Data dictionary ancillary_band category · 3 bands
A new KPI: each guest's extras band (None, Low, High) on paid seats, bags, upgrades and lounge per trip, rather than on fare alone.
Built from fare_revenue_12meventamount
club_tier Holiday Return-Trip Families
Megan
Representative customer
312,000 guests · 16.1%
OTA Fare Hunters
Kieran
Representative customer
284,000 guests · 14.7%
Weekday Business Commuters
Graham
Representative customer
226,000 guests · 11.7%
Northern Lights Adventurers
Freya
Representative customer
198,000 guests · 10.2%
Visiting-Family Flyers
Amara
Representative customer
176,000 guests · 9.1%
Golds Nearly Polaris
Daniel
Representative customer
128,000 guests · 6.6%
Lapsing Former Golds
Patricia
Representative customer
104,000 guests · 5.4%
Premium Cabin Leisure Couples
Henri
Representative customer
96,000 guests · 5.0%
Points Redemption Planners
Owen
Representative customer
88,000 guests · 4.6%
Student and Gap-Year Backpackers
Lily
Representative customer
82,000 guests · 4.2%
Connecting Long-Haul Transit Guests
Ravi
Representative customer
74,000 guests · 3.8%
Bleisure Weekend Extenders
Sofia
Representative customer
66,000 guests · 3.4%
Ski Season Groups
Tom
Representative customer
58,000 guests · 3.0%
First-Time Flyers
Grace
Representative customer
41,000 guests · 2.1%
Industries · Travel & Hospitality
Hotel, airline and travel teams know loyalty programs are blunt: a handful of tiers describing millions of distinct journeys. Neuralift builds an AI model of your guest base from your own booking and stay data, finds the segments inside it, and keeps it current as travel changes.
Every segment comes with a field guide: who they are, how you'd recognise them, and what sets them apart from everyone else.
Use the arrows, swipe, or pick a segment on the map.
Holiday Return-Trip Families
Representative customer Not a real person, Neuralift does not process or store PII
How you'd recognise them
OTA Fare Hunters
Representative customer Not a real person, Neuralift does not process or store PII
How you'd recognise them
Weekday Business Commuters
Representative customer Not a real person, Neuralift does not process or store PII
How you'd recognise them
Northern Lights Adventurers
Representative customer Not a real person, Neuralift does not process or store PII
How you'd recognise them
Visiting-Family Flyers
Representative customer Not a real person, Neuralift does not process or store PII
How you'd recognise them
Golds Nearly Polaris
Representative customer Not a real person, Neuralift does not process or store PII
How you'd recognise them
Lapsing Former Golds
Representative customer Not a real person, Neuralift does not process or store PII
How you'd recognise them
Premium Cabin Leisure Couples
Representative customer Not a real person, Neuralift does not process or store PII
How you'd recognise them
Points Redemption Planners
Representative customer Not a real person, Neuralift does not process or store PII
How you'd recognise them
Student and Gap-Year Backpackers
Representative customer Not a real person, Neuralift does not process or store PII
How you'd recognise them
Connecting Long-Haul Transit Guests
Representative customer Not a real person, Neuralift does not process or store PII
How you'd recognise them
Bleisure Weekend Extenders
Representative customer Not a real person, Neuralift does not process or store PII
How you'd recognise them
Ski Season Groups
Representative customer Not a real person, Neuralift does not process or store PII
How you'd recognise them
First-Time Flyers
Representative customer Not a real person, Neuralift does not process or store PII
How you'd recognise them
This segment Everyone else
Four tiers can't describe two million travellers. Ask your model who's close to their next tier, who's quietly drifting and what keeps each of them flying with you, and it comes back with the plays, ranked. Here's an example plan for Aurora Air, a Neuralift demo brand.
KPIs, in priority order
club_tier This use case Move members up a tier, above all Gold to Polaris, where share of wallet follows. trips_band Add a trip a year for members flying two or three times. ancillary_band Grow paid extras per member trip, not just points earned. Channels
Status challenges move them
44%take a status challenge 7%other members
Still reading, no longer flying
52%open Aurora Club email 21%everyone else
They plan around points
142Kaverage points balance 18Keveryone else
They come back every season
2.6winters flown in five years 1.4other leisure travellers
Bring the use case that matters most. Neuralift ties what its deep learning discovers in your guest data to the KPIs behind it, and comes back with the guests to act on and the plays to reach them.
| Keyguest_id | Cleanedfare_revenue_12m | |||||
|---|---|---|---|---|---|---|
| g_38612 | 11,240 | Low | 6 | 0.94 | 22 | 0.93 |
| g_52417 | 284 | None | 19 | 0.00 | 1 | 0.33 |
| g_29805 | 1,960 | High | 124 | 0.71 | 2 | 0.40 |
| g_17904 | 1,840 | Low | 63 | 1.00 | 4 | 0.25 |
| g_31620 | 612 | Low | 12 | 0.40 | 3 | 0.60 |
| g_41207 | 2,310 | High | 92 | 0.64 | 1 | 0.14 |
Data dictionary ancillary_band category · 3 bands
A new KPI: each guest's extras band (None, Low, High) on paid seats, bags, upgrades and lounge per trip, rather than on fare alone.
Built from fare_revenue_12meventamount
Data dictionary booking_window_days integer · days
The median days between booking and departure, with duplicate and cancelled bookings removed.
Built from eventevent_tspax_ref
Data dictionary direct_share float · 0 to 1
The share of a guest's bookings made direct rather than through an OTA, with channel codes cleaned across systems.
Built from channeleventpax_ref
Data dictionary trips_12m integer · trips
Round trips flown in the last 12 months, built from flown segments rather than bookings, so cancellations don't count.
Built from eventroutepax_ref
Data dictionary weekday_skew float · 0 to 1
The share of a guest's flights Monday to Thursday, with times normalized to each departure airport's time zone.
Built from event_tsroute
Discovery starts by transforming the data you already have. Neuralift turns it into an AI-ready asset, engineers the features your use case needs, creates new KPIs and documents every one.
It fuels the deep learning, and it's yours to keep. Your data science teams can build on it, and so can your own agentic and AI applications.
Our team is always happy to talk through your customer data challenges and opportunities. We usually respond within a day, and we love helping teams see what their customer data can tell them.