Resume
Summary Marketing
- Cours
- Marketing
- Établissement
- Universiteit Gent (UGent)
Samenvatting van de papers gegeven door Prof. Dr. D. Van den Poel
[Montrer plus]
Vendeur
S'abonner
Aperçu du contenu
THE ADDED VALUE OF SOCIAL MEDIA DATAIN B2B CUSTOMER ACQUISITION SYSTEMS: A
REAL-LIFE EXPERIMENT
Examine the added value of Facebook data in the customer acquisition process. Results: Facebook is
the most informative data source to qualify prospects, and is complementary with the other data
sources in that it further improves predictive performance. Facebook data challenge current best
practices in customer acquisition.
INTRODUCTION
Focus on business-to-consumer (B2C) applications.
Within business-to-business (B2B) environments, the potential of social media has already been
recognized, but the adoption of social media is slower compared to B2C companies.
There is a need to explore the added value of big data applications and analytics in business
environments, thereby taking into account the data, tools and algorithms that can be used.
In the history of customer relationship management (CRM), the acquisition process has received less
attention compared to retention and customer lifetime value (CLV). The underlying reasons are that
the customer acquisition process is more complex, less data of poorer quality are available, and
customer acquisition is typically more expensive compared to retention campaigns.
Evaluate the value of data extracted from the prospects’ social media page (Facebook pages), and
compare it with data extracted from their website, and data that the focal company buys from a
specialized vendor.
LITERATURE REVIEW
B2B ACQUISITION FRAMEWORK
Decision Making Unit (DMU): organizations’ buying decisions are taken by this group of people.
“Sales funnel”: 4 stages
1. Only a list of suspects: all potential new customers
a. make use of a profiling method
2. Reduce the list to a list of prospects that can be identified
a. companies start the sales process
b. starts when feedback on the first list of prospects is received
3. Qualifying the prospects → list of leads
a. based on intuition, feeling and simple rules
4. Lead is converted to a real customer
→ iterative method
• Phase 1: estimate and evaluate the quality of the probability of prespects to become a
customer, based on the similarity with customers
• Phase 2: with feedback data, model which prospects will be converted into customers,
based on information from previous successful conversions
Marketing 1 Papers 2019-2020 Casier Tessa
1
,SOCIAL MEDIA IN A B2B SALES CONTEXT
The use of social media within a B2B context provides new opportunities to improve sales force
effectiveness.
Three stages in which social media can be used:
1. Prospecting
• sales representatives use social media to identify potential buyers
2. Qualifying leads
• the quality of these leads is examined using information available on social media
3. Managing relationships
• social media can be used to manage the relationships with existing customers
Similar: 3-step process using social media: (the difference is that the relationship stages are
expanded over several categories)
1. crating opportunity
2. understanding customers
3. relationship management
Social media can and should be integrated into the entire sales process.
→ Social media are important in a B2B context
Social media are recognized as a tool to make the sales process less costly and more effective and are
seen as an extension of traditional customer relationship management (CRM), leading to Social CRM
activities.
The current qualitative focus on social media in the literature ignores important opportunities,
related to the big data nature of social media.
Focus on the ‘qualifying stage’ of the sales process. Move social media use in a B2B context from a
purely describing, qualitative view to a data-oriented which uses information systems to collect,
clean and analyse the data based on machine learning techniques.
SOCIAL MEDIA AS A DATA SOURCE
Main challenge when qualifying leads: the lack of qualifying characteristics
• The seller can only rely on data that is either publicly available or available for purchase, as
there is no formal relationship with the prospect yet.
Data collection: 3 sources
• commercially purchased data
o expensive
o poor quality → provide ‘best estimates’ of data → missing values
• websites
• social media
o information is updated more frequently than on websites
o different information types are available
3 types of data that help qualifying B2B customers
• company characteristics
• customer behaviour
• attitudinal information
Marketing 1 Papers 2019-2020 Casier Tessa
2
,Given the rich information present in social media data, we hypothesize social media data to be most
predictive for customer acquisition, as data quality is the best driver to boost an analytical model’s
performance.
METHODOLOGY
DATA
Literature review: external sources are crucial to obtaining information on prospects.
3 types of data sources:
• purchased commercial data from a specialised vendor
• data from the prospects’ web pages
• data from the prospects’ Facebook pages
COMMERCIAL DATA
Type of information: company size, industry type and other business demographics.
WEB DATA
Text mining
• raw text cleaning + stop word removal
• producing a document-term matrix
Also include expert knowledge (= the information that is deemed important by the salespersons
based on previous experience)
FACEBOOK PAGES
All information about a prospect that can be found on the Facebook page of a prospect.
Data drawn form the Facebook page: 2 categories
• company characteristics
• attitudinal information
MODELS
PHASE I MODELS
2 models:
• Initial model
o assumes no knowledge about converted vs non-converted prospects (Phase I)
o look-alike model or profiling
• Rank the prospects based on their predicted score (represents probability of becoming a
customer)
EXPERIMENT
List of the prospects, ranked from high to low probability of becoming a customer
Prospects were randomly assigned
PHASE II MODELS
Model converted vs. non-converted prospects
Marketing 1 Papers 2019-2020 Casier Tessa
3
, MODEL PERFORMANCE
Evaluation based on:
• AUC
o the probability that a randomly chosen positive observation is scored higher than a
randomly chosen negative observation
o focus on the entire range
• Lift over random selection
o focus on the observations with the highest predicted probabilities
RESULTS
Facebook data is better than commercial data, but only slightly better than website data.
We have superior performance of combined models.
DISCUSSION
The sales process can be improved by using social media in a way that was not explored yet, i.e. using
a data mining approach to social media in a formal information system.
Automatic handling of Facebook pages is a valuable tool to
• improve the qualification prediction of prospects into leads worth pursuing
• reduce the time needed to screen the Facebook page drastically
ADDED VALUE OF FACEBOOK INFORMATION
Social media pages are the most informative.
The two data types within Facebook data contain different information, which renders the Facebook
pages the most interesting data source.
The company characteristics contained within the Facebook information do a slightly better job in
predicting successful conversions compared to characteristics in commercial information.
COMBINATION OF DATA SOURCES
Phase I models:
• combining all data sources yielded best performance in terms of AUC, while the lift curves
were more in favour of a model combining Facebook data and commercial data
Phase II models:
• the model combining Facebook data and commercial data proved to be the best
It may be worth to build models based on freely available data only.
Social media provide opportunities to search online for potential prospects, although developing this
search models again requires effort and time.
ITERATIVE PROCESS OF THE SALES FUNNEL MODEL
Phase II model performs better compared to the Phase I model. In Phase II, we explicitly model good
or converted prospects. In Phase I, we aim to identify good prospects by comparing them to existing
customers.
Marketing 1 Papers 2019-2020 Casier Tessa
4
Les avantages d'acheter des résumés chez Stuvia:
Qualité garantie par les avis des clients
Les clients de Stuvia ont évalués plus de 700 000 résumés. C'est comme ça que vous savez que vous achetez les meilleurs documents.
L’achat facile et rapide
Vous pouvez payer rapidement avec iDeal, carte de crédit ou Stuvia-crédit pour les résumés. Il n'y a pas d'adhésion nécessaire.
Focus sur l’essentiel
Vos camarades écrivent eux-mêmes les notes d’étude, c’est pourquoi les documents sont toujours fiables et à jour. Cela garantit que vous arrivez rapidement au coeur du matériel.
Foire aux questions
Qu'est-ce que j'obtiens en achetant ce document ?
Vous obtenez un PDF, disponible immédiatement après votre achat. Le document acheté est accessible à tout moment, n'importe où et indéfiniment via votre profil.
Garantie de remboursement : comment ça marche ?
Notre garantie de satisfaction garantit que vous trouverez toujours un document d'étude qui vous convient. Vous remplissez un formulaire et notre équipe du service client s'occupe du reste.
Auprès de qui est-ce que j'achète ce résumé ?
Stuvia est une place de marché. Alors, vous n'achetez donc pas ce document chez nous, mais auprès du vendeur tessacasier. Stuvia facilite les paiements au vendeur.
Est-ce que j'aurai un abonnement?
Non, vous n'achetez ce résumé que pour €5,49. Vous n'êtes lié à rien après votre achat.
Peut-on faire confiance à Stuvia ?
4.6 étoiles sur Google & Trustpilot (+1000 avis)
73314 résumés ont été vendus ces 30 derniers jours
Fondée en 2010, la référence pour acheter des résumés depuis déjà 14 ans