Programmatic Advertising Types: A Buyer’s Map

Programmatic advertising is the automated buying and selling of digital ad inventory, executed in real time through technology platforms rather than direct negotiation. The main types are open auction (real-time bidding), private marketplace deals, preferred deals, and programmatic guaranteed, each offering a different balance of reach, control, and cost.

Understanding which type fits your objective is not a technical question. It is a commercial one. The wrong format at the wrong stage of a campaign is just spend with extra steps.

If you are mapping out how programmatic fits into a broader go-to-market plan, the Go-To-Market and Growth Strategy hub covers the upstream decisions that should shape your channel mix before you start buying inventory.

What Is Programmatic Advertising, and Why Does the Type Matter?

Programmatic advertising automates the process of buying digital media. Instead of calling a publisher, agreeing a rate, and sending a booking form, a demand-side platform (DSP) bids on impressions on your behalf, often within milliseconds of a page loading. The publisher’s supply-side platform (SSP) receives bids, the highest wins, and the ad serves.

That is the mechanics. The commercial reality is more nuanced. Programmatic is not one thing. It is a family of buying mechanisms, each with a different relationship between buyer, seller, price, and inventory quality. Treating them as interchangeable is one of the more common and more expensive mistakes I have seen teams make.

Early in my career, I was heavily focused on lower-funnel performance. Clicks, conversions, cost per acquisition. It felt scientific. Over time I came to understand that a significant proportion of what performance channels were being credited for was going to happen regardless. Someone who was already in market, already familiar with the brand, already close to a decision. The programmatic channel intercepted them, claimed the conversion, and the attribution model called it a win. The real question, the one that matters commercially, is whether you are reaching people who would not have found you otherwise. That is where the type of programmatic you choose starts to matter enormously.

Open Auction (Real-Time Bidding): Maximum Reach, Minimum Control

Open auction, also called real-time bidding or RTB, is the default programmatic buying model. Inventory is made available to all buyers simultaneously. Your DSP submits a bid, and if it wins, your ad serves. The whole process takes under 100 milliseconds.

The appeal is scale. Open auction gives you access to billions of impressions across thousands of publishers. If reach and cost efficiency are your primary metrics, this is where you start.

The trade-off is context and quality. Publishers typically make their least premium inventory available in open auction. You may find your brand appearing next to content you would not choose to be associated with. Brand safety tools help, but they are filters applied after the fact, not guarantees. The Forrester intelligent growth model has long argued that sustainable growth requires both reach and relevance, and open auction optimised purely for volume can undermine the second in pursuit of the first.

Open auction is the right choice when you are building awareness at scale, testing creative across a broad audience, or working with a limited budget that cannot support the floor prices in premium environments. It is the wrong choice when brand context matters, when you are targeting a narrow professional audience, or when you are in a regulated category where adjacency risk is high.

Private Marketplace Deals: Curated Access at a Price

A private marketplace (PMP) is an invitation-only auction. A publisher or group of publishers makes a defined set of inventory available to a select group of buyers, typically at a higher floor price than open auction. You still bid in real time, but you are competing in a smaller pool against buyers the publisher has chosen to work with.

The quality difference is meaningful. PMP inventory tends to be premium placements on reputable publisher sites, the kind of inventory that publishers reserve for direct buyers or trusted programmatic partners. You get better brand safety, better viewability, and often better audience data tied to the inventory.

I have seen PMP deals work particularly well in B2B contexts, where the publisher’s audience data is more valuable than the raw impression volume. If you are targeting finance decision-makers, for example, a PMP with a business publisher that has verified subscriber data is categorically different from open auction targeting with third-party segments. This connects directly to the kind of audience-first thinking covered in B2B financial services marketing, where context and credibility of placement carry real weight in how a message lands.

The limitation of PMPs is that you are still bidding. There is no guarantee you will win the impressions you want. If the floor price is high and competition within the deal is strong, your actual delivery may fall short of plan. For campaigns that require predictable reach against a specific audience, that uncertainty is a problem.

Preferred Deals: Fixed Price, No Guarantee

A preferred deal is a one-to-one arrangement between a buyer and a publisher. The publisher offers specific inventory at a fixed CPM to a single buyer, who gets first right of refusal before that inventory goes to auction. If you pass on it, it moves down the waterfall to PMP or open auction.

The advantage is price certainty and exclusivity of access. You know what you are paying. You have a relationship with the publisher. And you get access to inventory that other buyers in the open market do not.

The limitation is still the lack of delivery guarantee. You are not committed to buying, and the publisher is not committed to delivering a fixed volume. It is a handshake arrangement executed through technology. For brands that need to plan media weight precisely, this ambiguity creates forecasting problems.

Preferred deals tend to suit advertisers who have established publisher relationships and want to maintain them in a programmatic workflow, without going back to fully manual insertion orders. They are also useful in endemic advertising contexts, where appearing on a specific type of publisher site matters as much as the audience data attached to the impression. The piece on endemic advertising goes into more depth on why publisher environment can be a targeting variable in its own right.

Programmatic Guaranteed: Automation With Commitment

Programmatic guaranteed (PG) is the closest programmatic gets to a traditional direct buy. A buyer and publisher agree on a fixed volume of impressions, at a fixed price, for a defined audience. The deal is executed programmatically, but the commitment is mutual. The publisher guarantees delivery. The buyer guarantees spend.

This is the premium end of the programmatic stack. You get the best inventory, the most control, and the most predictable delivery. You also pay for it, and you carry the risk if your targeting parameters are too narrow to fill the agreed volume.

Programmatic guaranteed makes sense for brand campaigns with fixed flight dates, high-profile product launches, or situations where appearing in specific premium environments is non-negotiable. It also makes sense when you are working with a publisher whose audience you trust more than third-party data, and you want to lock in access before it goes to the open market.

One thing I have noticed over the years is that brands doing proper digital marketing due diligence before committing to programmatic spend tend to land on PG deals more often than they expected. When you actually audit what your media is doing, where it is appearing, and what quality looks like across your current buy, the case for paying more for guaranteed premium inventory often becomes clear very quickly.

Programmatic Across Channels: Display Is Just the Start

Most marketers think of programmatic as display advertising. Banner ads, leaderboards, MPUs. That is where the category started, but it has expanded significantly.

Programmatic video now covers pre-roll, mid-roll, and out-stream formats across web and app environments. Connected TV (CTV) has become one of the fastest-growing programmatic channels, with inventory from streaming services and smart TV apps available through the same DSP infrastructure used for display. Programmatic audio covers podcast and streaming music inventory. Digital out-of-home (DOOH) is increasingly traded programmatically, with screens in retail, transport, and outdoor environments available through automated buying platforms.

Each of these channel extensions carries its own inventory quality dynamics, measurement challenges, and buying norms. CTV, for example, has significant fragmentation across supply sources, and the same four buying types apply but with different floor prices and availability patterns. Audio programmatic is still maturing, and the measurement frameworks that work for display do not translate cleanly.

When I was growing an agency from a small team to over a hundred people, one of the things we learned early was that channel expertise and buying expertise are not the same thing. You can understand programmatic mechanics perfectly and still make poor decisions in CTV because you do not understand the content environment or the measurement conventions. The channel always has context that the technology does not supply.

For teams thinking about how programmatic fits into demand generation, it is worth looking at how pay per appointment lead generation models handle the intent-to-conversion gap, because programmatic often sits at the top of that funnel and the handoff to lower-funnel activity is where value leaks.

Audience Targeting in Programmatic: Where the Real Decisions Are

The buying type determines how you access inventory. The targeting determines whether that inventory is worth anything to you. These are separate decisions, and conflating them is where a lot of programmatic spend goes wrong.

Programmatic targeting broadly falls into a few categories. Contextual targeting matches your ad to the content of the page, without using audience data. Audience targeting uses behavioural, demographic, or intent signals to find specific users wherever they are. Retargeting reaches people who have already interacted with your brand. Lookalike targeting finds new users who share characteristics with your existing customers.

The deprecation of third-party cookies has shifted the balance back toward contextual and first-party data approaches. Contextual targeting, which had been somewhat dismissed as unsophisticated, has had a quiet rehabilitation. It turns out that appearing in a relevant content environment is actually a reasonable proxy for audience intent, and it does not require tracking individuals across the web.

For B2B advertisers, the targeting question is particularly sharp. Third-party B2B audience segments in programmatic platforms are often stale, broadly defined, or built from data that does not reflect current job function or buying authority. The BCG research on financial services go-to-market strategy makes a point that applies well beyond that sector: understanding who is actually in the buying process, at what stage, matters more than demographic approximation. Programmatic targeting built on job title segments often misses this entirely.

The brands I have seen get the most from programmatic are the ones who treat audience definition as a strategic exercise before they open a DSP, not something they configure in the platform on setup day. If you want a structured way to approach that, the website analysis checklist for sales and marketing strategy is a useful starting point for understanding what your current digital presence tells you about who is actually engaging with your brand.

Measurement and Attribution: The Part Most Teams Get Wrong

Programmatic platforms generate a lot of data. Impressions, clicks, viewability rates, completion rates, frequency, reach. It feels comprehensive. It is not.

The attribution models inside most DSPs are built to show the platform’s contribution in the most favourable light. View-through attribution, which credits a conversion to an ad impression the user may have barely noticed, inflates the apparent value of display and video programmatic significantly. Last-click models undervalue upper-funnel activity. Multi-touch models distribute credit according to assumptions that are rarely validated against actual purchase behaviour.

I spent a long time managing large programmatic budgets and watching attribution models tell a story that was internally consistent but commercially misleading. The channel that looked most efficient in the platform was often the one intercepting demand that other channels had created. When we ran incrementality tests, the picture looked quite different. Some of what programmatic was claiming credit for was going to happen regardless.

This is not an argument against programmatic. It is an argument for honest measurement. Use holdout tests where you can. Look at brand search volume as a proxy for upper-funnel impact. Compare performance in markets where you are running programmatic against markets where you are not. The Semrush analysis of market penetration covers some of the frameworks useful for thinking about incremental reach, which is the right question to be asking of your programmatic investment.

The goal is honest approximation, not false precision. A measurement approach that acknowledges its own limitations is more useful than one that produces confident numbers built on flawed assumptions.

Choosing the Right Programmatic Type for Your Objective

There is no universal answer to which programmatic buying type is best. There is only the right fit for a specific objective, audience, and budget.

If your objective is broad awareness and you have a large audience to reach with limited budget, open auction gives you the scale to do that cost-effectively. Accept the quality trade-offs and invest in brand safety and exclusion lists.

If you are targeting a specific professional audience and publisher context matters, a PMP with the right publisher partner will outperform open auction even at a higher CPM. The audience quality and brand environment justify the premium.

If you have an established publisher relationship and want to maintain programmatic efficiency without going back to manual buying, preferred deals offer a practical middle ground. You get price certainty without a delivery commitment on either side.

If you are running a time-sensitive campaign where delivery predictability is critical, such as a product launch or a seasonal push, programmatic guaranteed removes the uncertainty. You pay more, but you know what you are getting.

Most sophisticated advertisers run a mix across all four types, weighted toward the formats that match each campaign objective. The corporate and business unit marketing framework for B2B tech companies is a useful reference for thinking about how different campaign objectives map to different channel and buying strategies, particularly in organisations where corporate brand and product-level demand generation are running simultaneously.

There is a broader point here about how programmatic fits into a growth strategy rather than just a media plan. The Go-To-Market and Growth Strategy hub covers the planning layer above channel selection, which is where decisions about programmatic type should in the end be grounded.

When I was handed the whiteboard pen at Cybercom with no warning and a room of people waiting, the instinct was to reach for the safest, most familiar framework. The lesson I took from that, and from two decades of media planning since, is that the right answer is rarely the default one. Open auction is the default in programmatic. It is not always the right choice. Asking which type serves the actual objective is a more productive question than asking which type is easiest to set up.

For teams thinking about growth hacking and programmatic’s role in acquisition, the Crazy Egg overview of growth hacking is a reasonable starting point for understanding how programmatic sits within a broader acquisition mix, and the Vidyard piece on why go-to-market feels harder captures some of the structural reasons why programmatic alone rarely solves a growth problem.

About the Author

Keith Lacy is a marketing strategist and former agency CEO with 20+ years of experience across agency leadership, performance marketing, and commercial strategy. He writes The Marketing Juice to cut through the noise and share what works.

Frequently Asked Questions