Airline Distribution Costs

Full Report, 24 October 2017

Examination of direct versus indirect distribution costs for airlines

All data in this study has been sourced from publically available sources or ETTSA member companies on a

confidential basis. The confidential data has been anonymized and aggregated to ensure the source data cannot

be identified and the data cannot be ‘reverse’ engineered.

Commissioned by the European Travel Technology Services Association (ETTSA) and ECTAA (European Travel

Agents and Tour Operators Association)

Privileged Document. This document is subject to sensitive data protocols

Executive Summary: the distribution landscape - its drivers

The widely promoted narrative that direct distribution is ‘cheaper’ for an airline than indirect distribution is extensively reported

and used by airlines to support their position with respect to competition authorities and the general public.

However, no rigorous testing of this assertion is available in the public domain. This report aims to address this point. In

particular, this report examines the drivers of cost in the distribution process.

A number of key drivers are effecting change in distribution behaviour and costs:

Airlines that want to drive

direct sales via their own

websites have to replicate

the advertising and reach

of the online travel

agencies and ‘meta’

search companies.

This is proving expensive

and is not effective for

certain markets and the

European major LCCs,

EasyJet and Ryanair, are

now distributed via travel

agencies (through the

GDS) in order to access

the business market.

‘Ads’ are dominated by

the OTA who have

invested intensively in

non-branded ads creating

a step-change in

transparency and

customer service for

passengers but at a cost

per ad that will have to

matched by the airlines to

attract direct sales.

The GDS and OTA have

invested massively in a

range of consumer-facing

and back office

technology that greatly

enhances the consumer

experience and provides

service for business

passengers.

The airlines are finding it

challenging to offer the

same service; Lufthansa

has hired 17 business

partner companies to

help develop its direct

product

Travel Management

Companies and Online

Travel Agencies provide

substantial back office

support for businesses

and customer service for

the wider market.

They offer 24 hour multi-

language service centres

globally. This cost will fall

upon the airlines if not

provided by these

companies.

Major costs such as

credit cards and other

processing costs are paid

by the OTA and agents.

These costs will fall on

the airlines if they move

traffic directly to their

websites.

Customer acquisition,

competitive on-line

landscape:

Websearch:

Technology:

Customer service:

Payment, finance costs

and administration:

Page 2

This report has looked at the distribution landscape and drivers and has modelled objectively and accurately (subject to data

limitations) the impact of moving sales from indirect to direct channels. The cost differential is presently much smaller than

airlines contend. For network carriers the cost of direct distribution is €12.56 versus €14.21 for indirect.

Executive Summary: Detailed modelling shows limited impact of

increasing direct sales

Network airlines full distribution cost per booking €

The main model assumption used is that airline websales

increase to 60% from the base of 40% and direct sales

increase from 47% to 67% including the use of airline call

centres, ATO and CTO.

The impacts are complex with a number of key dynamics:

•  Substantial increase in average ads cost to pull consumers

from their current channels

•  Some costs are reduced due to lower GDS booking fees

and less agents commission.

•  Increased costs of customer service that agents provide for

customers and the credit card costs, some fraud costs and

the cost of managing customer changes would fall on the

airline.

•  The final estimate is a reduction of total cost per booking of

€0.11 but with substantial risks of losing market share,

especially in the business market, and a major

organisational challenge.

The ultimate loser may be the consumer due to less price transparency and potentially worse customer service.

Page 3

This report has been prepared in accordance with the scope of Infrata Limited’s appointment with its

client and is subject to the terms of that appointment. Infrata accepts no liability for any use of this

document other than by its client and only for the purposes for which it was prepared and provided.

The conclusions and recommendations contained in this Report are based upon information sourced

in the public domain and provided by others and upon the assumption that all relevant information has

been provided by those parties from whom it has been requested and that such information is

accurate. Information obtained by Infrata has not been independently verified by Infrata, unless

otherwise stated in the Report.

Certain statements made in the Report that are not historical facts may constitute estimates,

projections or other forward-looking statements and even though they are based on reasonable

assumptions as of the date of the Report, such forward-looking statements by their nature involve risks

and uncertainties that could cause actual results to differ materially from the results predicted. Infrata

specifically does not guarantee or warrant any estimate or projections contained in this Report.

Infrata disclaims any undertaking or obligation to advise any person of any change in any matter

affecting the Report, which may come or be brought to Infrata’s attention after the date of the Report.

Disclaimer

Page 4

Page 5

Introduction: description of the distribution modelling process

The analysis in this report is based on detailed knowledge and data of the travel industry’s business model and the players that

operate within.

However, every player in the industry operates the model slightly differently and has differing level of access to privileged information

which impacts their costs base. Whilst the authors had access to specific data to populate the cost model, other data required

‘considered’ estimation.

As a consequence, any numbers provided in this report will be the average number derived from within a range for a particular cost.

The methodology to produce this report has included the following stages:

Desk research:

There is a wide body of

literature covering airline

sales, marketing and

distribution. Infrata has

undertaken a

comprehensive review of

literature from industry

bodies, trade press and

academia.

A full list of the sources

and literature search is

provided in the appendix.

Primary research:

Infrata has sourced

information and

characterised the issues

through a series of

confidential interviews with

industry representatives

including airlines,

distribution companies,

travel agents and other

industry bodies.

A list of participants is

provided in the appendices

although some

respondents wish to

remain confidential.

Data platform:

Infrata has developed a

platform of the key data

that impact upon

distribution cost.

Some data was relatively

straightforward and easy to

source, such as the cost of

processing bookings or

credit card costs.

Other data, such as airline

‘ads’ cost was more

opaque and assumptions

have been made where

necessary. The data

platform allowed the

derivation of sensitivities of

certain costs to distribution

channels used.

Data modelling:

Infrata developed a model

that took into account all

the costs of distribution,

some of which are not

typically included in

distribution cost model.

The modelled applied ‘unit

rates’ of cost and channel

sensitivities to hypothesise

aggregate airline

distribution costs according

to their chosen distribution

mix.

Different costs applied to

the three main types of

carriers analysed.

Reporting:

The results are presented

as a range of costs that

can be compared to see

the magnitude of difference

between the varying levels

of direct distribution.

Content

•  Detailed Cost Summary

Supporting Facts and Analysis

•  Airline Distribution Market Dynamics

•  Airline Distribution Customers and Revenues

•  Airline Distribution Channels

•  Airline Distribution Model Output

•  Appendices and Model Assumptions

Page 6

Airline Distribution

Detailed Cost summary

Page 7

Detailed Cost Summary

The true cost of travel distribution is publically obscured by the airline industry as

cost elements are not fairly compared

The majority of airlines believe the cost of “direct’ distribution” (e.g. own website

and sales office) is “significantly” lower than selling via ‘third party intermediaries’

such as travel agents and online travel agents using GDS. This assertion is backed

up by various studies.

Our analysis in the chart (left) illustrates the gap between the airlines’ usual view of

distribution cost and our view of the full cost. The ‘classic’ airline view of distribution usually

contains the following:

•  ‘Distribution’ - comprising mainly agents’ commission, GDS booking fees, reservation

hosting fees;

•  Payment and finance - credit card costs, fraud costs, BSP costs, other IT costs;

•  Ancillary services - reservation and IT related costs supporting the sale of ancillary

services.

This ‘classic’ airline view typically excludes:

•  Online cost of customer acquisition including web-search (Google ads);

•  Offline marketing costs such as newspaper and TV advertising;

•  Cost of technological development and product enhancement;

•  Cost of customer service, sales offices, agents’ back office and merchant costs.

This study shows that these additional ‘non-accounted for’ costs substantially

narrow and for some types of airlines eliminate entirely the gap

The following analyses are based upon a model of ‘typical’ airlines with average costs.

The results will vary considerably by individual airline.

Source: Infrata

Page 8

Total Distribution Cost per booking (€ )

Classic

Airline

view

True cost

€13.43

14 13 12 11 10 9 8 7 6 5 4 3 2 1 - 15

Distribution

Customer

Acquisition

Payments

Ancillaries

€10.85

€2.58

!"#$%&'()*&+*,"(

-./&0"(1%2"(3/&'"#4(567(8*&"9#:;((

* Direct Sales comprise:

Web sales (40%) and ATO/CTO/CC (7%)

Detailed Cost Summary

Further, three sets of inter-related ‘drivers’ need to be modelled as they materially

‘impact’ outcomes for the airline

In developing a comprehensive cost /impact model, this study needed to take into account three interrelated cost drivers:

(1) Market dynamics, (2) Customers and revenue and (3) Cost per channel

(1)  Market dynamics:

Costs are impacted by the shifting structure and ongoing developments

in the airline industry:

•  Booking Direct:

The airline industry is shifting online and there is a trend towards

moving traffic to booking direct on airlines’ websites. The online

winners are increasingly those who can drive traffic to their

website.

•  Online advertising:

Google and other websearch ads are now the main way to drive

traffic to websites. ‘Ads’ are the first touch point of the consumer.

The most effective ads are paid, unbranded but these are

expensive with growing costs. They are ‘owned’ mainly by the

major OTA (Expedia, Travelocity) not the airlines.

•  Technology:

Technological upgrades/innovations to enhance the consumer

product and keep in touch with the consumer are being developed

by companies including Amadeus, Sabre, Travelport and

Expedia. Development is now being concentrated on mobile

technology. These are expensive and long term investment

programmes.

•  Airline network development: airlines seeking to grow in non-base

regions have to contend with the market power of base airlines.

The most cost-effective way to reach the market is to use all

distribution channels.

(2) Customers and revenue:

The effectiveness of the channels to market is driven partly by the types of

passenger that particular channels work for and the average revenue per

passenger of customers by channel. There continues to be a paucity of

information that would allow a more accurate modelling of the market

taking into accounts factors such precise revenues by channel.

Passenger channel shifting is a key factor in the model with, we believe,

major cost and revenue impacts:

•  revenue will change as higher revenue customers from TMC are

resistant to moving to airlines' websites

•  airlines selling direct may claw back some discounts previously shared

with travel agents

(3) Costs by channel:

There has been a close examination of the costs of distribution through all

the different channels.

The model allows the shifting of passengers from one channel to another.

Early important observations are that:

•  internal airline distribution systems costs appear relatively fixed by

channel, notably

•  commissions, search engine marketing (SEM), GDS are variable by

channel

The report presents a full cost evaluation of using each distribution channel.

Model: the data has been incorporated into an initial channel cost model showing the expected relativity of costs per booking by channel. The

model has been developed for three different types of airline: (1) Network - large home market, (2) Regional and (3) Network – small home market.

Page 9

Detailed Cost Summary

The full cost impact of moving bookings from indirect to direct is negligible for

network carriers (large home market) and negative for regionals and network

carriers (small home market)

Cost per passenger booked (€ per booking) – per carrier type, per level of direct booking

Our study modelled a specific ‘what-if’ scenario: Shifting the percentage of bookings from the airline’s current ‘indirect’ channel (’47%’)

to higher percentage of ‘direct’ channel (’67% direct’)’ by carrier type. The percentage shift represents estimated ‘real world’ examples.

For a network carrier with a large home market, the net effect of the channel shift is a reduction in ‘total’ distribution cost of up to €0.11.

This comprises a €1.64 reduction in ‘Distribution’ but compensatory increases in ‘Customer acquisition’ of €0.98 and payment, admin

and finance’ of €0.55.

Page 10

"Carrier type

Network

(large home market)

Regional

Network

(Small home market)

Channel /Scenario

47%

direct

67%

direct

Variance

47%

direct

67%

direct

Variance

47%

direct

67%

direct

Variance

Customer acquisition €

2.58 3.56 0.98 2.64 4.0 1.52 3.77 6.21 2.44

Channel cost € 6.75 5.11 (1.64) 7.3 5.83 (1.47) 7.75 5.73 (2.02)

Payment, finance

and admin €

4.04 4.59 0.55 3.46 4.09 0.63 4.04 4.59 0.55

Ancillary costs € 0.06 0.06 0.00 0.06 0.06 0 0.10 0.10 (0.00)

Total € 13.43 13.32 (0.11) 13.46 13.98 0.52 15.66 16.73 1.07

* 47% & 67% Direct Sales comprise:

Web sales (40% /60%) and ATO/CTO/CC (7%)

Source: Infrata, * IATA

Page 11

Detailed Cost Summary

Home market versus non-home is a vital driver of distribution cost

In modelling network carrier distribution dynamics, two important distinctions exist: (1) the concept of home market (versus a non-home market)

and (2) whether the home market is large or small

Two consequences of the above are:

1.  The difference between direct and indirect distribution costs will be further reduced (or

completely eliminated) in a network carrier’s home market (on a fully allocated cost basis);

and

2.  Network carriers with a larger home market will have lower overall indirect distribution costs

than carriers with smaller home markets

•  A network carrier’s home market is where the carrier has its

primary client base and where it normally originates from

•  Large network carriers are often legacy national flag carriers who

used to enjoy a monopoly or quasi-monopoly in their legacy

national market (what we now call their ‘home market’)

•  These flag carriers were frequently also state-owned or state-

controlled

•  Large network carriers often still enjoy a disproportionally strong

brand position in their home market (vs. non-home market). This

position relates to reputation, recognition, national identity, cultural

heritage and commercial presence

Characteristics:

Implications:

•  Carriers with a strong brand position have a much lower cost of

customer acquisition when distributing direct. This is because

customers tend to ‘default’ their travel searches to carriers they

know, recognise and relate to

•  This means that direct distribution for the carrier in its home market

will be lower cost to develop and maintain than in its non-home

market

•  Further the cost of indirect distribution in the carrier’s home market

will also be lower (than in the non-home market).

This is because the value (and hence bargaining power) of

intermediaries will be relatively lower in the airline’s home market

compared to non-home markets

Detailed Cost Summary

The blended difference in total booking cost per booking for Network Carriers

(large home market) is ‘immaterial’ and remains largely unchanged as airlines move

to greater direct distribution

Source: Infrata

Page 12

Blended

€13.43

€13.32

Channel Distribution Cost

per booking(€ )

Direct

€12.56

€12.77

.43

Distribution

Customer

Acquisition

Payments

Ancillaries

Channel Distribution Cost

per booking (€ )

47% Direct

€14.21

€14.42

Distribution

Customer

Acquisition

Payments

Ancillaries

The cost of the direct and indirect distribution channels between the low-direct and high-direct scenarios

Indirect

Blended

Direct Indirect

67% Direct

!"#$$%&'()*+*',' 47%' 67%'

!"#$$%&'-#&#$(%'

Indirect' Direct' Blend' Indirect' Direct' Blend'

Distribution

10.48 2.56 6.75 10.65 2.38 5.11

Payments etc

2.84 5.39 4.04 2.90 5.42 4.59

Ancillaries

0.06 0.06 0.06 0.06 0.06 0.06

Cust Acquisition

0.83 4.55 2.58 0.81 4.91 3.56

Total

€14.21 €12.56 €13.43 €14.42 €12.77 €13.32

!"#$%&'()*&+*,"(

-./&0"(1%2"(3/&'"#;((

* 47% & 67% Direct Sales comprise:

Web sales (40% /60%) and ATO/CTO/CC (7%)

Detailed Cost Summary

For Network Carriers (small home market) the impact of moving bookings from

indirect to direct distribution channels increases costs

Page 13

Blended Distribution Cost

per Segment (€)

47%

Direct

€15.66

€16.73

Distribution

Customer

Acquisition

Payments

Ancillaries

Direct Distribution Cost

per Segment (€)

47%

Direct

67%

Direct

€14.49

Distribution

Customer

Acquisition

Payments

Ancillaries

Indirect channel Distribution Cost

per Segment (€)

47%

Direct

67%

Direct

€16.78

€16.92

Distribution

Customer

Acquisition

Payments

Ancillaries

!"#$$%&'()*+*',' Indirect ' Direct' Blended'

!"#$%&'-#&#$(%'

47%'' 67%'' ’47%'' ’67%'' ’47%'' ’67%''

Distribution

12.35 12.53 2.56 2.38 7.75 5.73

Payments etc

2.84 2.90 5.39 5.42 4.04 4.59

Ancillaries

0.10 0.10 0.10 0.10 0.10 0.10

Cust Acquisition

1.49 1.29 6.34 8.63 3.77 6.21

Total

€16.78 €16.92 €14.49 €16.53 €15.66 €16.73

€16.53

67%

Direct

Source: Infrata

* 47% & 67% Direct Sales comprise:

Web sales (40% /60%) and ATO/CTO/CC (7%)

!"#$%&'()*&+*,"(

-<2/++(1%2"(3/&'"#;((

Indirect

Detailed Cost Summary

Little difference exists between ‘distribution’ costs of the Indirect channel and Direct

Channel when fully allocated costs are compared

Page 14

Direct

Cost per Segment

booked by channel

€2.56

€10.48

14 13 12 11 10 9 8 7 6 5 4 3 2 1 - 15

Distribution

‘Distribution’ only cost comparison

(Blended: home and non-home market)

!"#$%&'()*+*',' ./&0"(1%2"(3/&'"#4567(=*&"9#'

Indirect' Direct' Blend'

Distribution

10.48 2.56 6.75

Payments etc

2.84 5.39 4.04

Ancillaries

0.06 0.06 0.06

Cust Acquisition

0.83 4.55 2.58

Total

€14.21 €12.56 €13.43

!"#$%&'()*&+*,"(-./&0"(1%2"(3/&'"#4567(=*&"9#;((

Direct

€12.56

14 13 12 11 10 9 8 7 6 5 4 3 2 1 - 15

Distribution

Customer

Acquisition

Payments

Ancillaries

€14.21

Indirect

‘Properly’ allocated cost comparison

Blended

€6.75

€13.43

Source: Infrata

* Direct Sales comprise:

Web sales (40%) and ATO/CTO/CC (7%)

Detailed Cost Summary

Online advertising (e.g. Google) costs grow exponentially for a Network

Carrier (small home market) as it moves to greater direct distribution and

for network carriers with large home markets, it increases by nearly 70%.

€ Cost per booking

47%

Direct

67%

Direct

90%

Direct

47% 67% 90%

Large home market

(>60% originating traffic)

0.61 1.01 1.8

Small Home Market

(<30% originating traffic)

0.82 1.71 3.0

For carriers operating in large home markets (>60% bookings orginating from within its home terrritory) ads cost increase from an estimated at

€0.61 per booking at ‘low‘ level of direct sales (47%) to €1.01 where there is a greater emphasis on direct sales (67%) - as incremental direct

sales are picked up in the home terrtory where there is market and brand pull.

However for Carriers with small home markets (where <30% of the airline’s sales are inside its home market ) the cost is projected to increase

from €0.82 to €3.00 (47% to 67% direct sales respecitively)

Large home market

Small home market

Source: Estimates based upon ads industry sources and airline specific data

* 47% & 67% Direct Sales comprise:

Web sales (40% /60%) and ATO/CTO/CC (7%)

90% levels included to

illustrate the impact of

dynamics

Page 16

Airline Distribution

Market Dynamics

Page 16

Market dynamics

Overview of shifting market dynamics

The development of distribution approaches by airlines and

distribution costs is a function of the dynamics of the overall

aviation market.

The research undertaken in the course of this study has

revealed the major external factors that impact distribution

costs for individual airlines and for the industry overall.

These major factors have been analysed in the following pages

and their potential impact has been quantified where possible.

However, some of these issues, whilst informing the study, are

not easily quantified and would require more research and

dialogue, especially with airlines.

Various industry studies suggest that traditional distribution

costs have fallen in recent years from an average of 16% to

well under 10% but these exclude the growing cost of

‘customer acquisition’.

This has been facilitated by a major sub-industry of software

providers providing an array of solutions in optimising ways

of:

1.  providing consumer choice

2.  presenting airline product to the market

3.  efficient allocation of seat capacity.

The major areas of cost increase pertain less to the logistics

of taking reservations, decrementing inventory and issuing

tickets and more to communicating with potential customers.

The modelling process used a detailed knowledge of these systems and

processes to model present and potential future airline distribution regimes

Page 17

Category Dynamic

Online Airlines aim to increase bookings via their own websites including ‘direct connect’ (led by LCC and

American Airlines in US).

Search Engine Ads Airlines needs to ‘invest’ heavily in customer acquisition via Google ads (others also exist) to

ensure traffic to own-brand.com site.

Ancillary revenues Alternative business models and declining average ticket prices drive revenue enhancement

activity to a wide range of ‘ancillary’ revenue activity.

Technology (1) Airlines aim to optimize their bookings on their most profitable channel mix - employing

sophisticated revenue and channel management tools.

Technology (2) Airlines are facing new distribution players (e.g. speedmedia) which employ new technology (e.g.

mobiles) and business models (e.g. Google’s Trips).

Market Dynamics - Overview

Airline distribution costs are impacted by online sales, customer

acquisition and technological development

Key market dynamics impact airline distribution costs. These dynamics either simplify the ‘chain’ thus reducing certain cost (e.g.

online booking) whilst other dynamics increase ‘complexity’ (e.g. increased advertising costs and distribution technology).

Page 18

Market Dynamics - Online Sales:

Online sales are now over 50% of tickets and growing

Online Travel Sales 2015 and % of Market

(including airline web sales and OLTA)

Online sales overall have continued to grow in the last three years as a

percentage of total travel sale (see table). Global online travel sales

exceeded $523bn in 2016, over half of the global total.

Airline direct sales are growing fastest in emerging markets and are

growing less quickly in ‘mature’ US and European markets.

Issues around direct sales produced by the research are:

•  Airlines consider websales as an efficient way to bypass Travel Agents

•  Websales are most effective at selling on ancillary products although

IATA’s NDC and other processes are enabling TMC/TA to do this

•  Business passengers require the ‘high value’ services delivered by

TMCs; this channel has proved resilient due to the requirement for

value added services like back office, customer service and payment

on account

•  Airlines are being outspent and are losing visibility online due to heavy

marketing push by OTA and meta

•  Airlines with heavily concentrated markets can rely upon native search

but expansion or more widely distributed markets will require very

substantial advertising cost.

•  Lufthansa is pushing its direct channel by ‘penalising’ the indirect

channel with a €16 surcharge on GDS bookings

Airline web sales are the largest individual channel used for distribution and provides the greatest lever of ‘disintermediation’

Two differing levels of Websales by airline group are modelled (base and high direct) in the model

Key trends

Source: Tnooz

Page 19

60%

50%

40%

30%

20%

10%

52%

54% 54%

48%

49%

50%

Online Travel Sales as % of Total

2014 2015 2016

Europe

Market Dynamics - Search Engine ads:

Airlines need to compete with ‘heavy-spending’ OTAs and Metas to secure

online sales (1 of 2)

•  ‘Ads’ are a key tool in online sales

•  OTA and airlines agree keywords with the search engines

and pay to appear when these words are typed by the

passenger into their browser

•  Passengers routinely use common keywords to begin their

search for a flight and the best fare

•  Paid ads should ensure that the OTA or airline appears on

the first page

•  Not appearing on the first page can badly affect airlines’

sales

Top 15 Flight Company Paid Search UK – 2014

Source: Morningstar, searchmetrics

The analysis of keywords provides three main lessons

•  the OTA / Meta are outspending the airlines heavily

•  Foreign airlines are relatively ‘invisible’ in non-base

markets

•  Using OTA/Meta allows travel providers to reduce their

keyword spending

Key trends

•  ‘Project Paid Visibility’ on Google shows the visibility of an entered internet

address in the advertisements area on Google compared with an applied

keyword set.

The main advertisers in UK and Germany are OTA / Meta

•  In UK 12 OTA / Meta compete with 3 airlines

•  In UK 8 companies dominate paid search, 5 dominate in Germany

•  EasyJet appears low on paid search – success of long term branding will

have pushed native

•  In Germany 12 OTA / Meta v 2 airlines, 1 tour operator appear on equivalent

list

•  Only LH appears as a foreign airline in either list

Page 20

/.'

00'

1/2'

2//'

20/'

034'

15677'

1581.'

2506/'

05/17'

052.7'

05.02'

85748'

45../'

7' 15777' 05777' .5777' 85777' /75777' /15777'

9:;<#=>;*):?()?@A'

B)C()*+")&;=#D*?()E'

F@*G"%H;I"+?()?@A'

B@J"#$*#?()E'

KL<%=;#?()?@A'

K#*DM%+?()E'

NAD*(#$$%:?$%+'

B#*+E;$@+%?()E'

9:#>%&*@<%:E#:A%+?()E'

!"%#<H;I"+*?()?@A'

O#D#A?()?@A'

K-))A%:*?()E'

P:;Q*"#;:C#D*?()E'

R<)=)?()?@A'

9:#>%&S))?()E'

T:)M%(+'T#;='U;*;-;&;+D')$'V))I&%'

Market Dynamics - Search Engine ads:

Airlines need to compete with ‘heavy-spending’ OTAs and Metas to secure

online sales (2 of 2)

Ads are complex for airlines to manage effectively.

•  Cost of paid ads are growing and conversion costs getting very

close to third party fees the airlines are trying to escape from.

•  Some airlines do not appear in the top results for their brand

searches including searches in home markets.

•  Potentially space for competitors and OTA’s to generate the

majority of brand impressions and traffic.

•  Brand traffic is the highest converting due to loyal customers -

generally representing majority of revenue generated by the

channel.

•  Airline websites are heavily competing with OTAs and adding

other traditional travel agency products like hotels, rent a car,

etc.

Keyword Spend

($m)

Impress-

ions (m)

Avg.

Clickthrough

Rate

Avg.

Cost/

Click

($)

# of

Advert-

isers

Cheapflights 9.9 191 3.1% 1.70 74

Flights 2.7 59 3.3% 1.41 52

Expedia 2.6 53 13.5% 0.36 4

Cheap Tickets 1.8 69 2.8% 0.92 67

Priceline 1.7 22 18.2% 0.43 8

Orbitz 1.6 21 19.6% 0.40 4

Travelocity 1.6 26 15.8% 0.40 13

Cheap Airline

Tickets

1.5 22 3.8% 1.79 72

Airline Tickets 1.5 44 2.5% 1.37 74

US Airways 1.3 18 14.6% 0.49 12

Southwest 1.2 19 14.3% 0.45 3

US Ad spend and click through 2014 – Top

Travel Keywords

Source: Everymundo

Ads are one of the major cost drivers of the airline direct distribution cost

model

The average cost per click in Google AdWords is between $1 and $2 on the

search network. The average CPC on the Display Network is under $1. The

most expensive keywords in AdWords cost $50 or more per click. If a consumer

searches 22 times for the best flight there is the potential to incur (in a most

extreme case) up to around $50 in ads cost for the airlines or the online travel

agent and ultimately the air passenger.

Key trends

Page 21

Market Dynamics - Airline Ancillary Revenue:

Distribution companies are making major investments to sell ancillary

services for the airlines

30%

15%

25%

Baggage fees

Onboard Retail (food,

duty free)

Sale of FFP miles

Travel Retail (hotel, hire

cars, insurance)

Other 'a la carte'

services

Airline Ancillary Revenues 2014 by Main Category

Total Non-US Ancillary Revenue 2014 Estimated $36.5 bn

Ancillary revenues for the top 10 performing airlines globally rose to

almost $26 billion in 2015 compared to $8.4 billion generated in

2008, a CAGR of 17.5%. Total global was $59.2bn.

US low cost airline Spirit generated the highest amount of ancillary

spend per passenger at $51.80.

Ancillary revenue is generated by activities and services that yield

revenue for airlines beyond the simple transportation of customers

from A to B.

Revenue from optional services including onboard sales of food and

beverages, checked baggage, premium seat assignments, and early

boarding benefits, was $36.7 billion of the projected global 2015

total.

The remaining share, at $22.5 billion, comes from non-fee activity

such as the sale of frequent flier miles to program partners (a major

revenue in the US) , and commissions earned on the sale of services

to travelers, such as hotel accommodation and car rentals.

The IATA NDC project and other initiatives aim to facilitate the sale of

‘ancillaries product’ for GDS and OTA.

Ancillary revenues will be an important factor in future

distribution channel development

Key trends

Airlines websites are currently the best channel for selling ancillary products. Airlines can tailor the display to promote their

products – airlines claim that this is not possible at the moment on the OTA and on agents’ GDS screens. However, recent

innovations have led to increased sales of ancillaries by TA.

Page 22

Source: IdeaWorks /Cartrawler

Market Dynamics - Technology:

Airlines need to match similar levels of investment made by GDS to

continually ensure satisfactory levels of functionality and connectivity

Major investment by GDS, OTA and other major IT providers spread

over areas of mobile, ‘big data’, Cloud, API and compliance. Ongoing

service offering includes:

•  Data Centre (Amadeus used as example):

•  37 Petabytes+ of storage and over 16,500 infrastructure devices.

•  peak processes 39,000+ end user transactions per second and over

47 billion SQL executions daily.

•  5,500+ IT changes and over 540 application software loads daily

•  Airlines products:

•  GDS connectivity

•  inventory hosts

•  revenue management

•  e-commerce

•  ticketless access, merchandising solutions and self-booking tools.

•  cloud availability, NDC compliant XML connectivity, revenue

optimisation and financial suites

•  Travel agencies, meta-search engines, travel management

companies and corporations products:

•  cloud-based new generation selling platform

•  search engines

•  front-office customisation and conversion tools

•  merchandising solutions

•  ancillary services

•  fare families

•  Other travel providers

•  data, connectivity and solutions for hospitality, rail etc.

“Travel technology providers can

spread development and maintenance

costs over dozens or hundreds of

travel providers, thus increasing

capabilities and reducing costs”

Leading industry analyst

Page 23

Market Dynamics - Technology: Example Lufthansa

For true cost comparison, the costs associated with the technology

development of ‘Direct’ needs to be included

Platform for booking business trips, online

booking engine, travel management system.

Content Aggregator - GDS Content and Direct

connections with airlines and other travel

suppliers in the business travel eco-system

Midoffice for the international travel industry.

Travel production and distribution platform;

solutions for dynamic travel production

Back office & reporting systems for the travel

sales segment, with interfaces to all

common front and middle office providers,

tourism service providers and credit card

companies.

Technologies provide solutions for areas including:

•  Internet booking engine (IBE)

•  Content aggregator

•  Mid-office

•  Back office

•  XML, API (programming languages and interfaces)

•  Payment services

The systems integration of multiple technology providers provides

substantial challenges to airlines:

•  Integration with existing platforms

•  Quality control

•  Guaranteed supply, ongoing product investment and

innovation

•  Internal team familiarity with technology provider products

Multi GDS B2C /B2B booking engine, XML

API, Low cost carriers, Payment services, PCI

DSS proxy solutions and fraud prevention

solutions.

Key trends

Lufthansa ‘Direct Connect’ project lists 17 technology partners from 5 countries – providing connectivity and functionality across numerous

markets, affiliates, channels etc. is one that requires expertise and careful management. Carriers seeking to replicate reach and functionality

of direct connect can expect similar required effort and investment. IATA NDC ancillary product implementation may ultimately require a

similar scale of effort. These costs may be absorbed by LH or passed on to the consumer via the Travel Agent.

Sample selection of LH Direct technology partners

Source: Lufthansa

Page 24

Airline Distribution:

Channels

Page 25

Channels:

Distribution, Sales and Marketing goals vary by airline type

BA, Air France, American

Flybe, Aer Lingus, Alaska

Finnair, Icelandair

Widely dispersed market, high levels of business traffic, major trunk routes, targeting all

market segments. Requires ‘comprehensive’ distribution and sales and marketing effort

utilising all available channels

‘Local’ highly targeted market and important relationship with partner carriers. Distribution

goals of low cost and ease of access for major corporate accounts and business market.

Network (Large home)

Highly distributed market across global markets. Distribution and sales and marketing goals

or maximising penetration across numerous markets. Need to build/maintain consumer

awareness but control cost.

Regional

Network (Small home )

These customer groups formed the basis of the distribution cost model

The modelling process has segmented the airline market into three main types; clearly this is subjective and different

categorisations are possible.

Characteristics

Page 26

Channels:

Overview

The Infrata business model has been designed to

hypothesise and assess the impacts of changing distribution

regimes by different customer types.

The complex airline market has been segmented into three

airline types

1.  Network

a.  Large home market

b.  Small home market

2.  Regional airlines

The major market characteristics of each is described herein

with the potential distribution impacts.

The airline market is changing and one of the key dynamics

is the growing share of LCC. These have 39% of the intra-

European market in 2016 although their growth has slowed.

The structure of the network airlines is changing in response

with the development of LCCs by BA/IAG (Vueling), LH

(Eurowings) and a regional airline Air France’s (Hop).

Key dynamics of the customer base that drive the model

are the fares by channel and the mix of fares by carrier

type.

Further dynamics are the variances in fares by area of

sale and by point-to-point and connecting passengers.

Fares actually collected are a closely guarded airline

business secret and there have been considerable

restrictions on data.

Page 27

Channels:

‘Channel Complexity’ is driven by the market dynamics and interaction

between 6 main channels to consumer bookings

Airlines have 6 main channels to market including:

•

Web

•

OTA

•

TMC

•

Call centre / ATO /CTO

•

Affiliate

•

GDS are also discussed in this report as a channel

although it exists as a facilitator serving the business to

consumer contacts of the TA, OTA and TMC.

Airline internal costs are also examined as a major cost and

one that changes according to channel mix.

Customer acquisition costs impact on all channels and are

sensitive to and driven by channel mix.

Ancillary revenues and the cost of distributing ancillary

products is an area of growing concern for the airlines.

“…the increasing sophistication of

passenger services systems, the complexity

of each transaction is growing significantly.”

Amadeus Global Business Report 2013

Page 28

Consumer have several direct and indirect (primary and secondary)

channels by which they can access airline bookings

Page 29

Customer

Meta

Global

Distribution

Systems

Revenue

management

Space

Control

Pricing

Airline

Front Office Back Office

Public

SMEs

TMC

Customer

Search Service - Packaging Distribution - Comparison Inventory Owners

Off line

Revenue

Accounting

Invent

ory

IBE

Inventory

control

CRO

Leisure

OTA

Content

aggregators

Corporate

Direct connect

Source: Infrata

Channels - Web sales:

Airlines employ a range of channel to achieve sales. This mix varies by

Airline type

Airline Type Web CC /ATO /CTO TA OTA TMC

Network- large 40 7 10 18 25

Regional 35 10 10 25 20

Network-Small 40 7 10 18 25

Channels To Market (%) Used by Airline Groups

The channel mix of these groups are a major

driver of cost in the distribution cost model.

Network carriers targeting business market and

more distributed market typically utilise more

channels.

Agents / TMC are important channels into

business market

Opportunities for direct connect increase with

higher ‘local’ sales as the airline is likely to have

greater brand presence and marketing

effectiveness.

Source: Infrata based on airline discussions

The channel mix shown will form the ‘base’ scenario for the distribution cost model

Channel analysis

Page 30

Channels - Web sales:

Shoppers have a greater propensity to ‘comparison shop’ using OTAs,

than on Airline websites – Airlines website convert more bookings

Comparison Shopping

% Shoppers Using OTA and Airline Website 2015

Purchase

% Bookers Using OTA and Airline Website 2015

Source: PhoCusWright 2015

Websales are averaging around 50% in the US and the major European

markets:

•  Network airline websales estimated to account for 35%.

•  80% of LCC sales.

•  42% Regional

•  Network (high connecting) and Network (high non-base) are

expected to have lower websales at 30%

Airlines concentrating on websales as a sales channel due to following

‘pros’:

•  Perception that airline ‘owns’ the customer

•  High conversion % - see charts

•  Avoids GDS costs

•  Avoids agents commission, incentives, overrides, special fares

•  Superior channel for promoting ancillary products

•  Improved cash flow

•  Avoidance of direct price comparison

‘Cons’ often understated by airlines are:

•  High cost of customer acquisition – ads to drive traffic to the

website

•  Significant investments in technology by competitor OTA

•  Significant investments by OTA in ads and other customer

acquisition

Presently the trend is for shoppers to browse OTA but more ‘bookers’ use

airline website, see charts.

The conclusion for modelling purposes is that web sales will increase

as a % but will have to be supported by significant advertising.

Airlines’ costs are reduced by OTA providing service

Page 31

OTA Airline

USA UK France Germany

OTA Airline OTA Airline OTA Airline

OTA Airline

USA UK France Germany

OTA Airline OTA Airline OTA Airline

Channels - Web sales:

The online Travel Agents / Meta business model is based on a low-

margin, high-volume sales

OTAs offer airline products often in conjunction with hotel and other

components of the travel.

•  Predominantly leisure-oriented and their business model is based

on the commissions earned by OTA at the moment of the booking

and revenue coming from advertising.

•  Some OTA have business affiliates

•  Traffic comes to some OTA mainly from paid search (53%)

•  OTA tolerate a high ‘bounce rate’: their costs are spread over a

large number of visits

•  OTA content is provided by both GDS and through airline API and

‘screen scrapping’.

•  OTAs aggregate the results into a single list or display them

according to their source.

•  Meta searches generate revenues through advertising and

referring clients.

•  Customers then can purchase the travel product via OTAs or by

accessing the supplier website.

OTA sales are characterised by large volume of traffic driven though paid search

OTA Size and Performance 2015

Company Total

Visits

(m)

Bounce

Rate

% Traffic

From

Paid

From

Social

Networks

From

Display

Ads

Booking.com 218 30% 34% 53.3% 1.36% 2.45%

Expedia 70 37% 33% 31.6% 0.95% 1.85%

Source: similarweb

Notes

•  Bounce Rate: where the user left site from the entrance page without interacting

with the page

•  Percent of traffic from search: the amount of visitors from organic search

•  Percent of traffic from paid search: the amount of visitors who entered the page

via keyword bought to increase the visibility. I.e. ”Cheap Flight"

•  From social networks: the amount of visitors who entered the page via ad/

reference on Facebook, Instagram etc.

•  From Display Ads: paid ads on (for instance) Google Display Network.

Analysis of the scale of paid ads to OTA supports our assumption

that airlines will have to invest heavily to drive traffic to their

websites especially in non-base markets

OTA analysis

Page 32

Channels - Travel Management Companies:

TMCs experience strong corporate customer loyalty who require high

value services. TMC services are not easily replicated by airlines

Overall TMC estimated to take 25-50% of the market outside of

the LCC (source: Infrata)

TMC market is ‘loyal’ due to business service offering:

•  Manage travel supply contracts

•  Procurement programmes

•  Help establish & enforce policies

•  Cost reduction & productivity enhancing services

•  Travel expense reporting

•  Travel policies

•  Visas and passports

•  Out of hours services

•  Offer important assistance in times of disruption.

•  Corporate agency and inplants: supplies specialist travel

services to business customers

•  The business travel community sees great value in TMC

service; this would be difficult for the airlines to replicate.

TMC traffic is considered to be difficult to channel shift due to

the service requirements of the business market

TMC analysis

Source: BHS, Uniglobe/CWT

Page 33

33%

27%

25%

18%

28%

22%

21%

39%

51%

54%

73%

Italy Poland UK Scandinavia

Revenue Sources Mix for Business Travel

Agencies

Fees

Override/SLA (includes

GDS incentives)

Commissions (air if applied,

cars, hotels, etc)

Channels - Call Centres:

Airline call centre, ATO, CTO have reduced in importance to airlines

Call centres: typically taking 3-5% of airline bookings as most direct sales

‘encouraged’ to website. Costs are difficult to ascertain, costs per call

estimated at €20-30 per ticket issued.

Call agents estimated to handle 7-10 calls per hour

ATO/CTO: increasingly a sales support function with a greater

requirement for airlines with large international destinations. May also be

fulfilled by a General Sales Agent.

TMC/TA/OTA currently process majority of calls. Expedia has 15,000 call

centre staff.

Airline costs have been derived from a number of sources and have been incorporated into

the model. Assumptions have been made regarding cost development in each scenario

Source: M Hanke/OLTA discussions

Call centre, ATO, CTO increase with higher ‘other end sales’

Call centre analysis

Page 34

Channels - Affiliate

Affiliate programs are a growing but is an ‘obscure’ distribution channel.

Costs to airlines may have ‘escaped’ previous studies

Affiliate partners have the rights to distribute airline products by agreement.

There is an emerging body of affiliate networks serving the airlines including

Affiliate Future, Commission Junction and Google Affiliate Network.

The British Airways Affiliate Marketing programme is an advertising

programme that rewards media or site owners for displaying British Airways

affiliate advertisements that link directly to ba.com for purchase on the site.

Affiliates earn commission on qualifying transactions which include a valid

flight, holiday, car rental, or hotel booking made on ba.com. The link must be

the last one used to direct the customer to make their booking on ba.com.

Commissions vary by product, site type and region.

Turkish Airlines operates its Affiliate program with Digitouch in Turkey and

Tradedoubler worldwide. Turkish Airlines only deals with price-comparison

(meta-search) websites and have agreements with all local and global meta-

search websites through its Affiliate program.

Commission is earned through online customers purchasing air tickets.

Commission claims are generated when a passenger buys a ticket online.

No commission is earned if the passenger only makes a reservation.

Also, no commission is earned if the purchased ticket is cancelled within 30

days.

Turkish Airlines Affiliate programme

Affiliate activity is performance-based marketing in which a

business rewards one or more affiliates for each visitor or

customer brought by the affiliate's own marketing efforts.

Affiliate programs extend airline’s reach through using partner’s

platforms. Affiliates include other travel providers, agencies and

meta search. Examples of airlines using affiliates are

Singapore Airlines using Affiliate Futures and the Turkish

Airlines program described below.

Affiliate programs have been included in the model with assumptions

on commission rates and channel percentage derived from web

search and discussions with airlines

Ticket Type Commission Rates in All

Classes

Domestic Ticket Sales €0.50 (per PNR**)

International Ticket Sales €5 (per PNR**)

*Cookie period: valid for visitors’ immediate session on price

comparison sites. Permission is granted for a 30 minute cookie

period.

**Reservation code (PNR): the same reservation code can

include more than one passenger.

Source: Turkish Airlines

Affiliates analysis

Page 35

Channels - Travel Agents:

Travel agents provide services to clients which would otherwise have to

be covered by airlines

Agencies have seen airline ticket commissions steadily diminish since

2000.

This put pressure on the agency community to re-invent their role, their

service and their value to the end customer whether this be a corporate

or leisure traveller.

Travel agents now receive the majority of their income (fees) from clients

not airlines.

Travel agents have changed their revenue model from supplier-led

commissions to a client services and retail model where the end user

pays for the service they purchase.

Travel agents not booking much scheduled leisure travel, estimated 5%

of their sales. Leisure passengers flying on scheduled services not as

part of a group will book mainly on OTA. (Source: Infrata).

Travel agents undertake considerable number of customer service functions on

behalf of airlines – not paid by airlines

N%&%(+%='9W'#(Q>;Q%*'<%:X):E%='X):'#;:&;$%*'#$='()$*@E%:*'

•  80% of UK bookings with itinerary changes are never

ticketed - huge amount of uncharged time spent by

agents preparing and amending itineraries never

confirmed, ticketed and billed. These costs are borne

by the agents, not the airlines.(1)

•  42% of UK bookings are changed prior to ticketing.

This could be due to a number of reasons: agents

asked to hold multiple flights, routes, classes and

dates. (1)

•  Agents required to utilise multiple distribution

channels to obtain data, acquire best price and meet

client product requirements.

•  Offer important assistance in times of disruption.

•  Fees paid by airlines to agents have been reduced to

nil in many markets and 1-3% in others. (2)

The impacts of airlines selling a smaller share through TA sold

are:

•  Some lower costs due to agents commission, payment

processing

•  Some higher costs due to increased burden of customer

support, merchant costs, fraud costs that would normally

have been provided by TA

Travel Agents analysis

(1) Source: Travelution / Infrata

(2) Source: Amadeus ‘Service Fees and Commission Cuts’

Page 36

Channels - Airlines:

Airlines internal departments support distribution and marketing. Many of

these costs increase when going ‘direct’

Internal Function Distribution Cost Impact Model Factor

Customer support Major switches: agent to

direct increases airline

work

Derived estimated

cost per call and calls

per passenger

Fraud Some impact of switching

from agents to direct

Adjustment to % of

passenger revenue

Credit card Credit card commissions

sensitive to direct sales

Credit card cost as %

of revenue

adjustment

Cash flow Some increase in

payment speed with direct

sales

Cash flow factor

Revenue accounts Minimal impact on

administration

Small adjustment to

model

Revenue management Minimal impact on

administration

Small adjustment to

model

Other costs Minimal impact on

administration

Small adjustment to

model

Processes and infrastructure required to support distribution

Airline internal costs covers a range of activities required to

process bookings and revenue.

They also include commercial structures although these

are included in customer acquisition in the model.

The major internal support functions in terms of cost are

•  Credit card commissions

•  Fraud

•  Customer service

These costs are particularly ‘channel sensitive’ as they are

largely provided by the TA, TMC, OTA – moving to direct

sales will bring them in-house.

The other costs such as revenue accounts, revenue

management are largely insensitive to channel used.

Airline internal expenses will move with channel and are

a major driver of cost

Airline internal costs analysis

Page 37

Channels - Airline internal costs:

Commercial structures are often not fully considered when allocating

contribution to distribution costs

The number and cost of the support staff and infrastructure

varies greatly with the legacy airline having 100-200 and the

low cost having fewer than 50. The legacy airlines have

several thousand call centre staff are the LCC have far fewer.

The distribution cost implications are:

•  Different structures are required to support different

channel mixes

•  LCC typically award higher seniority to websales support

•  Network carriers have more expensive field sales, GSA and

group sales team

•  Customer support function higher level of seniority and

proportion of cost for LCC (network passengers may use

TA/TMC)

The differing commercial structures impact the way airlines allocate budget and report on costs

Source: Infrata

Typical ‘Network’ Airline and LCC Structures

Brick and

mortar

agents and

tour

operators

Network

CEO

Head of

Marketing

Advertising

Own Website

Field Sales

GSA

Head of Sales

Brick and

mortar

agencies

OLTA

LCC

CCO

Call Centres

Head of

Sales

Own Website

& OLTA

Head of

Marketing

Tour

Operators

Specialty

Outlets

Advertising

There is a generalised relationship of airline

structure to higher levels of direct distribution.

•  Costs down: payment processing, field sales,

agency support, GSA support

•  Costs up: customer support / call centres

Airline commercial structures analysis

Note that these are representative structures and they vary

among airlines even in the same category.

Page 38

Channels - Customer acquisition:

Offline and online display use a wide range of media – buying and

managing this is a major cost - rarely included in analysis

Airlines spread marketing between offline and online: Major

categories of marketing spend are:

•  Online: Digital / internet display costs has plateaued

(excluding google ads)

•  Offline: Newsprint still largest spend group

•  Other categories starting to decline for network but growing

for LCC

•  TV major growth area for LCC competing in leisure market

European Network Carrier

European LCC

Source: bradtop100 2010

Offline sales and marketing is vital to raise awareness and vital to drive ‘native search’ traffic to airlines’ websites

Distribution Channel Mix of Network and LCC (2010)

Offline channel costs based upon information derived

from airline/published sources are used in the model

Offline media analysis

Page 39

Channels - Ancillary products :

Currently, airlines prefer to sell ancillaries on their own websites.

Ancillary services are becoming more important to many airlines,

depending on their competitive environment. These services

are either flight-related (extra legroom, priority boarding) or

additional non-airline products like car rental, hotel, or insurance.

Airlines choose to sell these services predominantly through

their own websites, to maximize upsell and increase loyalty.

Realising that travel agencies represent a key distribution

channel also for flight-related ancillary services, an increasing

number of airlines are distributing this content either directly

from their own inventory, via an aggregator, or via GDSs. Rich

ancillary content is being enabled by XML messages, including

those defined as part of the IATA NDC XML standard.

Ancillary revenues vary considerably by airline.

The model used ‘mid point’ ancillary revenue per

passenger for network and regional

Source: IATA

Ancillaries analysis

Page 40

Ancillary Services Distribution

Consumers

TMC /OTA

Airline Reservation Systems

Airline

Website /

e-commerce

engine

GDS

New

Entrant

XML and other

interfaces

Model Output

Page 41

Modeling Scenarios & Drivers

Five categories of variables were modelled to identify the cost impact by carrier type:

1.  Internal organisational structures leading to differing reporting and budget lines

2.  Differing airline levels of connecting and point-to-point traffic

3.  Differing geographic markets (home market versus non-home markets)

4.  Differing marketing arrangements with the indirect channel

5.  Non standard allocation of full loaded costs

The model aims to quantify these dynamics in a fully allocated cost manner to

demonstrate the possible impacts of different regimes of direct / indirect distribution.

The following pages show:

-  The composition of the model

-  The headline results and a comparison of all airline types impact of moving to more

direct distribution

-  Analyses of the impact of moving to more direct for all the airline types shown is

a series of results charts

Page 42

Page

Our model uses a dynamic relationship between the three known areas of cost – Distribution, Payment and Customer Acquisition.

1.  Inventory management system costs

2.  GDS booking fees

3.  Travel Agents’ commission

4.  IT & infrastructure/utility costs

5.  Customer service

1.  Merchant costs

2.  Website and tech product costs

3.  Ancillary services

4.  Other airline internal costs (administration)

1.  Online advertising (Google ads, etc)

2.  Offline advertising (TV, radio, print media etc)

3.  Meta search referrals

4.  Other airline internal costs

(sales staff, offices, marketing team)

Customer Acquisition

‘Classic’ distribution costs included in most industry

comparisons of cost. Clearly understood

relationship to direct/ indirect distribution channels

Sales administration – some generally included in

industry comparisons. Link to distribution channels

not present in most analyses.

Cost of acquiring customers, largely not included in

industry channel cost comparisons, link to

distribution channels not present in most analyses.

Payment, finance and administration

Distribution

In addition there is a growing separate cost and revenue stream meeting the needs of Ancillary services

Model Output - Composition:

Airline sales & distribution costs ‘fall’ into 3 main categories

Page 43

1.  Corporate incentives, bonuses, discounts

2.  Corporate traveller extra benefits

3.  Travel Agent bonuses, IT integration costs

Other Acquisition Costs

Cost of making customers stop using a neutral

distribution channel and start using an airline

controlled one.

Often significant, but costs not included in study, as

reliable data is unavailable

Model Output - Composition:

In addition to the costs, the model is driven by the interaction between

customer groups and the channel

Sources : Credit Suisse, HRS, HRG, Tnooz.

The airline distribution model comprises 3 sets of variables. Effective modelling required detailed

understanding of each sub-component.

In the modelling we tested the impact of flexing different components to see the impact on the ‘system’ including

customer group, channel mix and cost per booking per channel

The model was developed to

address the following questions:

•  What is the average cost per

booking by channel?

•  What are the major cost

elements that build up to

average total cost

•  What is the cost impact for

airline in switching from

indirect to direct channels

The analyses in this section

show the results of calculating

fully allocated costs by channel

•  Network (large

home market)

•  Network (small

home market)

•  Regional airline

Direct

‒  Website (airlines)

‒  ATO /CTO /Call

Centre

‒  Meta search

Indirect

‒  TMC

‒  TA incl. Direct

Connect

‒  OTA

‒  ‘Direct connect’

•  Cost by channel

Scenarios by airline

type:

•  47% direct

•  67% direct

Page 44

Customer Groups Channels Costs

•  Distribution

•  Finance, payment &

admin

•  Customer acquisition

•  Ancillary

Model Results

Model Output - Composition:

The model is based around three types of airline

1.  The model considered the following channels:

•  Airline website

•  Call centre / ATO / CTO and field sales

•  Affiliates

•  Travel agents

•  Online travel agents and meta

•  Travel management companies

2.  The model considered customer acquisition costs

•  Online

•  Offline

3.  The model considered internal ‘distribution-associated’

costs:

•  Merchant and credit card costs

•  Customer service

•  Fraud

•  Airline internal costs: revenue management, revenue

accounts, other departments

The model was used to assess distribution costs for each of the four categories of airline

Source:

•  Data to populate the model has come from airlines, travel agents, TMC, OLTA, metasearch engine providers, ETTSA and its affiliated members.

•  Data has also been sourced from websearch and journal reviews.

Airline Type Characteristics

1 Network (large)

‘Baseline’ group with typical mix of

channels

2 Regional

Managing channels to deal with less

market power than larger rivals

3 Network (Small)

More distributed market with

challenging ‘customer acquisition’

Page 45

Model Output: Cost implications- Network airlines:

Network airlines (large home market) have 0.8% reduction in total

distribution cost with shift to more direct sales

The charts and data show the present full cost of distribution separated

into the four main cost groups. The main assumption is that airline

websales increase to 60% from the base of 40% and direct sales

increase from 47% to 67% including call centre, ATO and CTO.

The model assumes an increase of 0.5% average ads cost for every 1%

increase in web share due to the difficulty in attracting the more resistant

consumers.

The analysis shows that some cost groups decrease with a move to

more direct sales. Distribution costs are reduced as fewer bookings go

through the GDS and incur booking fees and there is also savings from

less agents commission.

The benefit is negated to some extent because the airlines have to

increase their advertising, in particular websearch ads to drive traffic to

their website.

Also there are increased costs of customer service that agents provide

for customers and the credit card costs, some fraud costs and the cost of

managing customer changes would fall on the airline.

Total cost per booking decreases by €0.11 due to:

•  Higher customer acquisition

•  Credit card costs imposed on the airline, not TMC/OTA

•  Distribution cost reduces with less GDS booking fees

The analysis indicates that Network carriers (large home market)

experience immaterial benefits from a major shift away from

indirect channels.

Page 46

Scenario

(Channel %)

Base

47%

High

67%

Variance

Customer acquisition (€) 2.58 3.56 0.98

Channel distribution (€) 6.75 5.11 (1.64)

Payment, admin.,

finance (€)

4.04 4.59 0.55

Ancillary (€) 0.06 0.06 0.00

Total (€) 13.43 13.32 (0.11)

Source: Infrata based on airline analysis

Network airlines full distribution cost per booking €

7?77'

/?77'

1?77'

2?77'

0?77'

6?77'

.?77'

3?77'

8?77'

Customer

Acquisition Costs

Channel Costs Payment &

Finance Costs

Ancilliary Costs

!)*+'<%:'<#**%$I%:'Y,Z'

-#*%' ";I"'=;:%(+'

Model Output: Cost implications- Regional airlines:

Regional airlines have 3.9% increase in total distribution cost with shift

to more direct sales

The charts and data show the present full cost of distribution separated

into the four main cost groups. The main assumption is that airline

websales increase to 60% and total direct sales to 67%.

The model assumes an increase of 0.5% average ads cost and

advertising cost for every 1% increase in web share due to the difficulty

in attracting the more resistant consumers.

The analysis shows that some cost groups decrease with a move to

more direct sales. Distribution costs are reduced as fewer bookings go

through the GDS and incur booking fees and there is also savings from

less agents commission.

The benefit is negated to some extent because the airlines have to

increase their advertising, in particular websearch ads to drive traffic to

their website.

Also there are increased costs of customer service that agents provide

for customers and the credit card costs, some fraud costs and the cost

of managing customer changes would fall on the airline.

Total cost per booking increases by €0.52 due to:

•  Higher customer acquisition

•  Credit card costs imposed on the airline, not TMC/OTA

•  Mitigated by distribution cost reducing with less GDS booking fees

Customer acquisition costs and payment costs eliminates

the benefit of the shift to direct.

Page 47

Source: Infrata based on airline analysis

Scenario

(Channel %)

Base

47%

High

67%

Variance

Customer acquisition (€) 2.64 4.0 1.52

Channel distribution (€) 7.3 5.83 (1.47)

Payment, admin.,

finance (€)

3.46 4.09 0.62

Ancillary (€) 0.06 0.6 0

Total (€) 13.46 13.98 0.52

Regional airlines full distribution cost per booking €

Model Output: Cost implications- Network airlines:

Network airlines (small home market) have 3.9% increase in total

distribution cost with shift to more direct sales

The charts and data show the present full cost of distribution separated into

the four main cost groups. The main assumption is that airline websales

increase to 60% and total direct sales to 67%.

The model assumes an increase of 4% average ads cost and advertising

cost for every 1% increase in web share due to the difficulty in attracting the

more resistant consumers.

The analysis shows that some cost groups decrease with a move to more

direct sales. Channel distribution costs are reduced as fewer bookings go

through the GDS and incur booking fees and there is also savings from less

agents commission.

The benefit is negated to some extent because the airlines have to increase

their advertising, in particular websearch ads to drive traffic to their website.

Also there are increased costs of customer service that agents provide for

customers and the credit card costs, some fraud costs and the cost of

managing customer changes would fall on the airline.

Total cost per booking increases by €0.6 due to:

•  Higher customer acquisition

•  Credit card costs imposed on the airline, not TMC/OTA

•  Mitigated by distribution cost decreasing with less GDS booking fees

•  Very high cost of attracting ‘marginal’ passengers in markets where the

airline has a relatively weak marketing position

Customer acquisition costs and payment costs eliminates the

benefit of the shift to direct.

Page 48

Network airlines (small home market) airlines full

distribution cost per booking €

Source: Infrata based on airline analysis

Scenario

(Channel %)

Base

47%

High

67%

Variance

Customer acquisition (€) 3.32 4.93 1.62

Channel distribution (€) 7.3 5.74 (1.56)

Payment, admin.,

finance (€)

3.94 4.47 0.53

Ancillary (€) 0.12 0.11 (0.02)

Total (€) 14.69 15.26 0.57

Appendices and Model Overview

Page 49

Description of Airline Distribution Cost Model

Airline Type

Airline Types

(can expand and may

need Overall)

Distribution Channels

Distribution Channel

Multiple airlines per type

or multiple sources per

airline type and channel

Revenue

Distribution

Costs

Customer

Acquisition

Costs

Fixed Costs Fixed Costs

Variable

Costs

Variable

Costs

Margin

Base Output:

Margin per PB

Revenue and distribution cost

per channel for each airline