OVBO
predictive overbooking

Taking OB strategy
to the next level.

OVBO predicts no-show passenger by passenger, then sends an optimal lid straight to your PSS every day. Your RM team sets the rules and gets back to the core business.

  • 29,355
    Optimized flights
  • 5,471,892
    Optimized passengers
  • +15 years
    Revenue management experience

the challenge

Overbooking is a margin decision made with incomplete information.

$635

Average compensation paid to passengers involuntarily bumped from flights in the US, 2018.

  • Denied boarding is expensive in more ways than one

    Negative PR, unhappy passengers, frustrated airport staff — and compensation payments on top.

  • No-show behaviour is hard to read

    Without passenger-level prediction, balancing revenue against customer satisfaction stays guesswork.

  • Playing it safe costs real revenue

    Conservative overbooking leaves seats unsold on flights that could have gone out full.

what OVBO does

Automatically optimize your overbooking strategy and stay focused on core RM work.

Efficient

OVBO optimizes every flight considering:

AI no-show predictionfor each passenger on the flight Potential overbooking revenue Denied boarding cost

Automatic

Full integration with your PSS, sending daily lid updates.

No manual tasksfor the RM team Daily data refresh Analytics for decision-making

Flexible

A clear interface where RM teams set the parameters that matter.

Lid limits and flight rules Peak days Denied boarding cost by market

the data

PNR data explains no-show more than we believe.

Long AP passengers tend to show up less at the airport

1,000,95 0,900,85 1020 304050
Show-up rate by advance purchase, in days before departure.

Passengers whose PNR was modified also show up less

~0,96
Regular PAX
~0,81
PAX with PNR modifications
Show-up rate, regular bookings vs. modified bookings.

+20 features available in PSS data to understand no-show behaviour

Graphs and statistics shown here are based on trends observed among our customers and are provided for illustration only. Actual results vary by airline and by operation.

how it works

From PSS data to a lid recommendation, every day.

  1. 01

    Processing

    OVBO processes and stores airline data from the PSS, plus custom integrations where needed.

  2. 02

    Training

    Flown data and OVBO-calculated features train an AI model to predict future no-show.

  3. 03

    Prediction

    Each day the model scores every passenger’s no-show probability across all upcoming flights.

  4. 04

    Lid recommendation

    Using the RM analyst’s settings, the optimizer produces an optimal lid for every flight.

  5. 05

    PSS

    OVBO writes the lid recommendation straight back to the PSS.

accuracy

Numbers speak for themselves.

OVBO’s error distribution sits clearly tighter than a traditional approach. Less spread means you can push the lid with more confidence.

Absolute error mean

~1,5%
OVBO
~3,5%
Traditional approach
Lower is better.

Error (%) distribution — prediction vs. real boarded

OVBO sigma 0.038 mean −0.003 Traditional sigma 0.069 mean 0.0042 −0,4−0,2 0,00,20,3
Narrower curve, closer to zero — fewer surprises at the gate.

Graphs and statistics shown here are based on trends observed among our customers and are provided for illustration only. Actual results vary by airline and by operation.

expected impact

Increase RASK by more than 3% with OVBO.

OVBO performs best where there is most to win: high load factors and higher no-show rates.

↑ load factor above 88%

≈ +1% – 2% RASK
≈ +3% RASK
Not in our scope
≈ +1% – 2% RASK
← no-show below 6% no-show above 6% →

OVBO performance improves with higher %LF and no-show rates

main features

What you get that legacy tooling doesn’t.

Legacy solutions OVBO

Online platform

A UI where RM analysts customize their own strategies.

Legacy
OVBO

Predictions with machine learning

+30 features feed the model that scores each passenger’s no-show probability.

Legacy
OVBO

Denied boarding cost optimization

Optimizes margin, not just revenue. Set denied boarding cost per market.

Legacy
OVBO

Automatic alerts

Configurable alerts confirm the process ran and flag flights worth a second look.

Legacy
OVBO

OB rules management

Set lid rules to keep unwanted limits out — peak dates, last-day flights and more.

Legacy
OVBO

Snowball effect prevention

Every recommendation reads nearby flights’ load factor, so bumped passengers don’t cascade.

Legacy
OVBO

track record

Built by industry natives, already live in production.

Integrated with

Our partners

let’s talk

Want to see OVBO on your own flight data?

Send us a note and we’ll walk you through the model, the interface, and what the PSS integration looks like on your side.