ArticlePDF Available

Abstract and Figures

Most of today’s digital marketing campaigns which are sent through email and mobile messaging are bulk campaigns which deliver the same message at the same time to all customers, regardless of their needs and preferences. The outcomes are bad customer experience, low engagement and low conversion rates. Modern marketing automation tools aim to facilitate personalized communications, such as scheduling of individual marketing messages based on each individual subscriber’s profile. This research focuses on the problem of automatically deciding on the optimal date and time for sending consent-based personalized marketing messages. We specifically focus on the case of repeat consumers of consumer packaged goods (CPG) which require regular replacement or replenishment. The objective is to timely anticipate the needs of consumers in order to increase their level of engagement as well as the rate at which they repurchase products. The proposed solution is based on a regression model trained with transactional data and instant messaging metadata. We describe the way such a model can be created and deployed to a scalable high-performance environment and provide pilot evaluation results that suggest a significant improvement in marketing effectiveness.
Content may be subject to copyright.
(IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 11, No. 4, 2020
Predicting the Optimal Date and Time to Send
Personalized Marketing Messages to Repeat Buyers
Alexandros Deligiannis1, Charalampos Argyriou2, Dimitrios Kourtesis3
Research & Development Department
Apifon S.A.
Greece
Abstract—Most of today’s digital marketing campaigns which
are sent through email and mobile messaging are bulk campaigns
which deliver the same message at the same time to all customers,
regardless of their needs and preferences. The outcomes are bad
customer experience, low engagement and low conversion rates.
Modern marketing automation tools aim to facilitate personalized
communications, such as scheduling of individual marketing
messages based on each individual subscriber’s profile. This
research focuses on the problem of automatically deciding on the
optimal date and time for sending consent-based personalized
marketing messages. We specifically focus on the case of repeat
consumers of consumer packaged goods (CPG) which require
regular replacement or replenishment. The objective is to timely
anticipate the needs of consumers in order to increase their level
of engagement as well as the rate at which they repurchase
products. The proposed solution is based on a regression model
trained with transactional data and instant messaging metadata.
We describe the way such a model can be created and deployed
to a scalable high-performance environment and provide pilot
evaluation results that suggest a significant improvement in
marketing effectiveness.
KeywordsPersonalized marketing automation; customer re-
lationship management; conversion rate optimization; customer
engagement; machine learning; XGBoost regression; cloud com-
puting; data privacy
I. INTRODUCTION
The objective of marketing communications is to reach
and engage customers through effective marketing campaigns.
Effectiveness is predicated on several different properties of a
marketing communication activity. Two highly critical proper-
ties are how relevant the marketing message is to the needs
and interests of each customer and how well it is timed [1].
The ability of an organization to deliver engaging marketing
communication experiences that achieve high conversion rates
requires solutions for optimizing both content and timing.
Most of today’s digital marketing campaigns sent through
email and mobile messaging are bulk campaigns which deliver
the same message at the same time to all customers, regardless
of their needs and preferences. The typical consequences are
bad customer experience, low engagement and low conversion
rates [2]. Marketing professionals attempt to improve outcomes
through audience segmentation, e.g. by dividing recipients into
smaller groups. In this way, they can differentiate the type
of message and time of delivery for different groups [3]. In
an ideal setting, these customer segments would be defined
based on customers’ past purchase history and engagement
behaviour relative to similar past campaigns. However, such
data segmentation is rarely available or practical to analyse.
Modern organizations seek to adapt the message and timing
of marketing communications to the needs and preferences
of individual customers (i.e. marketing to segments of one)
in order to achieve the highest marketing effectiveness [4].
Although personalization at scale remains a hard problem
for marketing communication professionals, such marketing
automation technology will allow organizations to leverage
unique customer profiles derived from past transaction and
engagement data to deliver the right message at the optimal
date and time to each individual customer [5].
A growing number of marketing technology providers
are investing research and development resources to address
the challenge of personalized marketing at scale [6]. Taking
into account recent research advances in machine learning
and predictive analytics [7], this paper presents some of the
research results obtained from project PRIME - an R&D effort
by business messaging technology provider Apifon [8].
PRIME focuses on developing new services for consent-
based personalization of business-to-consumer mobile commu-
nications, utilizing data from past purchase transactions and
past exchanges of instant messages between the business and
the customer. In this paper, we focus on the part of this research
that relates to optimizing the time that marketing messages are
scheduled for delivery to repeat buyers of consumer packaged
goods (CPG) via instant messaging. Specifically, we look at
the case of products that require frequent replenishment - such
as food & beverages, cosmetics or household products.
Our proposed solution utilizes predictive models which are
trained with data from previous purchase transactions as well
as past message exchanges. The objective is to improve the
timing of messages relative to subscribers’ predicted needs and
to increase the rate of follow-on purchases.
Section II outlines some of the published research relating
to this work, while Section III presents the scope and focus
of our research. Section IV provides a walk-through of the
required data processing approach and Section V discusses
the date and time optimization model in detail, from how
predictive models can be created based on a scalable machine
learning approach to how they can be deployed to a cloud
infrastructure. Section VI presents the evaluation method and
Section VII discusses the results of a real world pilot case
study. The results suggest a measurable improvement in mar-
keting effectiveness as measured by customer engagement and
conversion rates (product repurchases).
www.ijacsa.thesai.org 90 |Page
(IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 11, No. 4, 2020
II. RE LATE D WOR K
A. Time-Aware Recommender Systems
Much of the published research results in the field of
personalized recommender systems integrate complementary
contextual information aiming to improve the efficacy of
recommendations, notably the timing and frequency of in-
teractions between consumers and content or products. Ding
and Li observed that Collaborative Filtering methods are not
sensitive to changes in user purchase interests over time [9].
They suggested that any product or item which was recently
rated by a user should have a bigger impact on the prediction of
the particular user’s future behaviour compared to an item that
was rated by the same user a long time ago. They proposed an
algorithm to compute time weights for different products/items
rated by a user, in a manner that assigns a decreasing weight
to old data, thereby introducing a personalized interest decay
factor.
Koren also observed that customer inclinations and tastes
are evolving and that capturing time drifting patterns in user
behavior is essential to improving the accuracy of recom-
menders [10]. He however argued that classical time-window
or instance decay approaches are limited, as they lose too
much signal when discarding data instances. He proposed a
more sensitive model which can better distinguish between
transient effects and long term patterns in customer interests
and evaluated the model on a large movie rating dataset by
Netflix, showing increased prediction accuracy.
Baltrunas and Amatriain introduced a time-aware recom-
mender system aiming to accurately predict a user’s music
tastes given the time of day, week or year [11]. Their approach
assumes that user preferences change with time but have
temporal repetition. The main idea of the approach is to
partition a user profile into smaller, time-dependent profiles
and use these micro-profiles for the prediction task, instead of
a single profile.
B. Repeat Purchase Prediction
Wang and Zhang worked on augmenting recommender
systems to promote repetitive purchases [12]. They adapted the
proportional hazards modeling approach in survival analysis
to the recommendation research field and proposed a new
opportunity model to explicitly incorporate time in an e-
commerce recommender system. The model can estimate the
joint probability of a user making a follow-up purchase of a
particular product at a particular time.
A common obstacle in predicting repetitive purchase events
is the “noise” introduced by customers’ irregular purchases.
Dey et al. addressed the problem by building a Poisson-Gamma
model to estimate the average purchase frequency, along with
a Dirichlet model for estimating the purchase probability of a
product category [13]. The approach builds on the observation
that regular purchases in e-commerce websites are a reliable
indicator of customer satisfaction and loyalty. Their model was
shown to predict a user’s average repeat purchase frequency
and dominant product category with satisfactory accuracy.
Many researchers formulate the problem of predicting
when a customer will make the next purchase as a typical clas-
sification task. The procedure of training a predictive model
for this kind of task does not significantly differ from other
classification tasks. Feature engineering, however, is likely to
be more complex for prediction tasks in e-commerce compared
to other classification problems. This is because a large number
of attributes is needed in order to capture customer preferences,
behaviors and interactions. Liu et al. have shown that ensemble
techniques which blend multiple classifiers together result in
superior performance in repeat buyer prediction [14].
Chamberlain et al. presented how ASOS.com, a global
online fashion retailer, distinguishes between loyal and non-
loyal customers to predict Customer Lifetime Value (CLTV)
and enable personalized shopping experiences [15]. The pre-
dictive models used by ASOS are trained with an unsupervised
learning approach on user session data (i.e. sequences of
products viewed by each customer) and afford high accuracy
in predicting repetitive orders.
C. Demand Forecasting
Being able to accurately predict product purchases and
repurchases enables enterprises to produce accurate revenue
forecasts, manage their stock levels, optimize their pricing and
adapt their operations. This is the realm of demand forecasting,
which is another stream of research that can inform the design
of solutions like the one presented in this paper.
Traditional forecasting methods are based on time series
analysis. They are therefore applied under the hypothesis that
past demand can provide a statistical estimation of future
demand. This works well for simple environments but fails
to work in complex domains where demand is affected by
multiple parameters. Other kinds of forecasting methods such
as causal modeling can tackle complex domains beyond the
limits of time series models. Machine learning methods could
actually be viewed as approaches to causal modeling since
they can incorporate time series, categorical variables, fuzzy
variables, text analysis, etc. [16].
Ferreira et al. presented the way that an online retailer,
Rue La La, is using transactional data and machine learning
to enable demand forecasting and optimize pricing decisions
on a daily basis [17]. The researchers used machine learning
techniques on top of historically lost sales to predict the
future demand of new products, and developed an algorithm
to dynamically suggest optimized prices.
D. Email Campaign Optimization
One last stream of research related to the work presented
in this paper is marketing campaign optimization - and more
specifically, optimization of email campaigns. Several studies
over the past decade have addressed the question of computing
the optimal time to send out an email campaign [18, 19]. For
instance, Paraliˇ
c et al. have recently proposed a method for
email campaign planning that is optimized to increase open
rates, taking into account the marketing campaign content
and the type of customer [20]. The researchers applied a
combination of classification models which group together
similar customers and specify the optimal time to send the
campaign message to each group, resulting in significantly
increased open rates (i.e. percentage of customers opening the
messages).
www.ijacsa.thesai.org 91 |Page
(IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 11, No. 4, 2020
III. RESEARCH SCO PE A ND GOA LS
A. Project PRIME
The research results presented in this paper are part of
an ongoing R&D project carried out at Apifon, a Euro-
pean company providing marketing communication technol-
ogy and telecommunication services to the global market
(www.apifon.com).
The project is titled PRIME (Predictive Personalization of
Conversational Customer Communications with Data Protec-
tion by Design) and focuses on the development of a new
services platform which enables personalization of business-
to-consumer mobile communications through the use of pre-
dictive analytics and machine learning technologies. The mo-
tivation is to make communication: (a) more personalized
and relevant to each customer’s preferences; (b) more direct,
interactive and content-rich; and (c) safer for both sides,
by protecting consumers’ personal data and ensuring GDPR
compliance for businesses.
B. PRIME Platform Capabilities
The capabilities offered by the PRIME platform are the
following:
Date and Time Optimization: The goal of any business
that promotes its products through direct marketing
campaigns is to find the best time to send out the
campaign content to its subscribers, so as to achieve
the highest possible conversion rate. The same is
true for the special case of messages which are not
part of a mass campaign but are addressed to indi-
vidual consumers as reminders for them to reorder
a consumable or disposable product that they had
purchased in the past. The PRIME platform offers
services to automatically determine the optimal date
range, day of the week and time of day to send
out a specific marketing message to each customer
individually, based on his/her unique profile.
Segment Recommendation: A great challenge faced by
many marketing professionals during their daily work
has to do with how to choose the recipient list for
a particular campaign message. Marketers often lack
the tools to be able to specify a highly relevant target
audience for a campaign. Messages are often sent
out to large lists of non-relevant customers, resulting
in low engagement rates and poor customer experi-
ence. Through automated segmentation, the PRIME
platform enables marketers to reach those subscribers
who are most likely to find the content of a specific
campaign relevant and compelling enough to engage.
Keyword Recommendation: The topic of a market-
ing communication activity (concepts, terms) and the
written language used (style, expression) are essen-
tial for personalization. The PRIME platform offers
personalized content enrichment for the message of
an upcoming campaign through automated keyword
recommendations.
Click-Through Rate Estimation: One of the challenges
that marketers face in their everyday work is how to
estimate the ROI (Return on Investment) of a mar-
keting communication activity at the planning stage.
Marketers often have little information to help them
estimate how effective a campaign can be before
executing it. The PRIME platform aims to produce a
reliable estimation of a messaging campaign’s Click-
Through Rate before it is actually sent [8]. This is tied
to predicting who of the recipients will successfully
receive the message and will engage with the content
(e.g., by opening a link in the message and following
a call-to-action).
Risk Factor Estimation: Another issue marketers face
is knowing when a customer is highly likely to cease
purchasing. Estimating the risk of losing a customer is
about computing a reliable indicator of the client’s risk
of leaving at any given time. The PRIME platform can
dynamically organize customers into groups, based
on their risk factor. This enables marketers to take
preventive measures, like sending out special offers.
C. Focus and Contributions of this Research
The solution approach and evaluation results that we dis-
cuss in this paper relate to the first category of the platform
capabilities which we outlined above: Date and Time Opti-
mization.
We present a solution to the problem of scheduling the
delivery of consent-based personalized marketing messages to
individuals who are existing customers of a retail enterprise.
Specifically, we focus on the case of marketing communica-
tions which are addressed to repeat buyers of consumer pack-
aged goods (CPG) products which require routine replacement
or replenishment, such as food & beverages, cosmetics and
household consumables and disposables. The objective is to
improve the timing of messages relative to the subscribers’
predicted needs, leading to an improved customer experience
and a higher rate of follow-on purchases.
The problem is tackled by training a regression model
that is able to accurately predict the number of days between
the last product purchase made by a particular customer and
their next purchase. Two types of data are being used to train
the model: (i) historical data describing purchase transaction
details (e.g. purchased products, transaction dates and value),
and (ii) historical data describing the customer’s engagement
with mobile message communications received from the retail
enterprise in the past (message delivery dates, message content,
delivery status, open/click actions, engagement frequency and
dates, etc.).
Combining these two types of data allows us to build a
model that can accurately determine the most likely date that a
specific product would need to be repurchased by an individual
customer. The date predicted by the model is subsequently pro-
cessed by a domain-specific personalization algorithm which
fine tunes the timing to the optimal day of the week and
optimal time of the day to actually deliver the message. This
finally allows the PRIME platform to automatically schedule
the delivery of the repurchase reminder messages for each
consumer.
We evaluated this solution using online (live) testing with
actual customers of a European retail brand in the market of
www.ijacsa.thesai.org 92 |Page
(IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 11, No. 4, 2020
baby products and collected data which suggests a significant
improvement in marketing effectiveness, compared to the
standard method of scheduling product repurchase reminders
- which uses fixed time intervals determined by the last date
of a product’s purchase. As discussed later in the paper, our
personalized scheduling approach resulted in a significantly
higher rate of both customer interaction and conversion from
reminder message to repurchase transaction.
IV. DATA PROCESSING
A. Collection of Message Exchange and Engagement Data
Messaging service providers, like Apifon, function as inter-
mediaries in the communication between different companies
and their customers. As such, they are in position to collect and
analyze the data generated during this messaging exchange.
This may include events regarding if and when a message was
delivered to the customer’s mobile device, if, when and how
many times a message was opened, and whether or not the
customer followed hyperlinks included in the message.
Provided that the consumer has given consent, the messag-
ing services provider is in position to collect a host of histor-
ical data describing the consumer’s engagement with mobile
message communications received from the retail enterprise,
throughout the lifetime of the customer relationship.
B. Integration of External Transactional Data
Message exchange and engagement data are one of the
two types of data being utilized in our approach. The second
type of data is purchase transaction details. This data needs
to be sourced from the information systems used by the retail
enterprise.
For this purpose, we developed a secure data exchange
infrastructure to facilitate data collection. We used a message
broker on top of the RabbitMQ open-source software [21],
utilizing the Advanced Message Queuing Protocol (AMQP).
This protocol provides a platform for both sending and re-
ceiving messages which remain secure until they reach their
destination [22]. Table I provides an example of a potential
purchase transaction entry.
C. Compliance with Data Privacy Policies
Compliance with the General Data Protection Regulation
(GDPR) is a fundamental requirement in the design of the
PRIME platform. Any type of data processing taking place in
the platform is only possible with the data subject’s consent.
The way that this consent is being obtained depends on the
data privacy and marketing communication policies of the
retail enterprise that maintains the direct relationship with the
customer (i.e. the data owner).
Inside the platform, there is a series of data management
services that are built to support GDPR-specific requirements,
such as ensuring the data subject’s right to access and right to
be forgotten. These terms describe the right of the customer
to receive a copy of all the personal information stored by the
platform, as well as to remove this information permanently.
Additional measures that ensure GDPR compliance of
the solution presented in this paper include data encryption
(for instance, hashed phone numbers) and data segregation.
Segregation is an architecture design property which ensures
that no single data processing node can have read access to the
full set of data relating to any individual. Customer data which
includes personally identifying information are kept separate
from the computed user profile data generated by the predictive
model and are maintained only for the time period required for
the necessary calculations [23, 24].
D. Data Parsing and Feature Engineering
Once the data are imported into the PRIME platform
they are taken through a pre-processing pipeline to extract
and transform primary features into a format suitable for the
predictive models, but also to derive new, secondary features
based on a set of business domain rules.
The features used by the predictive model presented in this
paper are shown in Table II. These features are acquired either
by performing simple calculations on the incoming data, or
by applying more complex data transformations. The “Value
Cluster” and “Gender” features are an exception because
they require two additional computation processes. The values
for the Value Cluster feature are derived with a “K-Means”
unsupervised learning algorithm based on the “Sum Order
Value” feature of each customer (Table II). The “K-Means”
algorithm determines the optimal number of clusters to be used
using a Silhouette score [25]. The Gender feature values are
derived with a special classification model which has been
previously trained with a dataset of male and female names
labelled by gender.
V. DATE AND TIME OPTIMIZATION
A. Classification of Regular / Repeat Customer
The first step in the execution of our optimized message
scheduling solution is answering the question of whether
someone is a regular customer or not. The answer determines
if automated date and time optimization will be attempted. In
the case of a new or irregular customer, a different execution
path could be followed. Newly acquired customers for whom
there is little data available do not present the same opportunity
for optimization that repeat and regular customers provide.
The answer to the question of what makes a repeat and
regular customer is very much dependent on the business
domain. The definition of regular customer for a food retail
business would probably be quite different from that of a
cosmetics retailer. Our pilot test bed was a European retail
brand in the market of consumable and disposable baby prod-
ucts. Anonymized customer purchase transaction data were
sourced from the information systems used by the company
and analyzed, focusing on a specific product.
The definition of repeat and regular customer that we
eventually arrived at, for the business domain of this particular
pilot company, is customers who: (i) made at least 2 purchases
of a specific product in the last 2 months, and (ii) also had
less than 20 days of variance between their purchases of that
product, ever since the first time they bought the product.
To arrive at this definition we tested the accuracy of the
regression model that will be introduced in the next section
against six sets of customers of the pilot company who were
www.ijacsa.thesai.org 93 |Page
(IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 11, No. 4, 2020
TABLE I. TYPICAL INCOMING TRANSACTION ENTRY EXAMPLE.
Key Explanation Type
Account Id Unique identifier of the company (retail enterprise) Numerical
Transaction Date Date of the transaction Date
Customer Code Unique identifier of the customer who made the transaction Numerical
Transaction Code Unique identifier of the transaction Numerical
Order Source How the customer made the transaction Categorical
Products List of products purchased in the transaction List of Categorical
Total Price Total amount of money paid for the transaction List of Categorical
Payment Method How the customer paid for the transaction Categorical
Order Status Status of the payment Categorical
TABLE II. FEATURE S USE D TO F EED T HE PR ED ICT IV E MOD EL .
Feature Explanation Type
Last Order Date of the most recent transaction of each customer Date
Minimum Consumption Duration Minimum no. of days between two consecutive transactions per product unit Numerical
Average Consumption Duration Average no. of days between two consecutive transactions per product unit Numerical
Maximum Consumption Duration Maximum no. of days between two consecutive transactions per product unit Numerical
Average Fluctuation Standard deviation in no. of days between two consecutive transactions per product unit Numerical
Days from First Days lapsed since the first purchase of each customer Numerical
Days from Last Days lapsed since from the last purchase of each customer Numerical
Last Product Amount Number of products included in the last purchase per subscription Numerical
Transaction Day Preferred purchasing day of week per customer Categorical
Transaction Hour Preferred purchasing time of day per customer Categorical
Mean Order Value Average money spent per purchase per customer Numerical
Sum Order Value Total money spent per customer to date Numerical
Value Cluster Customer’s cluster assignment based on total money spent to date Categorical
Gender Customer’s gender prediction based on gender classification model Categorical
segmented with respect to different combinations of values for:
(i) total number of product purchases done by the customer,
(ii) number of recent repurchases, and (iii) consistency in
the time span recorded between the customer’s repurchases.
The values for these three metrics can be seen in Table
III under “Transactions”, “Recent purchases” and “Average
Fluctuation”, respectively.
TABLE III. PREDICTIVE PERFORMANCE ANALYSIS WITH ALTERNATIVE
DE FINI TI ONS O F REP EAT/REGULAR CUSTOMER.
Transactions Recent Purchases Average Fluctuation MAE R2
2 2 <15 4.48 0.37
2 2 <20 4.41 0.45
3 2 <15 5.16 0.39
3 2 <20 4.88 0.43
3 3 <15 5.05 0.40
3 3 <20 4.91 0.42
We measured the predictive accuracy of the regression
model on the aforementioned customer data set in terms of
the MAE and the R2score [26]:
Mean Absolute Error (MAE): The average of the
absolute errors (1). The MAE units are the same as
the predicted target, which is useful for understanding
whether the size of the error is of concern or not being
robust to outliers. The smaller the MAE is, the better
the algorithm’s performance [27]. In (1), N is the
total number of errors and |xix|equals the absolute
errors.
MAE =1
N
N
X
i=1
|xix|(1)
R2Score: A statistical measure that represents the pro-
portion of the variance for a dependent variable that’s
explained by an independent variable in a regression
model. The R2value varies between 0 and 1, where 0
represents no correlation between the predicted and
actual value and 1 represents complete correlation
[28].
As displayed in Table III, there was one particular set of
customers for whom the regression model produced predic-
tions with the smallest MAE and simultaneously the highest
R2score. This was the set of customers who had made at least
2 purchases of a specific product in the 2 prior months and
who had less than 20 days of variance between their purchases
of that product.
B. Prediction of Repurchase Date using Regression
The next step in the solution is estimating the future point
in time when a specific repeat/regular customer will need to
repurchase a consumable or disposable product that they have
been using in the past.
This estimation problem can be mathematically formulated
as a regression problem of trying to predict a continuous value,
i.e. the number of days from the customer’s previous purchase
to the next one. The target feature that our regression model
is expected to produce is “Days to Next Purchase”.
www.ijacsa.thesai.org 94 |Page
(IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 11, No. 4, 2020
In search of the best approach to develop our regression
model we tested a variety of machine learning algorithms
including Logistic Regression [29], Random Forests [30], K
Nearest Neighbors [31], CatBoost [32], LightGBM [33]. The
algorithm that we found most balanced in terms of accuracy
and processing performance for our goals was the XGBoost
regression algorithm [34].
XGBoost belongs to the ensemble learning methods which
are based on the need to rely upon the results of more than
one machine learning models giving their aggregated output.
Bagging and boosting are two widely used ensemble learner
models. These two techniques can be used with several statisti-
cal models and the most predominant usage was achieved with
decision trees [35]. XGBoost is a scalable machine learning
algorithm that is based on boosting approach and incorporates
several useful features, such as parallel processing, high flexi-
bility, automated handling of incomplete values, inherent tree
pruning and built-in cross-validation [36].
A pre-processing phase comes before the training of the
main predictive model. This phase starts with the elimina-
tion of undue data features that appear to have a negative
impact to the algorithm’s predictive accuracy. After that a
data transformation process takes place in order to convert
all the categorical features into numerical representations.
All the available data are then subdivided into training and
testing subsets, following a fixed distribution of 80% and 20%,
respectively. The algorithm is then ready to get trained using
the training data and having multiple variable parameters,
which have to be adjusted in the next step. The trained model
is eventually able to make predictions on the “Days to Next”
feature, which is then getting rounded in order to express
number of days.
Configuring the XGBoost regression algorithm has to do
not only with the parameters inside the algorithm itself, but
also with the number of folds to be used during the cross-
validation process which is responsible for the main tuning.
Cross-validation is generally used to assess a machine learning
model’s ability to meet first-time data using a limited sample
each time (i.e. data not used during the training phase). The
number of different restricted samples is determined by the
number of folds [37].
The described procedures utilize the so-called Grid-
SearchCV function provided by the scikit-learn Python library
[38]. It is possible to set the number of kernels to be used
by the algorithm to the maximum permissible number by the
system. In our configuration, we set the number of folds to
5 and selected the algorithm evaluation metric to be the R2
score. The percentage shrinkage rate used in each update of the
algorithm to avoid overfitting (i.e. the inability of the algorithm
to adapt to new data despite its good performance in the test
data) was set to alternate between the values of 0.01, 0.02,
0.05, 0.1, and 0.3. The maximum number of features to be
used by the algorithm was declared to alternate between 3, 4,
and 7, while the maximum depth of node representation of the
algorithm tested up to levels 3, 4, and 5. XGBoost was let to
randomly test 50%, 60%, and 70% of the training data before
developing nodes, preventing the overfitting phenomenon. We
also limited the number of repetitions during the training of
the algorithm to not exceed 20 in order to prevent overfitting
(Table IV).
C. Fine Tuning of the Message Delivery Date and Time with
a Domain-Specific Personalization Algorithm
The last step in the execution of the solution is to run
a domain-specific personalization algorithm. The algorithm
accepts as input the repurchase date that was predicted by the
regression model and applies domain-specific business rules
to extract the optimal day of the week and time of the day to
deliver a reminder message to the customer.
Note that the actual date and time that the reminder
message should be delivered to the customer is not the same
as the repurchase date produced by the regression model. The
latter date is when the transaction is predicted to take place,
whereas the message needs to be come earlier and serve as a
timely reminder.
The day and time that the reminder message is sent to each
customer should be chosen based on their individual profile
and optimized to maximize conversion rate. Looking at past
purchase transaction data it is easy to determine if some day
of the week had a higher number of transactions associated
with it, which could potentially suggest that making a purchase
is more convenient for that customer on that specific day.
Conversely, looking at past message exchange & engagement
analytics data it is easy to determine if the customer is more
likely to open, read or engage with a mobile message during
a specific period within the day.
The algorithm analyzes the distribution of the transactions
of a customer on a daily basis throughout the week, and then
adds a weight to each day in order to reveal the one with
the maximum weight, i.e. the dominant day. To calculate the
dominant day, it is important to distinguish between the case
where there is only one day with the maximum weight and
the case where the customer has performed an equal number
of purchases in more than one days during a week.
In the first case, the candidate dominant days are the ones
having at least as many transactions as the average transactions
of the whole week. In the second case, the candidate dominant
days are the ones having the maximum number of transactions.
For each one of those days, the weight is given by formula (2),
where the days of week are numbered from 0 to 6 representing
the days from Monday to Sunday:
W eight =|#of daily transactions
|day of week predicted day of week| | (2)
After calculating the weights, the algorithm selects the day
with the maximum score (i.e. the customer’s most active day)
which is as close as possible to the day predicted by the
regression model. It also checks whether the predicted day
is the same as the day with the maximum weight (i.e. if the
adapted day is the same as the one originally predicted as
repurchase date). If this is not the case, the algorithm shifts the
predicted day of week to the closest day having the maximum
weight. At the end of the process, there is a personalized
product reminder scheduled for each customer, which is based
on their personal purchase and interaction preferences (i.e.
Adapted Predicted Days” feature).
Table V shows an example of applying the algorithm to
transaction history data to get the distribution of the customer’s
www.ijacsa.thesai.org 95 |Page
(IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 11, No. 4, 2020
TABLE IV. SELEC TED XGBO OS T HYPERPARAMETER TUNING VALUES.
Parameter Description Value
Learning rate Step size shrinkage used to prevent overfitting. 0.01
Max depth How deeply each tree is allowed to grow during any boosting round. 3
Max features The number of features to consider when looking for the best split. 10
Min child weight Minimum sum of instance weight (hessian) needed in a child. 3
Subsample Percentage of samples used per tree. 0.6
N jobs The number of jobs to run in parallel for both fit and predict. -1
Evaluation metric Metric function to evaluate the predictions on the test set. MAE
Early stopping rounds Finishes training of the model early if the hold-out metric does not improve for a given number of rounds. 20
purchases over the days of the week. As shown in the table,
the customer’s most active day of week is Wednesday, with a
total of 8 transactions.
TABLE V. EXAMPLE DISTRIBUTION OF A CUSTOMERS P URC HA SES .
Day Purchases
Monday 3
Tuesday 6
Wednesday 8
Thursday 0
Friday 1
Saturday 1
Sunday 0
It so happens that there is a unique maximum value in
the distribution, so the algorithm calculates the weights of
days that are above the average transactions (i.e. in this case
the average of Monday, Tuesday and Wednesday is equal to
2.7). Based on the respective aforementioned formula, their
respective weights will be 3, 1 and 4. In this particular
customer instance the predicted repurchase day was Sunday
but the algorithm eventually set the message delivery date to
Wednesday.
Our algorithm aims to move the message delivery date
earlier than the customer’s most active (dominant) day of the
week, rather than just close to it. The algorithm checks whether
the difference between the predicted repurchase date and the
dominant day is less than 5 days and then shifts the message
delivery to the immediately preceding dominant day. If the
difference is at least 5 days, the algorithm shifts the message
delivery date to the nearest dominant day, as previously. If the
personalized message delivery date is in the past, the algorithm
finds the same dominant day of week in the next week, and
therefore moves the reminder day accordingly (i.e. producing
a “Days to Remind” feature).
The optimal time to send a personalized message during
the day is generated as a feature during the pre-processing
phase (i.e. “Transaction Hour” feature in Table II). We also
paid attention to enabling configuration of ‘Do not Disturb’
hours, i.e. time periods when sending message notifications
to customers should not be allowed. The algorithm moves all
recommendations between 9 p.m. and 3 a.m. to 9 p.m., and
all recommendations between 3 a.m. and 9 a.m. to 9 a.m.
The combination of message delivery date and time which
are determined by the algorithm produce the exact date and
time that the personalized reminder will be scheduled.
D. Deployment Infrastructure
Regarding the deployment infrastructure, the solution has
been implemented on IBM Cloud as a scheduled Jupyter
Notebook with an Apache Spark configuration (i.e. on a server
cluster) using the capabilities of the IBM Watson Studio suite
[39, 40]. We used a regularly executing Python notebook that
acts like an updater to the respective database which contains
the mapping of date and time to send the next reminder
message and unique encrypted customer code. The deployment
model is depicted in Fig. 1.
Fig. 1. Overview of the adaptive model deployment architecture.
VI. EVALUATIO N
A. Pilot Test Context
As highlighted in the sections above, the research presented
in this paper addresses the problem of scheduling the delivery
of marketing messages to repeat buyers of CPG products
which require routine replenishment or replacement - such
as food & beverages, cosmetics or household products. The
objective is to increase both the customer engagement and
the conversion rate. An indicator of customer engagement
is the rate of messages that are read by the recipients (i.e.
seen rate), as well as the percentage of customers who click
on the hyperlink found inside the message text (i.e. click-
through rate). The ultimate goal is to optimize the date and
time that each individual consumer will receive a message that
prompts them to order the product again, such that the rate of
conversion from reminder message to product repurchase can
be increased.
To evaluate the effectiveness of our approach we conducted
a pilot test with one of Apifon’s clients, a European CPG
www.ijacsa.thesai.org 96 |Page
(IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 11, No. 4, 2020
brand in the market of baby products. This is a company
that sells direct to consumers. Its revenues and profitability
depend on its customers placing regular product reorders. The
company has been using mobile text messages to regularly
remind its customers to reorder consumable and disposable
products that they purchased in the past, following a fixed
scheduling approach. Whenever a customer orders a product
of a specific type, a reminder message is scheduled to be sent
ndays later (the preset number of days varies per product
type). The message is scheduled for delivery provided that the
customer has given explicit consent to receive such reminders.
B. Experiment Design
To evaluate our proposed method, we analyzed and com-
pared its effectiveness against the company’s standard practice
of scheduling reminders at fixed time intervals. The evaluation
metrics that were used to compare the two different methods
are the message seen and click-through rates and the customer
repurchase rate that are achieved by each approach.
Our system was deployed live for a period of one week
during September 2019 and managed the message scheduling
process for an experiment cohort of 40 repeat customers who
met the following criteria: (i) they made at least 2 purchases
of a specific product in the last 2 months, and (ii) they also
had less than 20 days of variance between their purchases of
that product, ever since the first time they bought the product.
We measured the message seen and click-through rates, as
well as the conversion rate for this cohort of 40 customers after
1, 2, 12 and 24 hours since the time that the reminder message
was delivered to each of them, to determine the percentage of
customers who received a reminder and actually went ahead
and reordered the product.
As a next step we collected historical data from September
2018 (one year earlier) and created a benchmark cohort of 40
repeat customers who met the same criteria as the experiment
cohort (i.e. identical number of recent product purchases and
same consistency in the time span recorded between repur-
chases). Following, we measured the message seen and click-
through rates, along with the conversion rate for these 40
customers after 1, 2, 12 and 24 hours since the time they
had received the purchase reminder message (under a fixed
scheduling approach).
This setup allows us to compare the impact of the schedul-
ing method on each cohort: (i) the benchmark cohort consisting
of 40 customers, who were sent product repurchase reminders
using the company’s standard method of fixed time intervals,
and (ii) the experiment cohort consisting of 40 customers, who
were sent reminders using the optimized scheduling method.
The hypothesis was that customers in the optimized
scheduling group would not only interact with the marketing
message they received (as measured by seen and click-through
events), but also would repurchase products at a rate higher
than that of customers in the other group, at a level of statistical
significance not lower than the conventionally accepted 95%
(p-value 0.05).
VII. EVALUATI ON RE SU LTS
The results of the pilot test are displayed in Tables VI,
VII and VIII. They suggest that the new method of timing the
reminder messages using the approach proposed in this paper
outperforms the approach of scheduling messages at fixed time
intervals. This is evident by both the message seen and click-
through rates, and the customer repurchase rates within 1, 2,
12 and 24 hours from sending the message.
The message seen rate and click-through rate of the ex-
periment cohort surpassed those of the benchmark cohort
in all time span measurements. The results are statistically
significant, with confidence levels over 95% across cases, as
can be seen from the p-values in Tables VI and VII.
Similarly, the repurchase rate for the experiment cohort
was higher than that of the benchmark cohort across all mea-
surements (Table VIII). The measured increase in repurchase
rate recorded after 1, 2 and 24 hours of sending the message
has a confidence level over 95%. This is not the case for
repurchases at the 12 hour time span. Fig. 2 illustrates how the
repurchase rates of the experiment cohort (i.e. date and time
optimized reminders) evolve over time compared to those of
the benchmark cohort (i.e. the standard method).
VIII. CONCLUSION
Modern marketing automation technology is evolving to
allow enterprises to achieve personalized communications with
their customers at scale - delivering the right message at the
optimal date and time to each individual subscriber. In the
past five years we have seen an increasing number of research
studies on the topic of email marketing personalization and
data-driven models for repeat purchase prediction. Some of
the most interesting work comes from research teams inside
online retailers like Amazon.com and ASOS.com, who have
direct access to large scale data.
Nevertheless, we have found limited existing research
focusing on the following topics: (i) personalization of mar-
keting communications via instant messaging platforms; (ii)
personalization of marketing communications for repeat buyers
of consumer packaged goods (CPG); (iii) personalization of
instant messaging marketing with data protection by design.
The research discussed in this paper focuses on the inter-
section of the aforementioned areas. Specifically, this paper
presents a solution to the problem of scheduling the delivery
of consent-based, personalized instant-messaging reminders to
repeat buyers of frequently replenished CPG products.
A notable contribution of the personalization approach
presented in this paper is the use of predictive models trained
on two distinct types of data: historical purchase transaction
data augmented with instant messaging engagement data. The
system is able to predict the next product repurchase date for
each customer, which is then taken through a domain-specific
personalization algorithm to determine the optimal day of the
week and hour of the day when a repurchase reminder message
should be sent to that customer.
The aim is to improve the timing of the repurchase rec-
ommendations sent to a subscriber such that they better reflect
their needs and preferences, leading to a better customer ex-
perience and resulting in a higher rate of follow-on purchases.
The evaluation of our solution took place using online (live)
testing with actual customers of a European retail brand in
www.ijacsa.thesai.org 97 |Page
(IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 11, No. 4, 2020
TABLE VI. COMPARATI VE M ESS AGE S EE N RATES F ROM T HE P ILOT T ES T.
Time lapsed since message sent Benchmark cohort seen rates Experiment cohort seen rates Seen rate diff p-value
1 hour 10/40 (25%) 17/40 (42.5%) +17.5% 0.0461
2 hours 13/40 (32.5%) 21/40 (52.5%) +20% 0.0323
12 hours 17/40 (42.5%) 25/40 (62.5%) +20% 0.0338
24 hours 21/40 (52.5%) 29/40 (72.5%) +20% 0.0295
TABLE VII. COMPARATI VE M ESS AGE C LI CK-T HRO UGH R ATES F ROM T HE PI LOT T ES T.
Time lapsed since message sent Benchmark cohort click-through rates Experiment cohort click-through rates Click-through rate diff p-value
1 hour 0/40 (0%) 6/40 (15%) +15% 0.0039
2 hours 1/40 (2.5%) 6/40 (15%) +12.5% 0.0212
12 hours 4/40 (10%) 10/40 (25%) +15% 0.0359
24 hours 5/40 (12.5%) 12/40 (30%) +17.5% 0.0251
TABLE VIII. COMPARATI VE C UST OM ER CO NVE RS ION R ATES F ROM T HE PI LOT T ES T.
Time lapsed since message sent Benchmark cohort repurchases Experiment cohort repurchases Repurchase rate diff p-value
1 hour 0/40 (0%) 3/40 (7.5%) +7.5% 0.0359
2 hours 0/40 (0%) 3/40 (7.5%) +7.5% 0.0359
12 hours 2/40 (5%) 5/40 (12.5%) +7.5% 0.1155
24 hours 2/40 (5%) 7/40 (17.5%) +12.5% 0.0356
the market of baby products. The results suggest a significant
improvement in marketing effectiveness, compared to the
standard method of scheduling product repurchase reminders
at fixed time intervals. Our personalized scheduling approach
resulted in a significantly higher rate of customer interactions
and conversions within 1, 2, 12 and 24 hours from sending
the personalized reminder. There was a notable increase in
the ratio of customers who engaged with the message after
receiving it (up to 17.5% higher), as well as an increase in the
ratio of customers who completed a repurchase transaction (up
to 12.5% higher). The results of the pilot test are encouraging
although the experiment is limited in two ways. Firstly, the
total sample size is not very large. Secondly, we used a bench-
mark comparison approach rather than a controlled experiment
setup. It is understood that these factors could potentially be
contributors to sampling error, however the research team has
tried to mitigate this by careful sample design and using data
from the same period of the year to neutralize the seasonality
effect. As part of future work we plan to design a larger-
scale experiment and generate additional insights into the
effectiveness of our proposed method.
ACKNOWLEDGMENT
This research has been co-financed by the European Re-
gional Development Fund of the European Union and Greek
national funds through the Operational Program Competitive-
ness, Entrepreneurship and Innovation under the project code
T1EDK-04550.
REFERENCES
[1] Chen, P. and Hsieh, H. (2012). Personalized mobile advertising: Its
key attributes, trends, and social impact. Technological Forecasting and
Social Change, 79(3), pp.543-557.
[2] Koch, O. and Benlian, A. (2015). Promotional Tactics for Online Viral
Marketing Campaigns: How Scarcity and Personalization Affect Seed
Stage Referrals. Journal of Interactive Marketing, 32, pp.37-52.
[3] Heimbach, I., Kostyra, D. and Hinz, O. (2015). Marketing Automation.
Business & Information Systems Engineering, 57(2), pp.129-133.
[4] Awadzi, D., Calloway, D. and Awadzi, D. (2018). Increasing Conversions
through Behavioral Retargeting. International Academic Journal of Social
Sciences, 05(01), pp.23-27.
[5] Deligiannis, A., Argyriou, C. & Kourtesis, D. (2019). Predictive per-
sonalization of conversational customer communications with data pro-
tection by design. IEEE/WIC/ACM International Conference on Web
Intelligence on - WI ’19 Companion.
[6] Dangi, H., & Malik, A. (2017). Personalisation in marketing: an ex-
ploratory study. International Journal Of Internet Marketing And Adver-
tising, 11(2), 124. doi: 10.1504/ijima.2017.084079
[7] Abakouy, R., En-naimi, E., Haddadi, A., & Lotfi, E. (2019). Data-driven
marketing. Proceedings Of The 4Th International Conference On Smart
City Applications - SCA ’19. doi: 10.1145/3368756.3369024
[8] Deligiannis, A., Argyriou, C., & Kourtesis, D. (2020). Building a Cloud-
based Regression Model to Predict Click-through Rate in Business Mes-
saging Campaigns. International Journal Of Modeling And Optimization,
10(1), 26-31. doi: 10.7763/IJMO.2020.V10.742
[9] Ding, Y., & Li, X. (2005). Time weight collaborative filtering. In
Proceedings of the 14th ACM international conference on Information
and knowledge management (pp. 485-492).
[10] Koren, Y. (2009). Collaborative filtering with temporal dynamics. In
Proceedings of the 15th ACM SIGKDD international conference on
Knowledge discovery and data mining (pp. 447-456).
[11] Baltrunas, L., & Amatriain, X. (2009). Towards time-dependant recom-
mendation based on implicit feedback. In Workshop on context-aware
recommender systems (CARS’09) (pp. 25-30).
[12] Wang, J. and Zhang, Y. (2013). Opportunity model for e-commerce
recommendation: Right product; right time. Proceedings of the 36th
international ACM SIGIR conference on Research and development in
information retrieval - SIGIR ’13.
[13] Dey, S., Mitra, P. and Gupta, K. (2016). Recommending Repeat Pur-
www.ijacsa.thesai.org 98 |Page
(IJACSA) International Journal of Advanced Computer Science and Applications,
Vol. 11, No. 4, 2020
Fig. 2. Comparing the repurchase rates of benchmark cohort and experiment cohort at 1, 2, 12 and 24 hour intervals.
chases using Product Segment Statistics. Proceedings of the 10th ACM
Conference on Recommender Systems - RecSys ’16.
[14] Liu, G., Nguyen, T., Zhao, G., Zha, W., Yang, J., Cao, J., Wu, M.,
Zhao, P. and Chen, W. (2016). Repeat Buyer Prediction for E-Commerce.
Proceedings of the 22nd ACM SIGKDD International Conference on
Knowledge Discovery and Data Mining - KDD ’16.
[15] Chamberlain, B., Cardoso, ˆ
A., Liu, C., Pagliari, R. and Deisenroth,
M. (2017). Customer Lifetime Value Prediction Using Embeddings.
Proceedings of the 23rd ACM SIGKDD International Conference on
Knowledge Discovery and Data Mining - KDD ’17.
[16] Cadavid, J. P. U., Lamouri, S., & Grabot, B. (2018). Trends in Machine
Learning Applied to Demand & Sales Forecasting: A Review.
[17] Ferreira, K., Lee, B. and Simchi-Levi, D. (2016). Analytics for an
Online Retailer: Demand Forecasting and Price Optimization. Manufac-
turing & Service Operations Management, 18(1), pp.69-88.
[18] Bonfrer, A., Dr`
eze, X. (2009). Real-Time Evaluation of E-mail Cam-
paign Performance, Marketing Science 28(2), 251-263.
[19] Zhang, X., Kumar, V. and Cosguner, K. (2017). Dynamically Managing
a Profitable Email Marketing Program. Journal of Marketing Research,
54(6), pp.851-866.
[20] Paraliˇ
c, J., Kaszoni, T. and Maˇ
cina, J. (2019). Predicting Suitable Time
for Sending Marketing Emails. Advances in Intelligent Systems and
Computing, pp.189-196.
[21] Naik, N. (2017). Choice of effective messaging protocols for IoT
systems: MQTT, CoAP, AMQP and HTTP. 2017 IEEE International
Systems Engineering Symposium (ISSE).
[22] Dobbelaere, Philippe & Sheykh Esmaili, Kyumars. (2017). Kafka versus
RabbitMQ
[23] George, D., Reutimann, K. and Tam`
o-Larrieux, A. (2018). GDPR
Bypass by Design? Transient Processing of Data Under the GDPR. SSRN
Electronic Journal, pp.6-17.
[24] Lachaud, E. (2019). Adhering to GDPR Codes of Conduct: A Possible
Option for SMEs to GDPR Certification. SSRN Electronic Journal.
[25] Sai, L., Shreya, M., Subudhi, A., Lakshmi, B. and Madhuri, K. (2017).
Optimal K-Means Clustering Method Using Silhouette Coefficient. In-
ternational Journal of Applied Research on Information Technology and
Computing, 8(3), p.335.
[26] Botchkarev A. (2019). A New Typology Design of Performance Metrics
to Measure Errors in Machine Learning Regression Algorithms. Inter-
disciplinary Journal of Information, Knowledge, and Management, 14,
pp.045-076.
[27] Willmott, C. and Matsuura, K. (2005). Advantages of the mean absolute
error (MAE) over the root mean square error (RMSE) in assessing
average model performance. Climate Research, 30, pp.79-82.
[28] Alexander, D., Tropsha, A. and Winkler, D. (2015). Beware of R2:
Simple, Unambiguous Assessment of the Prediction Accuracy of QSAR
and QSPR Models. Journal of Chemical Information and Modeling,
55(7), pp.1316-1322.
[29] Steyerberg, E. W., Harrell Jr, F. E., Borsboom, G. J., Eijkemans, M.
J. C., Vergouwe, Y., & Habbema, J. D. F. (2001). Internal validation of
predictive models: efficiency of some procedures for logistic regression
analysis. Journal of clinical epidemiology, 54(8), 774-781.
[30] Evans, J. S., & Cushman, S. A. (2009). Gradient modeling of conifer
species using random forests. Landscape ecology, 24(5), 673-683.
[31] Batista, G. E., & Monard, M. C. (2002). A Study of K-Nearest
Neighbour as an Imputation Method. HIS, 87(251-260), 48.
[32] Dorogush, A. V., Ershov, V., & Gulin, A. (2018). CatBoost:
gradient boosting with categorical features support. arXiv preprint
arXiv:1810.11363.
[33] Ke, G., Meng, Q., Finley, T., Wang, T., Chen, W., Ma, W., ... & Liu, T.
Y. (2017). Lightgbm: A highly efficient gradient boosting decision tree.
In Advances in Neural Information Processing Systems (pp. 3146-3154).
[34] Chen, T., & Guestrin, C. (2016). Xgboost: A scalable tree boosting
system. In Proceedings of the 22nd acm sigkdd international conference
on knowledge discovery and data mining (pp. 785-794). ACM.
[35] Sumana, B. and Santhanam, T. (2015). Optimizing the Prediction of
Bagging and Boosting. Indian Journal of Science and Technology, 8(35).
[36] Chen, T. and Guestrin, C. (2016). Xgboost: A scalable tree boosting
system. In: Proceedings of the 22nd acm sigkdd international conference
on knowledge discovery and data mining. ACM, pp.785-794.
[37] Zhang, L., & Zhan, C. (2017, May). Machine learning in rock facies
classification: an application of XGBoost. In International Geophysical
Conference, Qingdao, China, 17-20 April 2017 (pp. 1371-1374). Society
of Exploration Geophysicists and Chinese Petroleum Society.
[38] Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B.,
Grisel, O., ... & Vanderplas, J. (2011). Scikit-learn: Machine learning in
Python. Journal of machine learning research, 12(Oct), 2825-2830.
[39] Palmer, T. (2019). Predict and Optimize Business Outcomes with IBM
Decision Optimization for Watson Studio and IBM Cloud Pak for Data.
The Enterprise Strategy Group, Inc., pp.3-14.
[40] Bozhinov, I. (2019). AI and big data on IBM Power Systems servers.
U.S.A.: IBM Corporation.
www.ijacsa.thesai.org 99 |Page
... Deligiannis et al. [26] present models for sending marketing communications to subscribers, optimizing the moment to repurchase a product. The communications are sent by SMS, through a Rabbit MQ protocol [27], which enables sending and receiving mobile messages. ...
... Some models include email-related features (e.g., the number of emails sent and received) as well as the subscriber profile information (e.g., location, purchase transaction, and so on). Deligiannis et al. [26] Deligiannis et al. [24] Singh et al. [40] Paralic et al. [29] Conceição et al. [33] Sinha et al. [37] Singh et al. [41] Luo et al. [35] Piersma et al. [42] ...
Article
Full-text available
In the digital world, the demand for better interactions between subscribers and companies is growing, creating the need for personalized and individualized experiences. With the exponential growth of email usage over the years, broad flows of campaigns are sent and received by subscribers, which reveals itself to be a problem for both companies and subscribers. In this work, subscribers are segmented by their behaviors and profiles, such as (i) open rates, (ii) click-through rates, (iii) frequency, and (iv) period of interactions with the companies. Different regressions are used: (i) Random Forest Regressor, (ii) Multiple Linear Regression, (iii) K-Neighbors Regressor, and (iv) Support Vector Regressor. All these regressions’ results were aggregated into a final prediction achieved by an ensemble approach, which uses averaging and stacking methods. The use of Long Short-Term Memory is also considered in the presented case. The stacking model obtained the best performance, with an R2 score of 0.91 and a Mean Absolute Error of 0.204. This allows us to estimate the week’s days with a half-day error difference. This work presents promising results for subscriber segmentation based on profile information for predicting the best period for email marketing. In the future, subscribers can be segmented using the Recency, Frequency and Monetary value, the Lifetime Value, or Stream Clustering approaches that allow more personalized and tailored experiences for subscribers. The latter tracks segments over time without costly recalculations and handles continuous streams of new observations without the necessity to recompile the entire model.
... Companies have long utilized emails to connect with consumers and create an asynchronous conversation. Email marketing has been widely adopted as it enables to reach a wide audience at a relatively low cost, and to engage consumers [6]. Despite its advantages, developing an email strategy/campaign can be challenging, as many elements should be carefully developed, from the content, the format and the design. ...
... However, this risks in resulting in a vast number of unsolicited emails in an individual's inbox and creating a sense of frustration and unsatisfaction [7]. The issue of information overload in the digital context is not a new phenomenon, hence developing effective and efficient forms of communication is essential [6]. ...
Chapter
The retail sector is experiencing a time of change due to technological advances; also consumer habits are changing. This process of change has been accelerated by the Covid-19 pandemic, which is impacting individuals’ daily lives. Digitalization is not only impacting consumers’ shopping habits but also the communication between brands and consumers. This study focuses on email marketing, in specific fashion brands’ newsletters. The top 50 apparel brands according to the 2020 brand finance are utilized as a sample. It emerges that most brands have a newsletter, showing that it is still an important and widely utilized form of communication. Then, it analyses the messages utilized to drive individuals’ subscription. Three main themes are identified, cognitive, emotional and financial. Moreover, the information collected for the subscription process is registered. This allows to gain an insight into the personal information collected. Subsequently, it conducts a benchmark of the confirmation emails received. In total thirty-four confirmation emails are analysed, both the subject lines and the content. Interestingly, low levels of explicit personalization are found within the analysed messages. Overall, this study shows the relevancy of newsletters and provides suggestions for future research.
... In [1], the authors performed an empirical comparison of supervised machine learning models based on decision tree and logistic regression algorithms in order to improve the open rate and conversion rate of email marketing campaigns. In [34], the authors propose to use transactional features and instant messaging metadata to train a boosting tree regression algorithm to timely anticipate the needs of consumers in order to increase their level of engagement as well as the rate at which they repurchase products. ...
Thesis
Full-text available
The overall objective of this PhD is to explore and propose new approaches leveraging a large volume of heterogeneous data that needs to be integrated and semantically enriched, and recent advances in machine and deep learning techniques, in order to exploit both the increased variety of offers that an airline can make to its customers as well as the knowledge it has of its customers with the ultimate goal of optimizing conversion and purchase. The overall goal of this thesis can be broken down into three main research questions: 1) What piece of content (ancillary services, third-party content) should be recommended and personalized to each traveler? 2) When should a recommendation be made and for which communication channel to optimize conversion? 3) How do we group ancillary services and third-party content and can we learn what often goes together based on purchase logs?
... Likewise, the objective of the emailing technique is to gather leads or potential customers to create databases of contacts, and to send them customized commercial campaigns or newsletters. These techniques, despite being extensively exploited, are still highly effective when following good practices [16,17]. ...
Article
Full-text available
The present study aims to provide an overview of the digital marketing strategies in social network services (SNSs) used in America and Europe. This work, unlike previous research that analyze digital marketing strategies in SNSs in a single geographic area, presents as the main contribution and value the use of a comparative approach between continents; specifically, then, an approach that allows comparing the management strategies used in these two areas. The research takes a sample of forty organizations, twenty American and twenty European. The author applies a descriptive examination followed by a parametric analysis using the t-test procedure over a total of 158,208 publications on Twitter. The findings show significant differences in the management approaches applied in these two continents. Even though the digital marketing strategies in SNSs should be based on dialogue and interaction with the organization’s target audience, this point can be more or less important, depending on the continent. While European organizations confer great importance to the interaction with its audience, American organizations tolerate a greater degree of one-way communication. The author concludes that the paradigms governing the definition of digital marketing strategies in SNSs at the global level, a priori thought to be universal, probably require reformulation if they are to be well adapted to the specific geographic areas where these strategies are implemented.
... The goal of any business that promotes its own products through direct marketing campaigns is to find the best time to send out the campaign content, so as to achieve the highest possible conversion rate. A corresponding use case that the PRIME platform supports, is to automatically determine the right date and time to send out a specific direct marketing campaign message to each customer individually, based on his/her unique profile [37].  Segment Recommendation: A great challenge faced by many marketing professionals in their daily work has to do with how to choose the recipient list for a particular campaign message. ...
Article
Full-text available
One of the main goals of customer relationship management is to reduce or eliminate “customer churn”, ie loss of existing customers. This paper introduces a prototype algorithm to estimate a continuously updated indicator of the probability of an existing customer to cease purchasing from a subscription commerce business. The investigation is focused on the case of repeat consumers of subscription commerce products which require regular replacement or replenishment. The motivation is to help marketers to make targeted proactive retention actions by categorizing regular customers into groups of similar estimated churn risk. The proposed algorithm re-computes the probability of churn for each customer at regular intervals using past purchase transaction data and incorporating subscription-based business logic. We describe the detailed process from data collection and feature engineering to the algorithm’s design. We also present evaluation results of the algorithm’s performance based on a pilot test that took place on a consumables ecommerce business. The results suggest a significant capability of the proposed algorithm in capturing the purchasing intentions of repeat customers, regardless of the risk group they belong to.
Article
Full-text available
Circular Economy (CE) has been one of the most transformational tendencies for the past years. What seemed to be one more organizational hype, is now appearing as a global trend, affecting macro, meso and microenvironments, ranging from governments, global organizations (such as the UN), the whole private sector, science, to final consumers and individuals. Despite the numerous CE definitions, a common sense regarding what CE means is still subject of studies. This opens space for misinterpretation and misuse, as well as greenwashing and image depreciation risks. Consequently, some organizations tend to shape CE to their own definitions and paradigms rather than changing their businesses. This article builds on previous work and aims to establish a common-sense CE definition, separating it from its enablers and related concepts, which seem to be the root causes of misuse. We asked 44 worldwide CE experts PhDs the same question: “Using your own words, please describe what you understand by “Circular Economy”. Database was complicated and analysed through a coded framework and triangulated with the support of R statistical tool. The main outcome is a final definition proposal, along with a structured CE framework. It is expected this research will provide resources to allow standards organizations to establish formal cross-industry CE policies and regulations, leading to scales, targets, KPI's development for CE assessments and audits; and guide organizations and governments on their CE transition roadmaps.
Article
Full-text available
The goal of the research presented here is to describe an innovative approach to predicting the impact of a business messaging campaign, by estimating the percentage of message recipients who will engage with a message. The motivation is to facilitate business marketers to address the problem of estimating the return on investment coming from a potential messaging campaign. The presented solution relies on the processing of large scale business data, taking into account state-of-the-art predictive algorithms, GDPR compliance requirements, and the challenge of increased data security and availability. In this paper we discuss the design of the core functional components of a system that could make this possible, which encompasses predictive analytics, data mining and machine learning technologies in a cloud computing environment.
Article
Full-text available
Aim/Purpose The aim of this study was to analyze various performance metrics and approaches to their classification. The main goal of the study was to develop a new typology that will help to advance knowledge of metrics and facilitate their use in machine learning regression algorithms Background Performance metrics (error measures) are vital components of the evaluation frameworks in various fields. A performance metric can be defined as a logical and mathematical construct designed to measure how close are the actual results from what has been expected or predicted. A vast variety of performance metrics have been described in academic literature. The most commonly mentioned metrics in research studies are Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), etc. Knowledge about metrics properties needs to be systematized to simplify the design and use of the metrics. Methodology A qualitative study was conducted to achieve the objectives of identifying related peer-reviewed research studies, literature reviews, critical thinking and inductive reasoning. Contribution The main contribution of this paper is in ordering knowledge of performance metrics and enhancing understanding of their structure and properties by proposing a new typology, generic primary metrics mathematical formula and a visualization chart Findings Based on the analysis of the structure of numerous performance metrics, we proposed a framework of metrics which includes four (4) categories: primary metrics, extended metrics, composite metrics, and hybrid sets of metrics. The paper identified three (3) key components (dimensions) that determine the structure and properties of primary metrics: method of de-termining point distance, method of normalization, method of aggrega-tion of point distances over a data set. For each component, implementa-tion options have been identified. The suggested new typology has been shown to cover a total of over 40 commonly used primary metrics Recommendations for Practitioners Presented findings can be used to facilitate teaching performance metrics to university students and expedite metrics selection and implementation processes for practitioners Recommendations for Researchers By using the proposed typology, researchers can streamline development of new metrics with predetermined properties Impact on Society The outcomes of this study could be used for improving evaluation results in machine learning regression, forecasting and prognostics with direct or indirect positive impacts on innovation and productivity in a societal sense Future Research Future research is needed to examine the properties of the extended metrics, composite metrics, and hybrid sets of metrics. Empirical study of the metrics is needed using R Studio or Azure Machine Learning Studio, to find associations between the properties of primary metrics and their “numerical” behavior in a wide spectrum of data characteristics and business or research requirements Keywords performance metrics, error measures, accuracy measures, distance, similarity, dissimilarity, properties, typology, classification, machine learning, regression, forecasting, prognostics, prediction, evaluation, estimation, modeling
Article
Full-text available
Although email-marketing is highly profitable and widely used by marketers, it has received limited attention in marketing literature. Extant research either focuses on customers’ email responses or studies the “average” effect of emails on purchases. In this paper, we use data from a U.S. home improvement retailer to study customers’ email open and purchase behaviors by using a unified Hidden Markov and Copula framework. Contrary to conventional wisdom, we find that email active customers are not necessarily active in purchases, and vice versa. Furthermore, we find that the number of emails sent by the retailer has a non-linear effect on both the retailer’s short- and long-term profitability. Through a counterfactual study, we provide a decision support system to guide the retailer to make optimal email contact decisions. This study shows that sending the right number of emails is vital for long-term profitability. For example, sending 4 (10) emails instead of the optimal number of 7 emails can cause the retailer to lose 32% (16%) of its lifetime profit per customer.
Conference Paper
The goal of the research presented here is to support the development of a software platform allowing businesses to improve the way they communicate with consumers throughout the lifecycle of the customer relationship. The motivation is to make communication with consumers more personalized and relevant to each customer’s interests, more direct and interactive, while also ensuring data privacy compliance by design. Personalization is made possible by developing predictive models based on a combination of data from past purchase transactions and past exchanges of messages between the business and the customer. This paper provides an overview of the capabilities of the system and the platform architecture that makes use of predictive analytics, data mining, and machine learning technologies.
Chapter
E-mail marketing is one of the main channels of communication with existing and potential customers. The Open Rate metric is as one of the primary indicators of email campaign success. There are many features of e-mail communication affecting the behavior of individual recipients. Understanding and properly setting these features imply the success of email marketing campaigns. One of these features is the time to send the e-mail. In this paper, we present a methodology for predicting suitable email sending time. Analyzing the available data collected from e-mail communications between companies and customers creates a space for applying data mining methods to data. The proposed methodology for the generation of prediction models to determine the optimal time to send an email has been implemented and evaluated on a real dataset with very promising results.
Article
Article 41 GDPR requires the conformity of data controllers and processor adhering to an approved codes of conducts (CoC) is monitored by a third-party body. The monitoring body can be an internal or an external entity to the owner of the CoC. The paper questions the nature of the monitoring process set by the GDPR to enforce the conformity of adhering controllers and processors with approved codes of conducts (CoC). It aims to determine whether the monitoring process of CoCs is a disguised form of certification or an original enforcement tool introduced by the EU lawmakers in the data protection regulation. It shows that the monitoring process of CoC and certification have the same legal value in the in GDPR despite they have a different nature. It also shows that certification in Article 42/43 is much more monitored by supervisory authorities than the monitoring process of CoC.
Conference Paper
We describe the Customer LifeTime Value (CLTV) prediction system deployed at ASOS.com, a global online fashion retailer. CLTV prediction is an important problem in e-commerce where an accurate estimate of future value allows retailers to effectively allocate marketing spend, identify and nurture high value customers and mitigate exposure to losses. The system at ASOS provides daily estimates of the future value of every customer and is one of the cornerstones of the personalised shopping experience. The state of the art in this domain uses large numbers of handcrafted features and ensemble regressors to forecast value, predict churn and evaluate customer loyalty. Recently, domains including language, vision and speech have shown dramatic advances by replacing handcrafted features with features that are learned automatically from data. We detail the system deployed at ASOS and show that learning feature representations is a promising extension to the state of the art in CLTV modelling. We propose a novel way to generate embeddings of customers, which addresses the issue of the ever changing product catalogue and obtain a significant improvement over an exhaustive set of handcrafted features.
Article
Internet has highly transformed contemporary business practices and presented a new paradigm for business relationships and transactions. Mass market is dead and personalisation is the emerging trend, in fact it has become a necessity in e-commerce. Personalisation simply means individualising the shopping experiences for customers based on data collected about them by marketers. Over last few decades, personalisation has become key element in marketing strategy of e-commerce firms. While personalisation is a buzz word today but conceptually it still lacks clarity. Various academicians and practitioners have expressed different viewpoints on personalisation. In this paper, we try to synthesise various viewpoints on personalisation by analysing key themes, components and approaches in literature to describe the concept of personalisation. The paper also highlights customers' attitudes towards personalisation.