Buy attention, not impressions

Buy ads that get seen

Integrate attention tags for media planning to go beyond viewability by optimizing for ad formats and ad domains that are most likely to be seen.

Be seen, drive outcomes

Increase brand lift and sales lift by driving higher attention on every ad. Buying for higher attention generates higher overall views and view time, leading to better recall and outcomes.

Get higher return on ad spend

Quickly optimize your investment towards the partners, platforms, channels, and campaigns that drive higher attention, while reducing spend on lower attention inventory.

Attention, not impressions

Lumen Attention Bidding is a pre-bidding custom algorithm that buys ad units that will actually be seen. By integrating with your chosen DSP, Lumen Attention Bidding helps you drive more awareness with attention signals more effectively and efficiently.


Powered by the Lumen Attention Measurement & Planning Platform

The platform


How it works


Attention technology at scale

The Lumen Attention Engine is trained on the largest eye-tracking dataset in the world, with 650,000+ real-world eye-tracking sessions over 10+ years

We combine this human-first visual data with a dataset of 30B+ programmatic ad impressions and 300B+ social ad impressions to generate an attention model for your brand based on your ad campaign data

Lumen's Attention Bidding custom algorithm integrates with your select DSP and begins bidding based on the high-attention ad units identified by the attention model specific to your brand and objectives, driving higher attention at lower cost

Attention data is linked to DSP outcomes data to identify conversions

Attention data linked to CPM data to help you understand the true ‘cost of attention’

Attention data linked to brand lift and conversions data to help you understand the true ‘value of attention'

Patented eye-tracking platform

How the eye tracking works

Lumen’s software collects and combines two different types of data:

  • Browser extension records events on device
  • Webcam eye tracking records visual attention via front-facing camera

Data security

The software is enabled at the browser-level

No personally identifiable information (PII) collected, stored or reported

All data collection, storage and processing fully GDPR-compliant

Attention model

Combining placement and attention data

Placement data: What ads could be seen (what was viewable, in what location and for how long, etc.)

Attention data: What ads people did, in fact, see, and for how long

Predicting attention from placement data

Placement and attention data from eye-tracking data and contextual data is used to power a model of attention

Attention is measured based on a number of placement characteristics

Attention to ad predicted every 100ms of visual exposure

Placement data included in model

  • Ad format
  • % pixels in view,
  • Viewable time
  • Screen real estate
  • Scroll speed
  • Number of competing ads on screen
  • Ad positioning
  • Page geometry
  • Domain

Understanding the Attention Funnel

Lumen’s attention metrics measure how many people saw an ad and for how long

We combine these estimates into a single unit of attention: ‘attentive seconds per thousand impressions’ = % view x av. eyes-on dwell time x 1000 impressions


Attentive seconds across media


Lumen conducts bespoke media evaluation studies across a variety of media, enabling brands and media owners to understand the true value of different media opportunities

Use the same ‘attention currency’ to compare between TV, BVOD, all forms of digital, print, OOH and POS

Logical next steps