It’s no longer a secret that clustering passengers into segments translates to higher profitability. Then, the two important questions are what are the possible segmentation models, and which segments of passengers are of higher value to the travel business?
Well, there are a set of demography-based segments of passengers which would always be helpful to target marketing strategies of the business. For instance, segments of passengers based on combinations of 3 data points like nationality, amount of purchase, and last booking time can help to target those in the risk of churn (those with big gap, and less expenditure) grouped by the nationality, as well as identifying those passengers to be retained.
Given only a minimum number of 20 data points (e.g. age, gender, nationality, marital status, etc.), and targeting to build segments based on 3 data points, one can build 1,140 distinctive clusters of passengers! But, when businesses face overloaded data points of the passengers’ personal information and transactions, prioritising the endless number of segmentation would not only save a huge amount of time operationalizing them, but also identifying new marketing opportunities.
Here, different actionable and result-oriented passenger segmentation models are explained, and their relevance and importance in the case of travel industry are briefly discussed.
1- Geographic Segmentation
This is one of the traditional models which targets passengers based on location they are residing in. Following this strategy, regional differences of the passengers can be easily identified, and a large amount of data will be available.
Geographical modeling provides a convenient organizational framework so that the business can manage resources and services in different location within countries, regions, and cities. For instance, passengers in different states within the same country might have different preferences of time for travelling; as a result of state events and holidays. Criterias for geographic segmentation can be whether continent, country, region, state, or city.
2- Demographic Segmentation (B2C)
In case of the B2C, demographic segmentation can help the business to identify patterns in existing and future passengers’ needs. The criterias for this segmentation can based on age, gender, ethnicity, marital status, or education.
Although there is not a ranked list of criterias in demography-based segmentation, conventionally, the gender-based models have provood to provide better insights as gender is a more distinctive determinant to separate patterns. Perhaps, after age comes the age, ethnicity, and marital status, respectively.
3- Demographic Segmentation (B2B)
Unlike B2C, the demographic criterias for B2B are differently listed as industry type, number of employees in the business, the business lifecycle, financials, or involvement in technology.
4- Psychographic Segmentation
As conveyed by the name, the psychographic segmentation considers psychological patterns of passengers. The criteria might be social class, lifestyle, presence in social media, personalities, or values.
The idea is to identify attitude of distinctive clusters of passengers towards specific set of products that influences their subsequent purchase of those products. The model assumes passengers with similar psychological preferences can be separated, and as result important insights can be derived from these segments.
5- Behavioral Segment
In the behavioral segmentation scheme, passengers are separated according what they do based on insights from their actions. The criterias for behavioral segmentation are spending pattern, mode of payment, likes/dislikes, awareness, or interest.
The Cycle of Prioritization
Understanding different segmentation models is the first step towards planning the business resources. In practise, the process of selecting the model and its criteria for inclusion is a repetitive task which involves both data science and marketing teams.
Although design and implementation of the segmentation model is the task of data science team in the companies, insights of the marketing team always add to more interesting and analytic interpretations of the segmented clusters of passengers.
Basically, there are five steps towards deciding the most important segmentation model for the firms as (1) Model Selection, (2) Criteria Selection, (3) Model Value Validation, (4) Criteria Refinement, and (5) Model Operationalization.
Effective segmentation of passengers is an absolute must for the marketing teams in travel industries. However, due to the large number of segmentation models, it’s essential to select the most suitable model for the urgent needs of the firm which is in line with the marketing priorities.
In this article, five approaches for segmentation of passengers were mentioned and briefly reviewed to assist firms with prioritizing their resource allocation and marketing strategies according to each unique segmentation model.
In future, we will extend this introductory article by discussing more about technical implementation aspect of each passenger segmentations one by one. More specifically, the architectural design, choice of technology, and practical implementation strategies will be explained.
Rasoul is Data Science and Machine Learning enthusiast with a Ph.D. in Deep Learning and Scalable Information Retrieval.
He is working as a Data Scientist in the Data Analytics Department, GoQuo (Sdn Bhd) Malaysia where he is developing Marketing Automation and Recommender Systems aiming to empower Airline E-commerce, Ancillary, Loyalty, and Customer Analytical Platforms.