open exchange for cost reasons will often show weak results that have nothing to do with the ad itself.

Where Dr. DSP fits

Dr. DSP is Amazon DSP run as a managed product by Full Circle, which has managed more than $500M in revenue across 100+ brands, mixing open-exchange, PMP, and Amazon's own owned-and-operated inventory depending on what a given campaign actually needs. A reader who never buys anything from us should still leave this page knowing that "we're running RTB" and "we're running programmatic" are not interchangeable claims, and that the deal type behind a media plan says more about what it's optimized for than the word "programmatic" ever will.