Published

How a DOOH Ad Platform Replaced Unsellable Inventory

An ad platform team layered a custom image-expansion model and a 96%-accurate moderation algorithm into uploads in two weeks — turning every image into a billboard-ready, revenue-generating ad.

Significant Revenue Lift

From unsellable inventory

6–12 months

Implementation Time

Not disclosed

Project Cost

the challenge

A digital out-of-home advertising platform was losing revenue every time an advertiser uploaded an image. If dimensions didn't match available billboard formats, the ad couldn't run — and inventory went unsold. Making it worse, human moderators reviewed every creative manually before it could go live, creating a slow, expensive bottleneck between upload and launch. Every hour of delay was another hour of unmonetized inventory sitting idle.

what they built

Tenex built a custom AI model that automatically expands any uploaded image to fit any available billboard format — turning every upload into a viable ad unit. Alongside it, they developed a proprietary content moderation algorithm that handles the first review layer with 96% accuracy, dramatically reducing reliance on manual review. The system deployed in just two weeks. What started as a revenue-recovery tool became a platform differentiator: every image that walks in the door now walks out as revenue-generating inventory, with compliance handled before a human ever needs to look.

Tenex identified two compounding revenue problems: dimension mismatch leaving billboard inventory unsold, and slow manual moderation delaying ad launches. Rather than patching the existing process, they built two parallel AI systems. The first was a custom AI image expansion model that takes any uploaded creative and automatically generates versions sized for every available billboard format — turning previously unusable uploads into viable ad inventory. The second was a proprietary content moderation algorithm built to handle the first review layer with 96% accuracy, dramatically reducing the volume reaching human reviewers and accelerating time-to-launch. The two systems were designed to work together: an image uploads, expands into all formats, passes first-pass moderation, and surfaces for human review only when flagged. Both systems were built and deployed in two weeks. The immediate impact was the monetization of previously unsellable inventory, generating a significant revenue lift across the platform. What began as a revenue-recovery project became a differentiated platform capability.

best fit for

Digital advertising platforms, DOOH networks, and ad-tech companies struggling with creative compliance bottlenecks or inventory utilization gaps.

Ai ROLE

Not shared

impact

Significant Revenue Lift

Previously unsellable inventory — ads that couldn't run due to dimension mismatches — became monetizable, increasing revenue per upload across the platform.

96% Moderation Accuracy

A custom-built AI moderation algorithm achieved 96% accuracy on first-pass content review, replacing a slow manual process without sacrificing compliance standards.

2 Weeks to Deploy

The full solution — image expansion model plus moderation algorithm — went from concept to live production in under two weeks.

implementation complexity

Not shared

Arman Hezarkhani

Co-founder & Managing Partner

Tenex

Managing Partner at Tenex, leading enterprise AI transformation for mid-market and growth companies across product, process, and people to drive efficiency, margin, and growth.

Get an intro

Talk to this team

industry

Media & Entertainment

business organization

Operations

Marketing

AI TYpe

Computer Vision

Generative Design & Content

Process Automation (RPA + AI)

value type

Revenue Growth

Cost Reduction

Time Savings

Headcount Avoidance

other(firm) Case studies

frequently asked questions

How did a mid-sized digital out-of-home ad platform turn unsellable inventory into revenue with computer vision?

The mid-sized ad platform built two AI systems that work together: a custom image-expansion model that resizes any uploaded creative to fit every available billboard format, and a moderation algorithm that handles first-pass content review. Previously unusable uploads became viable ad units and compliance was handled before a human stepped in, producing a significant revenue lift across the platform.

What AI tools and models did the ad platform use?

The platform used a custom AI image-expansion model that automatically resizes uploads to any billboard format and a proprietary content-moderation algorithm that handles first-pass review at 96% accuracy. The approach combined computer vision, generative design, and process automation.

What results did the ad platform achieve?

Three outcomes: a significant revenue lift as previously unsellable inventory became monetizable, 96% accuracy on first-pass automated content moderation, and a full solution deployed in under two weeks.

How long did it take to deploy the system?

Time to results was under four weeks. The full solution — image-expansion model plus moderation algorithm — went from concept to live production in under two weeks.

Who is this computer vision approach best for?

Digital advertising platforms, DOOH networks, and ad-tech companies struggling with creative-compliance bottlenecks or inventory-utilization gaps.

Have a similar challenge?

Ask whether this would work for you, or describe what you're trying to solve.

TELL US WHAT YOU'RE EXPLORING

How a DOOH Ad Platform Replaced Unsellable Inventory

Significant Revenue Lift

6–12 months

Not disclosed

Significant Revenue Lift

96% Moderation Accuracy

2 Weeks to Deploy

Arman Hezarkhani

How One Restoration Firm Clawed Back $500 a Proposal

How a Healthcare Staffer Cut Time-to-Hire 15 to 5 Days

How a Mid-Market CFO Halved Board Reporting Time

Have a similar challenge?