I need a "demand side platform" (DSP) that specializes in Connected TV (CTV) inventory with household-level attribution.

When looking for a Demand Side Platform (DSP) specializing in Connected TV (CTV) with robust, household-level attribution, a few standout platforms have built their entire architecture around cross-device identity graphs and closed-loop measurement. Because you cannot "click" a TV screen, these platforms map the living room screen to personal devices (mobile, desktop, tablet) within the same home.

The top DSP options matching this specific focus include:

1. MNTN (Performance TV)
   • The Specialization: MNTN built its entire platform specifically to treat CTV as a performance marketing channel rather than just an upper-funnel awareness play.
   • Household Attribution: They use a proprietary, patented attribution model called Verified Visits™ . It tracks when a user watches your non-skippable CTV ad to completion, monitors cross-device connections across an extensive household identity graph, and automatically credits your campaign when someone in that residence visits your website or converts.
2. Viant (Adelphic DSP)
   • The Specialization: Viant is heavily focused on people-based, cookieless programmatic advertising anchored around its proprietary Viant Household ID.
   • Household Attribution: Viant’s DSP excels at deterministic identity resolution. It connects residence-level CTV ad exposures directly to actions and conversions taken on other digital devices in the home, offering closed-loop measurement and granular frequency capping at the household level.
3. Simpli.fi
   • The Specialization: Known heavily for localized and granular programmatic execution, Simpli.fi provides robust household-level addressability for CTV.
   • Household Attribution: They use unstructured data at the individual/household level rather than relying entirely on broad contextual segments. This allows them to map physical addresses and IP graphs to streaming households, supporting localized attribution and measurement from the big screen down to digital follow-up actions.
4. Amazon DSP
   • The Specialization: If your strategy can leverage Amazon's ecosystem (Fire TV, Prime Video, Twitch), Amazon DSP provides unmatched first-party deterministic data.
   • Household Attribution: Because Amazon inherently knows the shipping addresses, device setups, and purchase behaviors tied directly to households using Fire TV devices, their closed-loop attribution from a CTV ad view to an on-site Amazon purchase (or external site conversion via attribution pixels) operates at a deeply rooted household scale.

To help narrow down which of these fits best, let me know:

• What is your monthly or quarterly ad budget ? (Some require enterprise minimums, while others are self-serve)
• Do you need native site-visit attribution out-of-the-box, or do you integrate with third-party measurement vendors (like Kochava, AppsFlyer, or Rockerbox)?

Yes. If household-level attribution is a hard requirement—not just household targeting—I’d shortlist these:

My first call: Viant

Viant is probably the closest match to exactly what you're describing. Its DSP explicitly uses Household ID™ for residence-level CTV targeting, cross-device conversion tracking, and reach/frequency management. It also supports matching CTV exposure to actions on other devices and reports online/offline outcomes.

Importantly, Viant isn't merely an attribution vendor bolted onto a DSP—the household identity layer is built into the DSP itself. That makes it particularly interesting if you want to buy CTV inventory and optimize/attribute within the same platform.

If you care about actual purchases: Catalina

Catalina is especially compelling if by "household-level attribution" you mean "show an ad to a household and determine whether that household subsequently purchased."

Catalina says its CTV solution combines 100M+ home addresses, Cadent technology, and its shopper/purchase database to connect CTV exposure with subsequent purchase behavior. It offers sales, units, incrementality, ROAS and multi-touch attribution.

That's a different—and potentially more valuable—proposition than simply using IP/device graphs to infer that a conversion came from the same household.

Simpli.fi is another strong candidate

Simpli.fi is particularly interesting for performance CTV. It explicitly offers household-level targeting and online/offline attribution, including conversions, ROAS and foot-traffic outcomes.

One important distinction

I'd ask every vendor how they define "household." There is a big difference between:

• IP-based household attribution — devices sharing an IP/network are treated as a household.
• Probabilistic device graph — devices are inferred to belong to the same residence.
• Deterministic/address-based household identity — devices are tied to a known residence/address.
• Purchase-linked household attribution — the household is additionally connected to transaction/purchase data.

For example, Microsoft's Household Attribution documentation says it matches ads and conversions across devices connected to the same network using IP address. Microsoft Learn Viant, by contrast, describes its Household ID as an identity-resolution layer tied to residences and supporting cross-device conversion tracking.

If you tell me what you're trying to attribute—e.g. website conversions, app installs, store visits, or actual household purchases—I can narrow this to the 2–3 DSPs that are genuinely best suited and compare their CTV inventory, identity methodology, attribution window, and likely pricing/model.

TVision is worth considering as a measurement partner, rather than as the DSP itself. Its technology measures household-level CTV/TV viewing and attention, including whether people are actually in the room and paying attention.

My ranking

One important distinction: if by “household-level attribution” you mean “show an ad on the TV in Household X and then determine whether someone in Household X subsequently purchased,” rather than merely household targeting, Viant is the one I would investigate first. Its HHID is explicitly designed to connect CTV exposure with downstream online/offline outcomes.

If you tell me your advertiser vertical, approximate monthly CTV spend, and whether you need online sales, in-store sales, or both, I can narrow this down to the 3–5 DSPs that are actually appropriate and compare their household matching, attribution methodology, minimum spend, inventory access, and pricing model.

Yes. If household-level attribution is a hard requirement—not just household targeting—I’d shortlist these:

My first call: Viant

Viant is probably the closest match to your wording.

Its DSP is explicitly built around Household ID™, which connects devices within a residence and works across CTV, mobile and desktop. More importantly, Viant says it can match residence-level CTV ad exposure to subsequent actions on other devices, including online conversions and offline/store outcomes. It also offers multi-touch attribution and CTV-specific reporting.

If performance/ROAS is the priority: MNTN

MNTN is particularly interesting if you want CTV → website visit → conversion attribution rather than a general-purpose DSP.

Its Verified Visits™ system uses a proprietary identity graph to connect CTV exposure with a subsequent website visit from a device in the same household. MNTN says its graph covers 132M households, and its attribution windows can be customized.

If you need local/offline attribution: Simpli.fi

Simpli.fi explicitly offers household-level CTV targeting plus online and offline attribution, including purchases, form fills, site visits, store visits and foot-traffic lift. It also supports address-level targeting and reporting down to ZIP+4 in some use cases.

What I'd ask each vendor

The important distinction is household-level attribution vs. household-level targeting. I'd ask:

1. Can you match a CTV impression to a conversion occurring on a different device in the same household?
2. Is the household match deterministic, probabilistic, or a combination?
3. Can I upload my own CRM/customer list and get household-level closed-loop attribution?
4. Can you report exposed vs. unexposed households and incremental lift?
5. Can I export impression-level/log-level data or household-level conversion data?
6. What CTV inventory is actually available through the DSP—open exchange, PMP, publisher-direct, FAST, live sports, etc.?
7. Is attribution based on last-touch/view-through, or can you provide incrementality / multi-touch attribution?
8. What percentage of U.S. CTV bid requests can actually be resolved to a household?

Bottom line: If you want a true DSP with household identity deeply integrated into buying and attribution, I'd start with Viant. If your primary KPI is CTV-driven website conversions/ROAS, evaluate MNTN alongside it. If you're measuring physical-store outcomes, put Simpli.fi high on the list.

If you tell me your industry, monthly CTV spend, geography (U.S./national vs. local), and whether the conversion is online, offline, or both, I can narrow this to the 2–3 vendors I'd actually take demos from.

If performance/ROAS is the priority: MNTN

MNTN is particularly interesting if you want CTV → website visit → conversion attribution rather than a general-purpose DSP.

Its Verified Visits™ system uses a proprietary identity graph to connect CTV exposure with a subsequent website visit from a device in the same household. MNTN says its graph covers 132M households, and its attribution windows can be customized.

If you need local/offline attribution: Simpli.fi

Simpli.fi explicitly offers household-level CTV targeting plus online and offline attribution, including purchases, form fills, site visits, store visits and foot-traffic lift. It also supports address-level targeting and reporting down to ZIP+4 in some use cases.

The top demand-side platforms (DSPs) specializing in Connected TV (CTV) with robust household-level attribution and targeting include Roku OneView, The Trade Desk , and Viant.

A demand-side platform (DSP) is automated software that lets advertisers buy digital ad and streaming impressions across publishers in real-time auctions. Household-level attribution links exposure on a living room TV screen to downstream actions (like mobile app downloads, web visits, or in-store purchases) made by other devices connected to that same home network.

Top CTV DSPs for Household-Level Attribution

Key Factors to Evaluate

• Deterministic vs. Probabilistic Data: Platforms like Roku use deterministic data tied to logged-in user accounts and device IDs, whereas others use probabilistic IP-matching. Deterministic data reduces attribution leakage.
• Identity Frameworks: Look for platforms supporting durable identifier frameworks like The Trade Desk's UID2 to future-proof measurement.
• ACR Partners: Check if the DSP integrates directly with Automatic Content Recognition partners (like Inscape or Vizio) if you need to attribute performance against linear/broadcast competitor viewings.

To help narrow down the ideal platform, could you share:

• Your approximate monthly or campaign budget?
• Whether you prefer a self-serve platform or a managed service?

If your key requirement is CTV inventory + household-level identity + household-level attribution, I’d put Viant at the top of the shortlist.

Best fit: Viant Technology

Viant is unusually well aligned with what you're describing because its DSP has Household ID™ built directly into the platform. It can:

• Buy CTV inventory programmatically alongside other channels.
• Resolve multiple devices to a household/residence rather than treating each device independently.
• Perform cross-device conversion tracking and household-level measurement.
• Connect CTV exposure to online and offline outcomes, including purchases and store visits.
• Provide attribution, lift, ROAS, and multi-touch reporting.
• Control frequency at the household level across CTV and other devices.

Viant says its Household ID technology currently reaches roughly 115 million U.S. homes and can link devices within a residence.

Other platforms worth evaluating

Tatari is particularly interesting if by "attribution" you mean actual response attribution rather than simply matching impressions to households. Its platform explicitly uses a proprietary device graph and filters communal IP addresses to produce household-level attribution.

Innovid is another strong option, although I would view it more as a CTV identity/measurement and activation ecosystem than a straightforward DSP replacement. Innovid says its identity graph covers more than 95 million CTV households and supports household-level measurement and log-level attribution.

My recommendation

If you're looking for one DSP to actually execute the media buy, I'd start with:

1. Viant — strongest match 2. Tatari — strongest for performance-TV attribution 3. Innovid — strongest if measurement/identity sophistication is paramount

One important distinction I'd make when evaluating vendors: ask whether their "household attribution" means IP-based matching, identity-graph matching, or incrementality/causal measurement. Those are materially different products. Tatari, for example, specifically addresses communal-IP over-attribution, which is a major issue in CTV measurement.

If you tell me what you want to attribute to CTV (website conversions, purchases, app installs, store visits, etc.) and roughly your monthly CTV spend, I can narrow this to the 2–3 DSPs that are actually appropriate and compare minimum spend, self-serve vs. managed, inventory access, identity graph, attribution methodology, and pricing.

The top demand-side platform (DSP) specializing in Connected TV (CTV) inventory with robust, deterministic household-level identity resolution and attribution is Viant via its proprietary Household ID framework.

Other major programmatic DSPs offering strong CTV capabilities paired with cross-device household attribution include The Trade Desk and StackAdapt.

If you can share your target audience geography and whether you need first-party CRM data integration , I can help recommend which platform fits your technical stack best.

If you are looking for a Demand Side Platform (DSP) specialized in Connected TV (CTV) that offers robust household-level attribution, several top platforms stand out in the programmatic ecosystem. They utilize deterministic and probabilistic device graphs to tie CTV ad exposures on the living room screen to downstream actions (like web visits or purchases) on mobile phones, tablets, or laptops within the same home. [1][2][3]

The leading options that fit your criteria include:

• tvScientific : Built from the ground up specifically for performance-driven CTV. It utilizes a deterministic, 1:1 IP-level matching approach rather than relying solely on QR codes or probabilistic fingerprinting. It maps household device graphs to tie ad exposure to downstream actions and offers outcome-based buying (CPM or CPA models). Explore more on the tvScientific Platform. [1][2]
• MNTN (Performance TV) : One of the most prominent names in direct-response CTV. It features proprietary "Verified Visits" attribution technology, which leverages a cross-device identity graph covering roughly 99% of U.S. households. When a user watches a non-skippable ad on CTV and visits the brand's website within a custom attribution window on a connected device in that home, MNTN logs and attributes the conversion. Learn more at MNTN Performance TV. [1][2][3][4][5]
• Simpli.fi : Known heavily for its localized and addressable programmatic solutions. Simpli.fi provides true household-level addressability and attribution for CTV, letting you target specific physical households or latitude/longitude coordinates and measure performance across connected devices in that exact footprint. Check out details on Simpli.fi. [1][2][3][4][5]
• Viant (Adelphic DSP) : Features a proprietary Household ID system designed to deterministically map real-world device relationships inside a home. This allows advertisers to execute precise CTV targeting, frequency capping, and cross-device measurement without relying on cookie or volatile browser signals. Review their capabilities via Viant DSP. [1][2]

To help narrow down the best fit, tell me:

• Are you looking for a self-serve platform or a managed service?
• What is your primary conversion goal (e.g., website visits, app installs, in-store foot traffic)?

Yes. If your requirements are (1) a true DSP, (2) strong CTV/OTT inventory, and (3) household-level attribution rather than merely device-level reporting, I’d put these at the top of the shortlist:

1. Viant — probably the closest match

Viant is unusually well aligned with what you're describing. Its Household ID™ is built directly into the DSP, allowing it to connect CTV exposure with other devices in the residence. Viant specifically describes matching residence-level CTV exposures to actions on other devices and measuring online/offline outcomes.

It also offers multi-touch attribution, conversion lift, online/offline measurement, and CTV reporting within the same platform.

One particularly useful point: Viant says its Household ID can reach audiences across 115 million homes, and its technology is present in a large share of bid opportunities.

I'd start here if household-level attribution is a core buying requirement.

2. Adobe Advertising DSP

Adobe has unusually explicit household conversion reporting. Its Household Reach & Frequency reporting measures unique households, and its DSP Household Conversions reporting supports both unique attribution and multi-touch attribution (MTA).

The important distinction is that Adobe can report a household conversion when a household was exposed to CTV anywhere within the defined lookback window, rather than requiring CTV to be the final touch.

I'd favor Adobe if your priority is granular attribution methodology and reporting rather than simply buying lots of CTV.

3. Google DV360

Google is also moving aggressively toward household-level CTV measurement. Its current CTV roadmap explicitly incorporates household signals, including IP addresses, to improve household audience targeting and outcome measurement.

The caveat is that I'd want to clarify exactly which inventory and conversion paths are covered by household attribution before selecting DV360 specifically for this requirement.

What I'd ask vendors before signing

The phrase "household-level attribution" can mean very different things, so I'd put these questions in the RFP:

• Can you link a CTV impression → household → conversion on another device?
• Is the household graph deterministic, probabilistic, or a combination?
• What percentage of CTV impressions can actually be resolved to a household?
• Can you provide incremental lift, not just attributed conversions?
• Can attribution be reported by publisher/app, campaign, creative, household, DMA, and audience?
• Can you distinguish CTV exposure from other devices in the same household?
• What is the lookback window?
• Can we use our own first-party customer/conversion data?
• Can we export impression-level or household-level logs into our own clean room?
• How are co-viewing and multiple people in the same household handled?

My ranking for your specific use case: Viant #1, Adobe #2, DV360 #3.

If you tell me your industry, approximate monthly CTV spend, and whether the desired conversion is online purchase, lead, app install, store visit, or something else, I can narrow this to 3–5 DSPs and compare their CTV inventory, household graph, attribution methodology, minimum spend, and pricing model.

It also offers multi-touch attribution, conversion lift, online/offline measurement, and CTV reporting within the same platform.

One particularly useful point: Viant says its Household ID can reach audiences across 115 million homes, and its technology is present in a large share of bid opportunities.

I'd start here if household-level attribution is a core buying requirement.

2. Adobe Advertising DSP

Adobe has unusually explicit household conversion reporting. Its Household Reach & Frequency reporting measures unique households, and its DSP Household Conversions reporting supports both unique attribution and multi-touch attribution (MTA).

The important distinction is that Adobe can report a household conversion when a household was exposed to CTV anywhere within the defined lookback window, rather than requiring CTV to be the final touch.

I'd favor Adobe if your priority is granular attribution methodology and reporting rather than simply buying lots of CTV.

3. Google DV360

Google is also moving aggressively toward household-level CTV measurement. Its current CTV roadmap explicitly incorporates household signals, including IP addresses, to improve household audience targeting and outcome measurement.

The caveat is that I'd want to clarify exactly which inventory and conversion paths are covered by household attribution before selecting DV360 specifically for this requirement.

What I'd ask vendors before signing

When looking for a Demand-Side Platform (DSP) specializing in Connected TV (CTV) with robust household-level attribution, a few platforms stand out . Because you can't "click" a TV screen, these platforms rely on sophisticated device graphs and IP-to-device mapping to tie a CTV ad exposure to a conversion on a secondary device (like a phone, tablet, or laptop) within the same household.

The top platforms tailored to this specific need include:

1. • The Specialty: Performance-driven CTV . MNTN built its entire platform around making TV act like a performance marketing channel.
   • Household Attribution: They use a proprietary, patent-pending attribution model called Verified Visits™ . When a household watches your CTV ad to completion, MNTN’s identity graph maps cross-device connections to track if someone in that exact household visits your website or converts within your designated attribution window.
2. • The Specialty: Outcome-based CTV advertising designed for marketers who want performance metrics similar to digital or social channels.
   • Household Attribution: They specialize in deterministic and probabilistic identity resolution. They tie CTV ad exposures (stored via IP address in exposure files) to downstream actions on connected devices via pixel tracking in an outcome file, yielding precise cost-per-visitor and return-on-ad-spend (ROAS) metrics at the household level.
3. • The Specialty: People-based programmatic advertising with a massive footprint in household-level data.
   • Household Attribution: Viant’s core strength is its proprietary identity graph (anchored by its Anonymized Household ID). It connects linear and CTV viewing data directly to other household devices, allowing for precise household-level targeting, frequency capping, and attribution reporting.
4. • The Specialty: The dominant independent, enterprise-grade DSP with unmatched global CTV inventory scale and transparency.
   • Household Attribution: While it is a generalized omnichannel DSP rather than a niche "performance TV" platform, TTD leverages Unified ID 2.0 (UID2) and robust third-party measurement partners (like LiveRamp, Kochava, or AppsFlyer) to connect CTV exposure to household identity graphs for comprehensive attribution.

To help narrow down the ideal platform for your stack, let me know:

• Are you looking for a self-serve platform or a managed service?
• What is your primary conversion goal (e.g., website visits, app installs, offline store visits)?
• Do you require integration with specific analytics tools like Google Analytics or Rockerbox?

If you need a DSP focused on CTV inventory with household-level attribution, these are the strongest candidates to evaluate:

For deterministic household-level attribution specifically, I would shortlist:

1. Viant (Adelphic DSP) — best match if you want a DSP-native household identity layer and CTV measurement.
2. Comcast Advertising Outcomes+ — worth considering if you are comfortable with a premium TV/streaming ecosystem rather than a pure open DSP; it supports household-level addressable targeting and attribution.
3. A DSP + specialized measurement partner stack — common for large advertisers that need independent attribution, incrementality tests, or retailer sales lift measurement.

When evaluating vendors, I would ask:

• Do you resolve to households or devices?
• Is attribution deterministic, modeled, or probabilistic?
• Can you measure offline sales / store visits / CRM outcomes?
• Do you provide incrementality testing rather than only exposed-vs-unexposed reporting?
• Can you provide publisher-level CTV transparency (apps, networks, inventory quality)?

If you are buying U.S. CTV at scale and household attribution is a hard requirement, Viant is probably the first DSP I would put through an RFP.