Programmatic Advertising for E-commerce: What Moves Revenue

Programmatic advertising for e-commerce is the automated buying and selling of digital ad inventory across display, video, connected TV, and native channels, using real-time bidding and audience data to reach shoppers at scale. For e-commerce brands, the strongest programmatic options combine prospecting reach with retargeting precision, typically running across demand-side platforms like The Trade Desk, Google DV360, or Amazon DSP, depending on where your customers actually spend time and money.

The landscape has matured considerably in the last five years. What used to require a dedicated trading desk and six-figure minimums is now accessible to brands spending $10,000 a month. That accessibility is a good thing, mostly. But it also means more noise, more mediocre campaigns, and more clients convinced they’re running programmatic when they’re just running Google Display with a fancier dashboard.

Before getting into the platforms themselves, it is worth saying this clearly: programmatic is a channel within a funnel, not a funnel in itself. If you want to understand how it fits into a broader acquisition and conversion architecture, the work on high-converting funnels is the right place to start. Programmatic without funnel thinking is just expensive banner advertising.

What Makes Programmatic Different for E-commerce Specifically?

E-commerce has a structural advantage in programmatic that most brand advertisers do not: you have first-party transaction data. You know who bought, what they bought, when they bought, and how much they spent. That data, when fed cleanly into a DSP, changes the quality of your audience targeting significantly.

I spent several years running programmatic strategy for retail and e-commerce clients at iProspect. The brands that got the most from programmatic were not necessarily the ones with the biggest budgets. They were the ones who had clean CRM data, a clear view of their customer lifetime value, and the discipline to suppress converted customers from prospecting audiences. The brands that struggled were running the same creative to the same audiences for six months and wondering why ROAS was declining.

E-commerce also has a shorter consideration cycle than B2B or high-ticket consumer goods, which means programmatic retargeting can be aggressive without being inappropriate. A shopper who viewed a product page yesterday is still a warm lead today. That window is real, and it is shorter than most brands think. The paid acquisition data for DTC brands consistently shows that retargeting windows beyond seven days produce sharply diminishing returns for most product categories.

The other thing that makes e-commerce programmatic distinct is the role of the product feed. Dynamic creative optimisation, or DCO, pulls directly from your product catalogue and serves personalised ads based on what a user has viewed or what is algorithmically predicted to convert. This is table stakes now, not an advanced capability. If your programmatic setup is not running DCO for retargeting, you are leaving performance on the table.

The Trade Desk: The Strongest Independent DSP for E-commerce Scale

The Trade Desk is the platform I would recommend most often for mid-to-large e-commerce brands running serious programmatic budgets. It is genuinely independent, meaning it does not own media inventory and therefore has less incentive to push you toward placements that serve the platform rather than your campaign. That independence matters more than most marketers acknowledge.

What The Trade Desk does particularly well for e-commerce is cross-channel reach. You can run display, video, audio, connected TV, and native from a single platform with unified frequency management. For a brand running a seasonal push, the ability to coordinate messaging across a user’s streaming service, podcast, and browser in a single week is genuinely powerful. The platform’s Unified ID 2.0 initiative is also the most credible industry attempt to build a cookieless identity framework, which matters as third-party cookies continue to deprecate.

The honest limitation is the learning curve. The Trade Desk is not a self-serve platform you hand to a junior media buyer and expect results from on week one. It requires strategic setup, ongoing optimisation, and someone who understands bid shading, supply path optimisation, and audience taxonomy well enough to make smart decisions. If your team does not have that capability in-house, you need a partner who does, and you need to verify that they actually trade on the platform rather than just reselling managed service.

Amazon DSP: The Purchase-Intent Advantage No Other Platform Has

Amazon DSP is the most strategically interesting programmatic option for e-commerce brands, precisely because Amazon’s data asset is unlike anything else in the market. They know what people searched for, what they considered, what they added to cart, and what they bought. That purchase-intent signal, applied to programmatic targeting, produces audience quality that is genuinely difficult to replicate on other platforms.

The platform works best for brands in two situations: those already selling on Amazon who want to drive traffic to their product listings, and those selling off-Amazon who want to reach Amazon’s audience segments on third-party inventory. Both use cases are legitimate. The first is more straightforward to attribute. The second requires more careful measurement planning.

One thing I have seen repeatedly: brands that sell through both direct-to-consumer channels and wholesale or marketplace channels often underestimate how Amazon DSP can support the full business, not just the Amazon storefront. If you are thinking through how your channel mix affects your programmatic strategy, the analysis on direct to consumer versus wholesale is worth reading before you set your audience targeting parameters.

The practical limitation of Amazon DSP is access. Managed service requires significant minimum spend commitments, typically in the range of $35,000 to $50,000 per month depending on your region and relationship. Self-serve access through Amazon Ads is more accessible but less flexible. For brands below that threshold, The Trade Desk with third-party purchase data integrations often delivers comparable audience quality at lower entry cost.

Google DV360: The Integrated Choice for Google-Heavy Stacks

Display and Video 360, Google’s enterprise DSP, is the logical choice for e-commerce brands already running significant Google Ads spend and using Google Analytics as their primary measurement platform. The integration between DV360 and the broader Google Marketing Platform means audience lists, conversion data, and attribution flow more cleanly than they do when you are bridging multiple platforms.

DV360 also gives you access to YouTube inventory programmatically, which is a meaningful advantage for brands where video plays a role in the consideration experience. Connected TV through DV360 is improving, though The Trade Desk still has a more mature CTV offering in most markets.

The concern I have always had with DV360, and I have managed campaigns on it across multiple large retail clients, is that the Google ecosystem can create a kind of measurement tunnel vision. When your DSP, your analytics platform, and your attribution model are all Google products, you will tend to see Google-attributed results that look better than they might in a more neutral measurement environment. That is not a reason to avoid DV360. It is a reason to supplement it with independent measurement, whether that is a marketing mix model or a third-party attribution tool.

For brands running complex e-commerce operations, including those in the middle of a platform migration, DV360’s integration with Google’s tag infrastructure can simplify implementation considerably. If you are also handling a platform migration at the same time as launching programmatic, sequencing matters. Get your tracking and data layer right before you scale spend.

Criteo and Retargeting Specialists: When Reach Matters Less Than Precision

Criteo built its business on e-commerce retargeting and still does it well. If your primary programmatic objective is converting browsers into buyers using dynamic product ads, Criteo’s commerce media network and its access to retailer first-party data gives it a specific edge in that use case.

The platform has expanded beyond pure retargeting into prospecting through its Commerce Audience offering, which uses purchase behaviour signals from its retail media network to find new customers. The quality of that prospecting varies significantly by vertical and geography, but for brands in categories where Criteo has strong retail data coverage, it can be a cost-effective complement to broader DSP prospecting.

The honest assessment is that Criteo works best as part of a broader programmatic stack rather than a standalone solution. It is not where I would put the majority of programmatic budget for most e-commerce brands, but it earns its place in a retargeting layer alongside email and paid social. Speaking of which, retargeting only makes sense if your abandonment recovery sequence is functioning properly. If you have not looked hard at your abandoned cart email strategy, fix that before adding more retargeting spend. Email is cheaper and often more effective for that specific use case.

Connected TV: The Emerging Prospecting Channel Worth Taking Seriously

Connected TV deserves a dedicated section because it has moved from experimental to genuinely viable for e-commerce prospecting in the last two years. Streaming audiences are large, engaged, and increasingly addressable. The ability to reach a household watching a specific type of content, with a specific purchase history, and serve them a 30-second non-skippable ad is a meaningful capability.

The attribution challenge is real and should not be minimised. CTV does not produce the same click-through attribution signals as display or paid search. You are measuring it through view-through windows, brand lift studies, or incrementality testing, and each of those approaches has limitations. I have judged Effie entries where CTV was credited with significant sales lift, and I have also seen cases where the attribution methodology would not survive serious scrutiny. The channel can work. The measurement needs to be honest.

For e-commerce brands considering CTV, The Trade Desk and Amazon DSP are currently the strongest platforms. Roku’s OneView is worth evaluating if Roku inventory is strategically important to you. Start with a contained test, define your measurement approach before you spend, and do not let the creative production cost tempt you into running a single piece of video for six months. CTV creative fatigue is real.

CTV also has interesting implications for CPG brands selling through e-commerce channels, where the brand awareness function is as important as direct response. If you are working through a CPG e-commerce strategy, CTV is one of the few programmatic channels that can credibly carry both brand and performance objectives simultaneously.

Retail Media Networks: The Programmatic Channel Most E-commerce Brands Underuse

Retail media networks are, technically, a form of programmatic advertising. Walmart Connect, Target Roundel, Instacart Ads, and the growing number of retailer-owned ad platforms give brands access to first-party shopper data at the point of purchase intent. For e-commerce brands that sell through these retailers, the targeting precision is exceptional.

The challenge is fragmentation. Managing campaigns across five different retail media networks, each with its own interface, reporting logic, and creative specifications, is operationally demanding. Brands often underinvest in retail media not because the channel does not work, but because the operational overhead is underestimated at the planning stage.

There are aggregation platforms emerging that allow multi-retailer management from a single interface, including Skai and Pacvue. These are worth evaluating if retail media is a meaningful part of your mix. The measurement story across retail media is also improving, with more networks offering closed-loop attribution tied to actual purchase data rather than modelled estimates.

For brands in financial services or fintech operating marketplace models, the positioning dynamics of retail media have direct parallels in how financial marketplaces are structured. The financial marketplace positioning strategies framework applies more broadly than its category suggests, particularly around how to compete for visibility in a curated inventory environment.

How to Choose the Right Programmatic Mix for Your E-commerce Brand

The question I get asked most often is: which platform should we be on? It is the wrong question. The right question is: what is the specific job this programmatic spend needs to do, and which platform does that job best given our data assets, budget, and team capability?

I have seen too many e-commerce brands chase platform innovation without a clear business problem to solve. A brand that cannot attribute its existing display spend accurately does not need a CTV strategy. A brand with a 2% site conversion rate does not need more top-of-funnel programmatic reach. The discipline to sequence these decisions correctly is more valuable than any platform feature.

A practical framework for most e-commerce brands looks like this. Start with retargeting, because it is the highest-intent audience and the most defensible spend. Get your product feed, your audience segmentation, and your creative rotation right before you scale. Then layer in prospecting through whichever DSP aligns with your data assets and team capability. Add CTV when you have a measurement model that can handle view-through attribution honestly. Consider retail media when you have the operational bandwidth to manage it properly.

Budget allocation matters too. There is no universal right answer, but a reasonable starting point for a brand new to programmatic is to put 60 to 70 percent of programmatic budget into retargeting and the remainder into prospecting, then shift that ratio as prospecting audiences prove out. The demand generation benchmarks from HubSpot provide useful context for how programmatic fits into broader acquisition cost expectations by channel.

One more thing worth saying plainly: programmatic advertising is not a set-and-forget channel. The brands that get consistent returns from it treat it as an active discipline, with weekly optimisation, regular creative refreshes, and ongoing audience testing. The brands that treat it as a background channel and check in monthly will consistently underperform. That is not a platform problem. It is an attention problem.

Programmatic is one component of a broader acquisition system. If you want to see how it connects to the full funnel from awareness through conversion, the high-converting funnels hub covers the architecture in detail, including how each channel type earns its place in the sequence.

For further reading on how programmatic fits into broader lead generation and conversion strategy, Semrush’s lead generation breakdown and HubSpot’s website optimisation guide are both worth reviewing alongside your programmatic setup. And if you are thinking about how programmatic interacts with organic search at the funnel level, Moz’s work on the organic search conversion funnel offers a useful complementary perspective.

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