The Intelligence Layer ingests OOH movement data, programmatic signals, and CRM events into a unified audience graph with under 6-hour latency. Enables true cross-channel attribution with 8.3x average ROAS uplift when DOOH is combined with sequential digital retargeting.
Programmatic Intelligence: Real-Time Audience Data for DOOH + Digital Convergence
01The DOOH Intelligence Gap
Out-of-home advertising has always suffered from a fundamental measurement problem: you cannot directly observe who saw your billboard. Traditional OOH planning relied on census-based footfall estimates and panel-based reach modelling — accurate to within 20–30% at best.
Programmatic DOOH has changed the medium. But most operators are still planning and measuring it like traditional OOH — leaving a significant intelligence gap between impression delivery and audience understanding.
The Njinga Intelligence Layer was built to close that gap: connecting real-time movement data, device-level audience signals, and CRM-based identity resolution into a unified audience graph that enables genuine cross-channel attribution.
< 6 hrs
LATENCY FROM IMPRESSION DELIVERY TO AUDIENCE SEGMENT REFRESH
02The Audience Intelligence Stack
Effective programmatic DOOH intelligence requires four data layers operating simultaneously:
Layer 1: Location Intelligence — aggregated, anonymised mobile device movement data linked to DOOH screen proximity. This identifies which device types (and by inference, audience cohorts) were within viewable range of each screen during impression delivery.
Layer 2: Identity Resolution — probabilistic matching between on-location device signals and known audience profiles from CRM, first-party web, and paid media interaction data. This converts anonymous impressions into qualified audience segments.
Layer 3: Digital Retargeting Activation — identified audience cohorts who received OOH impressions are activated in digital channels (Meta, Google, programmatic display) within the 6-hour latency window — creating true cross-channel sequential messaging.
Layer 4: Conversion Attribution — CRM and transaction data is matched against both OOH impression cohorts and subsequent digital touchpoints, enabling multi-touch attribution that includes OOH contribution.
THE PROGRAMMATIC INTELLIGENCE STACK
LOCATION INTELLIGENCE
Device proximity · Dwell time · Movement patterns
IDENTITY RESOLUTION
Probabilistic matching · CRM alignment · Segment build
DIGITAL ACTIVATION
Sequential retargeting · Cross-channel delivery · < 6hr latency
CONVERSION ATTRIBUTION
Multi-touch model · CRM matching · OOH contribution score
03Screen Selection Intelligence
Not all DOOH screens are equal — and the difference is not just audience volume. Screen context, dwell environment, and audience intent state are the variables that drive campaign effectiveness.
The Njinga Intelligence Layer scores DOOH inventory across four dimensions:
Audience Density Score — the volume and quality of audience cohorts within the screen's viewable radius during your target daypart, validated against first-party conversion data from previous campaigns.
Intent Environment Index — a contextual score based on the commercial environment surrounding the screen. A screen outside a high-end automotive dealership scores differently for a luxury automotive campaign than the same reach volume in a transit environment.
Cross-Channel Amplification Score — how effectively audiences reached via this screen respond to subsequent digital retargeting. Screens whose audiences show high response rates to follow-on digital messaging score higher regardless of raw reach.
Attribution Confidence Rating — the degree to which impressions delivered via this screen can be matched to identifiable audience cohorts, enabling reliable contribution modelling.
8.3x
AVERAGE ROAS UPLIFT WHEN DOOH IS COMBINED WITH SEQUENTIAL DIGITAL RETARGETING VS. DIGITAL-ONLY
04The Convergence Attribution Model
The core value proposition of DOOH + digital convergence is not reach extension — it is attribution maturity. For the first time, OOH media can be assigned a measurable contribution to pipeline and revenue.
The Njinga convergence attribution model operates in three stages:
Stage 1: Cohort Construction — audiences exposed to OOH impressions are cohorted by screen, daypart, creative, and audience segment. These cohorts form the basis of all subsequent attribution analysis.
Stage 2: Digital Journey Mapping — each cohort is tracked across subsequent digital touchpoints — paid search, social, programmatic display, direct web — to construct a cross-channel journey map that includes OOH as a first or mid-funnel touchpoint.
Stage 3: Conversion Contribution Scoring — each channel's contribution to conversion is calculated using a probabilistic multi-touch model, validated against incrementality test results where available. OOH contribution is expressed as a weighted percentage of total conversion attribution.
05Deployment: Connecting the Intelligence Layer
Deploying the Njinga Intelligence Layer for DOOH + digital convergence requires four integration points:
→DOOH SSP Connection — direct API integration with Vistar, Hivestack, or your existing programmatic DOOH buying platform. This enables real-time impression logging and screen-level performance data ingestion.
→Location Data Partner — aggregated, privacy-compliant mobile device movement data feed aligned to your target market geographies. Njinga has validated partnerships for South African and sub-Saharan African markets.
→CRM / Identity Data — first-party customer and lead data for probabilistic audience matching. HubSpot and Salesforce have native connectors; custom CRM environments are supported via API.
→Digital Activation Channels — Meta, Google DV360, and programmatic DSP connections for audience activation within the retargeting latency window.
Full integration and data flow validation can be completed within 14 days on the Scale tier.
14 days
TYPICAL DEPLOYMENT TIMELINE FROM INTEGRATION TO FIRST ATTRIBUTION REPORT
Ready to deploy this in your operation?
Book a Growth Architecture session. We'll map your current operational model and show you exactly where the OS delivers the highest-impact improvements.
RATE THIS INSIGHT
Was this resource valuable?