Programmatic display ROI

Programmatic display: the channel most marketers misprice

Programmatic display has a brand problem in the marketing community. Half the industry still thinks of it as worthless banner-ad spam, the other half runs it as a "must-have retargeting layer" without measuring whether it's actually incremental. The truth sits somewhere harder to navigate: programmatic display, run well, is one of the most underpriced impression inventories available to a performance marketer in 2026. Run badly, it's a pure tax on your paid-media budget.

This calculator factors in the real variables: CPM, viewability rate, CTR, post-view conversion, post-click conversion, fraud rate, and retargeting incrementality. Most "display ROI" calculators assume 100% viewability and treat the platform-reported click conversions as incremental. They're neither. The output of this tool is net ROI after the friction — which is typically 40–70% of the gross figure.

Benchmarks: programmatic display economics in 2026

The three layers of display that behave like different channels

Prospecting display (cold audiences). Lowest intent, weakest ROI, but it can work at scale with very specific audience targeting and strong creative. Not recommended for brands under $5M in revenue — you don't have the attribution rigor to separate signal from noise.

Retargeting display (site visitors, cart abandoners). The most common use case. Works at face value on last-click attribution, looks dramatically less impressive on incrementality-adjusted numbers. Genuinely effective when done with aggressive frequency caps (under 8 impressions per user per month) and exclusion lists that suppress already-purchased customers.

Contextual display (content-relevant placements). The post-cookie winner. Placements chosen based on page content, not user profile. Less targeting precision but zero privacy risk and zero signal-loss exposure. Growing share of modern programmatic spend — up from ~15% in 2020 to ~35% in 2026.

The incrementality check: why retargeting display looks better than it is

A classic test: pause retargeting display for 3 weeks and watch total conversions. If your retargeting-attributed revenue was $120K/month and total company revenue drops by only $45K, retargeting display's incrementality was 38% — the remaining $75K would have converted through email, direct, branded search, or paid social anyway. This is the 2021 Uber paper, the 2018 eBay experiment, and dozens of smaller DTC experiments all saying the same thing: retargeting display's incrementality is usually 30–60%, not 100%.

This doesn't mean "don't run retargeting." It means: model it at 40–50% incrementality, not 100%. Apply that haircut in this calculator's incrementality field to get honest ROI numbers.

The DSP selection: why it matters more than creative

DSP (Demand-Side Platform) choice affects outcome more than most buyers realize. The big four:

• The Trade Desk. The dominant independent DSP. Strongest data and inventory integrations. Typically requires $5K+/month commitment. Best for mid-market and above.
• DV360 (Google). Tight Google ecosystem integration. Strong YouTube integration. Widely used by agencies.
• StackAdapt. Strong self-serve platform. Good for small-mid brands. Native advertising integration.
• MediaMath, Xandr (Microsoft), Basis. Second-tier but still legitimate. Worth comparing if you're in a specific niche.

Below $30K/month in programmatic spend, you're usually better off on Google Display Network or through Criteo managed service than building a full DSP operation. The minimum spend to make DSP economics work is typically $25–50K/month.

Creative formats that still perform

Static 300x250 and 728x90 banners are at the bottom of the performance pile. The formats that actually move metrics:

• Native display. In-feed placements formatted to match publisher content. 2–4x CTR vs. traditional banners.
• Video (pre-roll, mid-roll, outstream). Higher CPMs but much higher engagement. Runs through the same DSPs.
• CTV / OTT. Connected TV advertising is programmatic under the hood. Tight measurement, growing fast, but CPMs are $30–60 — brand play, not direct response.
• Audio (Spotify, podcasts via programmatic). Spotify Audience Network, Acast, Art19. Highly engaged listeners, limited inventory.
• DOOH (digital out-of-home) programmatic. Billboards, transit screens — sold programmatically now through Vistar, Place Exchange.

Fraud and viewability: the two taxes everyone pays

IAB viewability standard is 50% of pixels in-view for 1+ second. Most DSPs will report 55–72% viewability as standard; anything under 55% is a creative or placement problem. Set viewability floors in your DSP (70%+ for premium buys) and accept that your effective CPM rises 20–40% when viewability is enforced — but fraud drops proportionally.

Invalid Traffic (IVT) — bots, domain spoofing, click farms — runs 3–12% on open exchange and 0.5–3% on PMP / programmatic guaranteed. Always run DV, IAS, or Moat pre-bid filtering. The fee (~$0.03–0.08 CPM) is cheap insurance.

When display actually wins vs. Meta, Google, and TikTok

Three scenarios where programmatic display is a better dollar than social:

1. Highly niche B2B audiences. If your ICP is "IT directors at mid-market hospitals," you can target that on The Trade Desk using Bombora or LinkedIn Audience Network integrations far more precisely than on Meta.
2. Geo-specific retail activation. Mobile DOOH programmatic around physical store locations is often unbeatable for quick-service retail driving foot traffic.
3. Contextual brand alignment. If your brand specifically needs to appear next to health content, finance content, or sports content — programmatic contextual buys get you adjacency that social can't.

Outside of these niches, display is usually a "layer" channel rather than a primary driver. It should support retargeting, post-click re-engagement, and brand continuity — not be the primary acquisition driver.

The B2B programmatic use case: account-based display

For B2B SaaS at $50K+ ACV, programmatic display targeted at a specific account list (via 6sense, Demandbase, or RollWorks) is one of the most efficient awareness investments available. Typical spend: $5–15K/month on a 500-account target list produces 40–70% account-reach and measurable lifts in direct traffic and demo requests from those specific accounts. Pair with the Attribution tool to see how different attribution models credit this channel.

Three programmatic display archetypes with full economics

Archetype 1: DTC home goods, $22M revenue, retargeting-heavy

$38K/mo on programmatic display, 80% retargeting via Criteo ($30K) + 20% prospecting contextual via StackAdapt ($8K). Criteo fee embedded at roughly 15% of media. IAS verification $0.05 CPM adds $1,900/mo. MFA block list from Jounce (free tier) active. Frequency cap: 6 impressions per user per month. Gross attributed revenue via last-click: $185K/mo. Pause-test at month 4 showed incremental lift was 42% — so true incremental revenue $78K/mo. Net ROI $78K incremental / $39,900 all-in cost = 1.95x on incrementals; gross reported ROAS 4.6x. Decision: maintain at current spend, do not scale prospecting until CTR on cold audiences clears 0.09% consistently (currently 0.055%).

Archetype 2: B2B SaaS, $45M ARR, account-based display

6sense at $60K/year + The Trade Desk managed at 16% of $18K/mo media = $2,880/mo platform. Targeting a 520-account list across IT Security and Cloud Infrastructure buyers. Account reach month 6: 68% of list hit with 10+ impressions. Measured lift: direct traffic from target accounts up 44%, branded search queries from target accounts up 31%, demo requests from target accounts up 22% vs. a matched holdout of 150 accounts that received no display. Attribution via Dreamdata ($1,800/mo) credits $540K pipeline influence in Q3, $135K closed-won. Net program cost: $18K media + $2,880 platform + $5K attribution-share + $5K 6sense allocation = $30,880/mo or $370K/year. ROI on closed-won $1.62M annualized / $370K = 4.4x; on pipeline influence 17x. Scale plan: expand to 900-account list in Q4.

Archetype 3: Regional QSR chain, $12M revenue, DOOH + mobile

Quick-service restaurant chain with 38 locations in the southeast. Runs $24K/mo programmatic DOOH via Vistar Media targeting billboards and transit screens within 3-mile radius of each store, plus $16K/mo in mobile programmatic via The Trade Desk geo-fenced to the same radius. Total $40K/mo. Measured via Foursquare Attribution ($8K/mo for foot traffic measurement): incremental store visits lift of 6.3% across the network, equating to roughly 14,000 additional visits/mo at $11.50 average ticket and 24% gross margin = $39,000 incremental gross margin. Net ROI 0.97x on margin basis — marginal but acceptable for a brand-spend rationale. Decision: cut 30% of DOOH spend on bottom-quartile screens (low viewability zones), reinvest in Samsung Ads CTV targeting households within delivery radius.

Programmatic stack pricing reference (April 2026)

Decision framework: scale, optimize, or cut programmatic display

Scale signals: retargeting incremental ROAS above 3x, prospecting CTR above 0.12%, viewability above 68%, MFA exposure under 5%, and clear lift in direct traffic or branded search month-over-month. Optimize signals: gross ROAS 2–4x but incremental under 2x — the fix is almost always frequency caps (drop to 6/mo/user) and tighter exclusion lists (suppress customers within 90 days of purchase). Cut signals: incremental ROAS under 1.5x after MFA blocking is live and verification is enforced; or budget under $30K/mo where DSP minimums and agency fees eat the economics. Always triangulate with Attribution output — last-click will overstate display's contribution, while a data-driven or position-based model usually shifts 20–40% of display-attributed revenue back to upstream channels. If you are under $25K/mo total display spend, collapse onto Google Display Network plus Criteo and do not try to run a multi-DSP program.