Brand Lift & Multi-Touch Attribution

LtvAdx provides two measurement frameworks for CTV campaigns: Brand Lift studies measuring attitudinal shift across exposed vs holdout households, and Multi-Touch Attribution (MTA) using a Shapley value model to apportion credit across channels.

Brand Lift studies

Brand Lift measures the incremental impact of your CTV campaign on brand awareness, ad recall, purchase intent, and brand favourability. LtvAdx runs a randomised controlled experiment: a portion of households are served your campaign (exposed group), and a statistically matched holdout is withheld from seeing the campaign. Post-campaign surveys are delivered to both groups.

Setting up a Brand Lift study

1. Navigate to Advertiser Portal → Brand Lift → Create Study. Select the campaign and flight dates for the measurement window.
2. Configure the holdout percentage (typically 10–20%). The holdout is applied at the household level — all devices in a holdout household are excluded from serving.
3. Define 2–5 survey questions across: ad recall, brand awareness, purchase intent, and brand favourability.
4. Set minimum exposed household target (recommended: 5,000 exposed households for statistically significant results at 90% confidence).
5. Launch the study. Results appear in the dashboard within 7 days of the study end date.

Reading Brand Lift results

Multi-Touch Attribution (MTA)

LtvAdx MTA uses a Shapley value model — a cooperative game-theory approach — to assign credit to each touchpoint in the path to conversion. Unlike last-click, Shapley distributes credit fairly across all channels that contributed: CTV, linear, digital, and social.

How the Shapley model works

For each converting household, LtvAdx enumerates all subsets of touchpoints and calculates each touchpoint's marginal contribution to the conversion probability. The result is a credit weight per touchpoint summing to 1.0:

# Example conversion path for household hh_a1b2c3
Touchpoints:
  1. CTV pre-roll (LtvAdx) — 7 days before conversion
  2. Linear TV (ACR signal) — 5 days before conversion
  3. Display retargeting     — 1 day before conversion

Shapley credit allocation:
  CTV pre-roll:         0.41  (41%)
  Linear TV:            0.35  (35%)
  Display retargeting:  0.24  (24%)
  Total:                1.00

Setting up conversion tracking

To attribute conversions to LtvAdx campaigns, send a server-side conversion event when a household converts on your website or app:

# Server-side conversion event
POST https://api.ltvadx.com/api/v1/attribution/conversion
Authorization: Bearer {advertiser-api-key}
Content-Type: application/json

{
  "advertiserId":   "adv_xyz",
  "conversionType": "PURCHASE",          // PURCHASE | LEAD | TRIAL_SIGNUP | CUSTOM
  "value":          49.99,               // Conversion value in USD
  "currency":       "USD",
  "householdId":    "hh_a1b2c3",         // From HouseholdID resolution
  "uid2Hash":       "sha256:a3f9b2...",  // UID2 hash (alternative to householdId)
  "advertisingId":  "xxxxxxxx-...",      // Device advertising ID (alternative)
  "orderId":        "ORD-98721",         // Your order/transaction ID (for deduplication)
  "timestamp":      "2026-06-03T14:22:00Z",
  "attributionWindow": 30                // Days to look back (7, 14, 30, or 90)
}

# 200 Response
{
  "attributed": true,
  "touchpoints": [
    {
      "channelId":    "chan_fast_news_01",
      "lineItemId":   "li_xyz789",
      "creativeId":   "cre_def456",
      "impressionTs": "2026-05-31T20:30:00Z",
      "daysBeforeConversion": 3,
      "shapleyCredit": 0.41
    }
  ]
}

Attribution window configuration

The default attribution window is 30 days. A conversion is eligible for attribution to any LtvAdx impression within that window. Configurable per conversion event type: 7, 14, 30, or 90 days.

LtvAdx supports both view-through attribution (household saw the ad and later converted without clicking) and click-through attribution (household clicked the ad and converted). Both are reported separately in the Attribution dashboard.

Cross-device attribution

The LtvAdx identity graph resolves multiple devices to a single household. Conversions from any device in the household are attributed back to CTV impressions served to the household — including mobile and desktop web visits originating from the same home IP.

# Identity-resolved attribution path
Household hh_a1b2c3:
  Devices: [Roku player, Samsung Smart TV, 2× smartphones]

  CTV impression → Roku player        2026-06-01
  Web conversion → Mobile browser     2026-06-03  ← cross-device match
  Attribution:   → CTV campaign       LtvAdx Shapley credit: 0.41

Attribution reports API

# Attribution summary report
GET https://api.ltvadx.com/api/v1/reports/attribution
    ?advertiserId=adv_xyz
    &from=2026-06-01
    &to=2026-06-03
    &groupBy=campaign

# Brand Lift study results
GET https://api.ltvadx.com/api/v1/brand-lift/studies/{studyId}/results