Connected TV (CTV) advertising is paid video delivered to internet-connected televisions, combining the emotional impact of big-screen storytelling with the precision targeting and measurable outcomes of digital media. For brand marketers and performance advertisers, it sits at the intersection of reach and accountability, and nearly 70% of the U.S. population now uses CTV, making it one of the most consequential channels in any modern media plan.
Here is the strategic TL;DR before we go deeper:
- CTV works across the full funnel. Use it for brand awareness at the top, retargeting and consideration in the middle, and outcome attribution at the bottom.
- CPMs run higher than social video but deliver completion rates that commonly exceed 90% in non-skippable premium environments, making cost-per-attention far more competitive than it first appears.
- Measurement has shifted. Over 27.4% of marketers now prioritize conversions and sales as their primary CTV KPI, moving away from legacy reach metrics toward business outcomes.
- Your immediate next step: audit your current identity and measurement infrastructure before you buy a single impression. Gaps in data clean room access or CRM linkage will limit what you can prove.
Table of Contents
- How connected TV advertising actually works
- CTV ad formats and creative specs that drive results
- How targeting and identity work in CTV
- Measuring CTV campaigns: from impressions to business outcomes
- Buying CTV inventory: platforms, methods, and selection criteria
- What CTV advertising costs and how to set a realistic budget
- How to integrate CTV into your omnichannel media plan
- Risks, supply quality, and privacy considerations for U.S. advertisers
- Step-by-step checklist to launch your first CTV campaign
- How Theartistevolution approaches CTV campaigns
- Key Takeaways
- Why CTV deserves a permanent seat in your media mix
- Theartistevolution builds and measures CTV campaigns end to end
How connected TV advertising actually works
CTV advertising follows a specific delivery chain that differs meaningfully from linear TV or desktop video. A viewer opens a streaming app on a smart TV, Roku, Amazon Fire TV, Apple TV, or gaming console. The app sends an ad request to an ad server or supply-side platform (SSP). That request enters an auction or a pre-negotiated deal, a winning ad is selected, and the creative is inserted into the stream before playback begins.
Two insertion methods dominate. Server-side ad insertion (SSAI) stitches the ad directly into the video stream at the server level, making it nearly impossible for ad blockers to intercept and producing a seamless viewing experience. Client-side ad insertion (CSAI) calls the ad from the viewer’s device, which is more vulnerable to blocking and can cause buffering. Premium publishers overwhelmingly favor SSAI, and you should prioritize it when evaluating inventory quality.
Buying methods split into three lanes. Programmatic real-time bidding (RTB) gives you speed and scale through demand-side platforms (DSPs). Private marketplace deals (PMPs) offer curated, publisher-direct inventory with stronger brand safety controls. Direct/guaranteed buys lock in specific placements and pricing, which suits brand campaigns with fixed flight dates. Approximately 84% of CTV advertising is now bought programmatically, so programmatic readiness is not optional.

Pro Tip: Before your first campaign goes live, confirm that your creative files are encoded to the publisher’s spec (H.264, 1920×1080, 23.98fps minimum), your impression trackers fire correctly in an SSAI environment, and your measurement pixel or log-level export is configured. These three items cause the majority of launch delays.
CTV ad formats and creative specs that drive results
Non-skippable pre-roll and mid-roll video are the workhorses of CTV. In premium, non-skippable environments, completion rates commonly reach the 90th percentile, far above mobile or desktop video benchmarks. That captive attention is the creative opportunity, and wasting it with a repurposed social video is one of the most common mistakes brands make.
Essential specs to plan for:
- Video length: 15 seconds (brand recall), 30 seconds (storytelling/DR), 60 seconds (high-consideration products)
- Codec: H.264 or H.265; container MP4 or MOV
- Resolution: 1920×1080 (16:9); 4K accepted by select publishers
- Bitrate: minimum 15 Mbps for 1080p
- Audio: stereo AAC, normalized to -24 LKFS per broadcast standards
- Ad pod position: confirm pre-roll vs mid-roll with your SSP or publisher, as pod length affects attention
Beyond specs, the creative approach determines whether your completion rate translates to brand recall or purchase intent. Two frameworks work on CTV:
Storytelling creative builds emotional resonance over 30–60 seconds. It suits brand campaigns where the goal is awareness, affinity, or category entry. The big screen rewards cinematic composition, ambient sound design, and a narrative arc that earns the viewer’s attention rather than demanding it.
Direct response creative front-loads the offer or product within the first five seconds, uses clear supers and a spoken call to action, and drives viewers to a URL or QR code. It suits performance campaigns where the KPI is a downstream conversion.

For interactive TV ads and shoppable overlays, the format is growing but still publisher-dependent. Overlays that display a product URL or QR code during playback are available on select platforms and work particularly well for CPG and retail brands with short purchase cycles.
How targeting and identity work in CTV
CTV targeting does not rely on cookies. Instead, it uses four primary identity approaches, each with distinct accuracy, scale, and privacy tradeoffs.
| Identity Approach | How It Works | Accuracy | Scale | Privacy Risk |
|---|---|---|---|---|
| Household ID / ACR data | Automatic content recognition matches viewing behavior to a household; no login required | Moderate | High | Low to moderate |
| Deterministic (hashed login IDs) | Publisher matches authenticated user email/ID to advertiser CRM data | High | Lower (login-gated) | Low when hashed |
| Probabilistic device graph | Statistical modeling links devices by shared signals (IP, behavior) | Moderate | High | Moderate |
| CRM onboarding / data clean room | Advertiser uploads first-party data; matched to publisher IDs in a privacy-safe environment | High | Depends on match rate | Low (clean room controls) |
Targeting capability is now the top investment criterion for CTV buyers, surpassing content quality. That shift reflects a market that has moved past “reach TV audiences” toward “reach my customers on TV.”
Practical guidance: use deterministic matching when your CRM list is large enough to generate meaningful reach (typically 100,000+ records) and your publisher supports authenticated data. Use probabilistic targeting for prospecting at scale where some identity variance is acceptable. Contextual targeting, which matches ads to content genre or show category, is a strong privacy-safe complement when identity signals are limited.
U.S. privacy rules, including state-level frameworks like the California Consumer Privacy Act (CCPA) and the growing patchwork of state laws, require that any CRM data used for targeting be collected with appropriate consent and processed through privacy-compliant pipelines. Data clean rooms, where neither party shares raw PII, are the standard mechanism for compliant first-party data activation in CTV.
Measuring CTV campaigns: from impressions to business outcomes
Start with the business outcome you need to prove, then build your measurement stack backward from there. Impressions and reach are table stakes; the question your CFO will ask is whether the campaign moved sales.
Three-tier KPI framework:
Tier 1: Awareness and reach. Unique household reach, frequency, and share of voice. These matter for brand campaigns and for setting baseline exposure before measuring downstream effects.
Tier 2: Engagement and attention. Video completion rate (VCR), audible and visible on complete (AVOC), and attention scores where available. A 90%+ VCR in a non-skippable environment confirms delivery quality but does not confirm impact.
Tier 3: Business outcomes. ROAS, cost per acquisition (CPA), incremental sales lift, and store visit lift. These require identity-based matching or incrementality testing to produce.
Because CTV ads are non-clickable, attribution relies on household-level matching, linking TV ad exposure to devices on the same Wi-Fi network and then connecting those devices to downstream conversions via identity graphs or authenticated platform data. This produces vendor variance: two platforms measuring the same campaign will often report different attributed conversions. The mitigation is to run a holdout/incrementality test alongside your standard attribution.
Streaming now accounts for nearly 44% of U.S. TV time, yet measurement fragmentation across publishers remains a real operational challenge. Ingest log-level exports from each publisher into a unified analytics environment, whether a data clean room, CDP, or BI tool, to reconcile completion, reach, and frequency definitions across sources.
For retail brands, data clean rooms and loyalty-card integrations now make it possible to connect CTV ad exposure directly to in-store sales, producing clearer ROAS visibility than was achievable even two years ago. Theartistevolution’s retail activation framework is built around exactly this kind of exposure-to-outcome linkage.
Incrementality testing: run a geographic or audience holdout test for at least four weeks to isolate the causal lift from CTV exposure. Results should directly inform budget allocation decisions, not just validate the campaign post-hoc. A practical approach to measuring marketing ROI starts with defining the counterfactual before the campaign launches.
Buying CTV inventory: platforms, methods, and selection criteria
The three main buying routes each serve a different need.

Programmatic DSP/RTB gives you access to broad inventory across multiple publishers through a single platform. It suits prospecting campaigns, audience extension, and any scenario where reach and efficiency matter more than placement exclusivity.
Private marketplace (PMP) deals connect you to curated, publisher-direct inventory at negotiated CPMs. You get stronger brand safety guarantees, better data transparency, and often access to premium content environments that are not available in open auction.
Direct/guaranteed buys lock in specific placements, time slots, and pricing. They are best for high-stakes brand campaigns, tentpole events, or situations where content adjacency is a brand requirement.
Platform callouts:
The Trade Desk is the dominant independent DSP for CTV. Its Unified ID 2.0 framework provides an open, privacy-forward identity solution that works across publishers without relying on cookies or proprietary walled-garden data. It is the strongest choice for advertisers who want cross-publisher reach with consistent identity resolution and transparent log-level reporting.
Amazon Ads brings first-party purchase and browsing data from Amazon’s retail ecosystem, making it particularly powerful for CPG, retail, and e-commerce brands. Its CTV inventory spans Fire TV, Freevee, and third-party streaming apps. The tradeoff is that its identity and measurement advantages are strongest within the Amazon ecosystem.
LiveRamp is not a DSP but a data connectivity platform. It enables CRM onboarding, identity resolution, and data clean room integrations that connect advertiser first-party data to CTV publishers and DSPs. For brands with large CRM files or retail loyalty data, LiveRamp is the infrastructure layer that makes deterministic CTV targeting and outcome measurement possible.
Selection criteria checklist:
- Inventory quality: does the platform provide SSAI-delivered, brand-safe, premium publisher supply?
- Identity access: can you activate your first-party data or CRM list?
- Reporting transparency: are log-level exports available?
- Tech integrations: does it connect to your clean room, CDP, or BI stack?
- Measurement support: does it support third-party verification (e.g., DoubleVerify, Integral Ad Science)?
Pro Tip: Before scaling any CTV buy, request a sample of log-level impression data from your platform and reconcile it against your own analytics. If the platform cannot or will not provide log-level exports, that is a transparency red flag.
What CTV advertising costs and how to set a realistic budget
CTV pricing runs primarily on a CPM (cost per thousand impressions) model, though cost-per-completed-view (CPCV) and guaranteed/fixed-rate buys are available for specific objectives.
Pricing guidance by buying method:
- Open programmatic RTB: generally the lowest CPMs, with more inventory variance
- PMP deals: mid-range CPMs with stronger inventory quality controls
- Direct/guaranteed buys: premium CPMs, justified by placement certainty and content adjacency
CPM levels vary based on targeting granularity (a tightly defined behavioral or CRM-matched audience costs more than a broad demographic), time slot (primetime and live sports command premiums), and publisher tier (premium streaming services carry higher floors than long-tail apps).
U.S. CTV ad spend was projected at $33.35 billion for 2025, reflecting the channel’s scale. For individual brands, the practical budget question is not “how much does CTV cost?” but “how much do I need to spend to reach statistical significance in my measurement?”
Budgeting checklist:
- Pilot budget: allocate enough to reach a minimum of 500,000–1,000,000 impressions before drawing optimization conclusions.
- Test vs. scale split: run 20–30% of initial budget as a test phase across two or three inventory sources before concentrating spend.
- Funnel alignment: brand awareness campaigns can run at lower frequency targets (2–3 exposures per household per week); direct response campaigns benefit from higher frequency caps (4–5) to drive conversion.
- Incrementality budget: reserve 10–15% of your pilot budget for a holdout group to validate lift before scaling.
- Minimum pilot duration: four weeks is the floor for meaningful outcome data; eight weeks is preferable for seasonal categories.
How to integrate CTV into your omnichannel media plan
Treat CTV as a measurable, addressable component within a unified media plan, not a standalone TV experiment. Its value compounds when CTV exposure feeds into cross-screen follow-up sequences and when its data reconciles cleanly with your broader analytics stack.
A practical campaign flow looks like this:
- Audience definition: build your target segment from CRM data, first-party web analytics, or a third-party data partner. Activate it through a clean room or identity platform.
- CTV exposure: deliver brand or consideration creative to the defined audience on connected TV. Log household-level exposure data.
- Cross-screen follow-up: retarget exposed households on mobile, desktop, or digital out-of-home using the same identity graph. Serve direct response creative that references the brand story from the CTV exposure.
- Conversion and measurement: match downstream conversions (web, app, in-store) back to CTV exposure using household matching or clean room attribution. Feed results into your BI dashboard.
- Optimization loop: adjust audience segments, creative rotation, and frequency caps based on outcome data, not just completion rates.
Data sources to sync:
- CRM and email lists (for audience activation and post-exposure matching)
- Web analytics (for view-through conversion tracking)
- Mobile app data (for cross-device attribution)
- Retail POS or loyalty data (for in-store sales lift)
- Clean room or CDP (for privacy-safe data joining)
For omnichannel campaigns to work, naming conventions matter more than most teams expect. Set a campaign taxonomy before launch: consistent UTM parameters, placement naming, and log-level export schemas so CTV data reconciles with your paid search, social, and display reporting without manual cleanup.
Video marketing’s role in digital advertising extends well beyond CTV, but CTV is where video’s reach and digital’s precision converge most cleanly in a living-room context.
Risks, supply quality, and privacy considerations for U.S. advertisers
The three highest-risk areas in CTV are inventory quality and fraud, inconsistent metric definitions across publishers, and identity/attribution variance. Each is manageable with the right controls.
Inventory quality and fraud remain real concerns in programmatic CTV. Spoofed inventory, where a low-quality app misrepresents itself as a premium publisher, can inflate impression counts without delivering real viewers. Mitigation: require ads.txt and app-ads.txt compliance from all supply sources, activate a third-party verification partner such as DoubleVerify or Integral Ad Science, and monitor invalid traffic (IVT) rates in your log-level data.
Inconsistent metric definitions create reporting confusion when you run across multiple publishers. One publisher’s “completion” may be defined differently from another’s. The fix is to ingest raw log-level exports into a single analytics environment and apply your own consistent definitions rather than relying on publisher-reported dashboards.
Identity and attribution variance stems from the fact that CTV attribution uses probabilistic household matching, which produces different results depending on the identity graph each platform uses. Running an incrementality holdout test gives you a platform-independent measure of true causal lift.
On privacy: U.S. advertisers must ensure that any first-party data used for CTV targeting was collected with appropriate consent under applicable state laws. Use data clean rooms for all CRM-to-publisher matching. Never pass raw PII into ad targeting workflows. Contextual targeting is a strong privacy-safe fallback when identity signals are restricted.
Pro Tip: Request your platform’s IVT rate and ads.txt compliance rate before committing budget. A platform that cannot provide these figures on request is not ready for a brand-safe buy.
Step-by-step checklist to launch your first CTV campaign
A CTV campaign moves through five phases: plan, build, pilot, optimize, and scale.
Here is the full checklist with an 8–12 week sample timeline.
Phase 1: Plan (Weeks 1–2)
- Define your primary business outcome KPI (ROAS, CPA, sales lift, or brand lift).
- Map your audience: build your target segment from CRM, first-party data, or a third-party data partner.
- Select your buying method: programmatic DSP, PMP, or direct, based on your budget and inventory requirements.
- Set your measurement architecture: identify your clean room, attribution partner, and log-level export destination.
- Establish naming conventions and campaign taxonomy for cross-channel reconciliation.
Phase 2: Build (Weeks 2–4)
- Produce creative assets to spec: 15-second and 30-second versions, encoded to H.264/1080p, audio normalized to -24 LKFS.
- Implement impression trackers and verify they fire correctly in an SSAI environment.
- Configure your measurement pixel or household-matching integration.
- Set up your holdout group for incrementality testing (10–15% of target audience withheld from exposure).
Phase 3: Pilot (Weeks 4–8)
- Launch with 20–30% of total budget across two or three inventory sources.
- Monitor daily: completion rate, frequency, IVT rate, and pacing.
- At week six, pull a mid-flight log-level export and reconcile against your analytics platform.
Phase 4: Optimize (Weeks 6–10)
- Rotate creative if VCR drops below 85% or frequency exceeds your cap.
- Shift budget toward the inventory source showing the strongest outcome signal.
- Adjust audience segments based on household-level conversion data.
Phase 5: Scale (Weeks 10–12)
- Confirm incrementality lift from holdout test before scaling budget.
- Expand to additional publishers or increase frequency on proven segments.
- Present outcome data to stakeholders using business-outcome KPIs, not just impressions.
Optimization rules during pilot:
- Do not optimize on completion rate alone. A 95% VCR with zero downstream conversions is a creative problem, not a media success.
- Pivot creative within 10 days if early engagement signals (attention, audio-on rate) are below benchmark.
- Scale only after the holdout test confirms positive incremental lift.
How Theartistevolution approaches CTV campaigns
Theartistevolution integrates CTV strategy, creative production, and measurement into a single managed workflow, which is where most brands lose efficiency when they try to coordinate these three functions across separate vendors.
The agency’s approach starts with a measurement architecture review before any creative brief is written. Audience definition, identity activation, and outcome KPI alignment happen in the planning phase, not after launch. Creative is produced to CTV-specific specs with distinct 15-second and 30-second versions, built for big-screen composition and sound-on environments. Measurement setup includes log-level export configuration, clean room or CRM linkage, and incrementality test design.
Across brand engagement campaigns, Theartistevolution has delivered measurable lifts in brand awareness and downstream conversion rates for clients in retail, healthcare, and CPG, with campaign outcomes tied directly to business KPIs rather than media metrics. The agency’s 18-year track record spans brand development through performance measurement, giving CTV campaigns the creative depth and analytical rigor that standalone media buys rarely achieve.
For teams that want a structured starting point, a marketing assessment surfaces gaps in measurement infrastructure, identity readiness, and creative specs before a dollar of media spend is committed.
Key Takeaways
Connected TV advertising delivers measurable business outcomes when measurement architecture, identity strategy, and creative production are aligned before the campaign launches.
| Point | Details |
|---|---|
| Outcome KPIs over impressions | Over 27.4% of marketers now prioritize conversions and sales as their primary CTV KPI. |
| Programmatic dominates buying | Approximately 84% of CTV inventory is bought programmatically; DSP and PMP readiness is required. |
| Identity drives attribution | Household-level matching and clean room integrations are the foundation of CTV attribution. |
| Pilot before scaling | Run a 4–8 week pilot with a holdout group before committing full budget to any inventory source. |
| Theartistevolution as your CTV partner | Theartistevolution manages strategy, creative, and measurement as a unified service for brands ready to launch or scale CTV campaigns. |
Why CTV deserves a permanent seat in your media mix
CTV has crossed the threshold from “emerging channel” to structural part of the U.S. media diet. Streaming accounts for nearly 44% of U.S. TV time, and that share is not going back down. What strikes me most about where CTV sits in 2026 is not the scale of the channel, which is well documented, but the gap between what brands are measuring and what they could be measuring.
Most teams still report CTV performance in reach and completion terms because that is what their DSP dashboard surfaces first. The brands pulling ahead are the ones that have built the measurement infrastructure to connect CTV exposure to actual revenue, whether through clean room attribution, loyalty data matching, or incrementality testing. That is not a technology gap. It is a planning and prioritization gap.
The other underappreciated reality: CTV creative quality has a disproportionate impact on outcomes compared to other digital video formats. The non-skippable, sound-on, big-screen environment amplifies both great creative and mediocre creative. Brands that invest in CTV-specific production, rather than repurposing social assets, consistently see stronger completion-to-conversion ratios. Theartistevolution’s model of integrating creative production with measurement setup reflects exactly that conviction: the channel rewards brands that treat it seriously from the first frame to the final attribution report.
Theartistevolution builds and measures CTV campaigns end to end
Most brands approaching CTV for the first time face the same three gaps: creative not built for the big screen, measurement infrastructure not configured before launch, and no clear path from impressions to business outcomes. Theartistevolution closes all three in a single engagement.

With over 18 years managing campaigns for brands in retail, healthcare, legal, and CPG, Theartistevolution handles campaign strategy and ongoing management from audience definition and creative production through log-level measurement and optimization. You get a team that writes CTV briefs, produces to spec, configures attribution, and reports on business outcomes, without coordinating three separate vendors. The starting point for most clients is a marketing assessment that maps your current measurement readiness and identifies exactly what needs to be in place before your first CTV flight. Request yours to get a clear picture of where you stand and what a pilot would look like for your brand.
Curated sources and further reading
- CTV Advertising: A 2026 Guide — Equativ: broad overview of CTV buying mechanics, programmatic share data, and 2026 trend context.
- What Is Connected TV (CTV) Advertising? — LiveRamp: definition, household penetration data, and clean room attribution use cases.
- A Guide to CTV Ad Measurement: 10 Metrics Worth Tracking — LiveRamp: deep-dive on measurement KPIs, attribution mechanics, and identity-based matching.
- CTV Measurement Explained: KPIs & Attribution — Equativ: targeting priority data, streaming share of TV time, and cross-platform measurement guidance.
- Understand MNTN Attribution — MNTN Help: technical explanation of household-level matching and attribution mechanics.
- Connected TV: Transforming Advertising Trends — Nielsen: analyst perspective on linear/streaming convergence and cross-platform reach management.
- The Trade Desk — Unified ID 2.0: identity framework documentation for privacy-forward CTV targeting.