Dynamic Pricing for a Regional Carrier

Dynamic pricing for a regional carrier: A fairsystems revenue optimization success story.

Overview

Our client was discounting early to fill seats, then watching late high-value demand buy those same seats at the price they had already given away.

Mohamed El Alali

Sr. Marketing Consultant / fairsystems
Pricing is the fastest lever a transport business has and the one most often set by habit. Fares get anchored to what the competition charged last season, discounting starts when the load factor looks uncomfortable, and nobody measures what the discount cost. fairsystems was engaged by a regional carrier to replace pricing instinct with a demand-based method.

The Client

Our client is a regional airline operating short-haul routes across a competitive market, with a mixed base of leisure travellers, business commuters and seasonal traffic. Fares were managed by a small revenue team using spreadsheets and long experience. That experience was real, but it did not scale across the route network, and it could not distinguish a route that was genuinely price-sensitive from one where the carrier had simply never tested a higher fare.
Roadmap
1
Assess
fairsystems analysed three years of booking curves route by route, separating genuine price sensitivity from the self-fulfilling effect of the carrier's own discounting behaviour. Our consultants interviewed the revenue team to capture the rules they applied intuitively, then tested those rules against outcomes to establish which reflected real market structure and which were habit. We segmented the network by demand pattern rather than by geography, which revealed that routes the client managed identically behaved very differently, and that several were being discounted into losses that seasonal averages had concealed.
2
Deliver
Our team built a demand-forecasting and fare-recommendation model calibrated per demand segment, with booking-curve thresholds that triggered fare movement on evidence rather than on nervousness about load factor. fairsystems deliberately kept the revenue team in control: the model recommends, an analyst approves, and every override is logged so the model learns from disagreement. We rebuilt the fare structure to protect late high-value demand, introduced controlled price testing on a subset of routes to gather genuine elasticity data, and integrated the recommendations into the reservation system the team already used.
3
Continue
fairsystems supports the client through a quarterly model review, incorporating the elasticity data that controlled testing continues to generate and recalibrating as competitive conditions shift. We train new revenue analysts on the method rather than the tool, so capability does not sit with individuals. Our consultants also work with the commercial team on route planning, applying the same demand segmentation to network decisions so that new routes are launched with a pricing model rather than acquiring one after the first difficult season.
Solution Details

Discounting early does not fill aircraft. It sells your best seats cheaply.

Discounting triggered by load factor anxiety rather than demand
Deliverable: fairsystems replaced intuition-driven discounting with booking-curve thresholds calibrated per demand segment, so fares move on evidence.
Late high-value demand buying seats already discounted
Deliverable: Our consultants restructured fare buckets to protect inventory for late booking segments, recovering yield that early discounting had been giving away.
Routes with different demand patterns managed identically
Deliverable: We segmented the network by demand behaviour rather than geography, exposing routes that had been discounted into losses under seasonal averages.
No real elasticity data, only the carrier's own past behaviour
Deliverable: fairsystems introduced controlled price testing on a route subset, generating genuine elasticity evidence instead of inferring it from self-fulfilling history.
Revenue expertise concentrated in a few individuals
Deliverable: Our team codified the revenue rules into an explicit method and trained analysts on the reasoning, so capability no longer depends on specific people.
Pricing tooling separate from the reservation system
Deliverable: We integrated fare recommendations into the reservation system already in daily use, removing the parallel spreadsheet process entirely.
No learning loop when analysts disagreed with the model
Deliverable: fairsystems built override logging into the workflow so every analyst disagreement becomes training data at the next recalibration.
New routes launched without a pricing model
Deliverable: Our consultants extended the demand segmentation into route planning, so new routes open with a calibrated fare structure rather than acquiring one by trial.

Result:

The carrier now prices from measured demand rather than from load factor nerves. Yield improved across the network, most sharply on the routes that had been quietly loss-making under seasonal averaging, and late high-value bookings are no longer competing for inventory already sold cheaply. The revenue team retained control throughout, which is why they use the model rather than work around it. New routes now launch with a fare structure derived from comparable demand segments instead of learning their pricing the expensive way.
0

Revenue per available seat increase

0

Yield improvement on price-sensitive routes

0

Reduction in unnecessary discounting

0

Load factor improvement

0

Fare decisions model-supported

Testimonials

How our services bring about success

We help organizations turn their operations into a competitive advantage and achieve breakthrough results.

All rights reserved by fairsystems. Copyright © 2013 – 2026
.