Factors Affecting Response Rate in Messenger Marketing
Оригинальность работы 94%
Table of Contents
1. Literature review14
Introduction to literature review14
Existing studies and gaps in chatbots and messengers fields15
1.3. Research design, sample description and gaps in existing literature16
Dependant variables and gaps in existing literature17
Independent variables and gaps in existing literature19
Conceptual framework from email marketing studies as a basis for messenger
Peculiarities of messenger-marketing and key definitions22
Advantages and value of messenger-marketing for business25
The principles of automated sales funnels in messengers and how it is applied on
Conceptual framework for current messenger marketing study39
Modelling and hypotheses development41
Data collection and sample50
Data preparation and methods of analysis51
3. Empirical results53
Results and hypotheses58
Limitations and further research63
Today the paradigm of communications is changing rapidly. There is a global shift towards faster and shorters ways of communications such as text messaging using instant messaging apps. This is evidenced by the increased popularity of such applications. According to Kommando Tech (Kommando Tech, 2019), 41 mln text messages are sent out every minute. Moreover, messengers are currently over 20% bigger than social networks. Around 2.9 bln people currently use WhatsApp and Messenger, and there are 2.52 billion people using messengers on mobile devices. This number is forecasted to reach around 3 billion in 2022. Thus, since messengers are becoming even more popular than social media, companies all over the world start to see them as a prospective channel to get closer to their customers.
In order to communicate with their clients, organizations are beginning to use various instruments of messenger marketing such as channels, chats or chatbots (Følstad, 2019). As for chatbots, specifically, at the moment, chatbots can be seen as one of the most fast-growing marketing tools. Currently, the chatbot market size is projected to grow from USD 2.6 billion in 2019 to USD 9.4 billion by 2024, at a CAGR of 29.7% during the forecast period (Markets Insider, 2019).
Concerning Russia, Russia shares global trend of increased popularity of instant messaging apps. In June 2019, messengers got the first place in terms of the most demanded smartphone functions in Russia. Besides, the popularity of messengers is increasing in Russia: 53% of respondents surveyed started using them more often in 2019 (Yakovlev, 2020)
Now, chatbots are widely used as part of sales funnels by the range of various firms from different fields. They use them to automate lead generation process, automate communications and sales as well as to improve their overall financial performance (Riy, 2019). Chatbots can be considered a valuable tool within sales funnel as they show a higher lead conversion rate, their functionality allow to automatically ask questions, collect answers,
Nowadays, conversion rates in messengers are relatively high as this is a brand new channel: it is not yet overfilled with advertisers, users are not yet tired of constant commercial offers and tend to perceive any types of communications positively in messengers (Serebryakov, 2019). So, at the moment, users are more open for communications in messengers in contrast to other channels, and they are more willing to share information with organizations in instant messaging apps. To make more profits and increase ROI marketing managers and individual entrepreneurs are looking for ways to optimize sales funnel in messenger marketing. But, generally, their knowledge is limited with STP principles (segmentation, targeting, positioning) which is the basis of any sales funnel in any channel (Feld, 2012). Although almost any specialist knows about the importance of segmentation, targeting and personalization for higher conversion rates, there is a clear lack of knowledge on such factors as text message design, timing, channel (specific instant messenger app) or content features which are the key elements of communications in instant messaging apps (Baturin, 2018). Having this knowledge could shed light on how to optimize messenger mailings campaign conversion rate and make the whole sales funnel more efficient.
Unfortunately, at the moment, messenger marketing is 90% unexplored area worldwide, and current study has a potential to lay the foundation for further research in this field. There is no research investigating the general phenomenon of chatbots in messengers despite the increased popularity of this trend. Also, none of the existing works considers chatbots in the context of messengers or automated sales funnels. Besides, none of the existing studies on chatbots focuses on factors affecting the response rate in chatbots in messengers, specifically.
Moreover, existing theory is either too general or too narrow and focuses on separate instruments or channels. Most of the reported studies consider chatbots and messengers as separate instruments and channels (Weißensteiner, 2018). There are some general works on how chatbots influence marketing (Zumstein, 2017; KACZOROWSKA-SPYCHALSKA, 2018; Ojapuska, 2018;) and some specific articles focusing more on technical aspects of chatbots (Hildebrand, 2019; Jain, 2018). In addition, there is a large number of studies
Some of the research also measure adoption level of chatbot technology.
In some of the studies, authors consider Facebook Messenger chatbots alone limiting their scope with this only channel (Pereira, 2018; Balasudarsun, 2018). So consolidating and putting text mailings data and response data for different messenger marketing channels such as Telegram / WhatsApp / Viber or VK into one research model would allow us to gain some unique knowledge on how conversion rates in messenger mailings differentiate by various instant messaging apps.
Most of existing studies share some common methodological gaps and gaps in research design. In their methodology field experiments within only one single organization from one single industry were mostly used. In addition, most of them considered small number of attributes and campaigns (2-30) (Duijst, 2017). Moreover, existing papers analyzed only aggregated conversion (direct mail response) data on overall campaign level which reduced the reliability of final results. Hence our study that will consider more than 3 million mailings and more than 25,000 responses. It will include data on approximately 500 companies from more than 20 industries. It will analyze more than 1500 campaigns (mass mailings) and consider text mailings conversions on individual recipient level.
The current paper will investigate the ways of messenger mailings response rate optimization as “response” is currently seen as the bottleneck in messenger marketing that has a potential to be optimized which would ultimately make the whole sales funnel more efficient (Borcov, 2019). In order to develop a conceptual model and put hypotheses on how various factors could affect response rate in messenger marketing, the literature on direct mail response rate optimization was chosen for the analysis. It is motivated by the fact that messenger marketing is based on the same fundamental principles as email marketing. Thereby, the goal of this research is to identify how timing, channel (specific instant messaging app), visual design (specific emoji and emoji count) and information intensity (text message length) affect response rate in messenger marketing.
According to theoretical knowledge from academic literature, the following hypotheses have been formulated. H1 - timing positively affects the probability of response
(the later a text mailing is sent the higher the probability of positive response will be). H2 - information intensity negatively affects response rate (the longer text message is the lower the probability of positive response will be). H3 - visual design positively affect response rate (the more emoji text message includes the higher the probability of positive response will be). H4 - top instant messaging apps by their response rate: 1 - Telegram, 2 - Viber, 3 - WhatsApp, 4 - VK, 5 - Facebook.
The research design of current paper will include more than 3 million messenger text mailings and more than 25,000 messenger responses acquired from «TextBack» servers – leading messenger marketing service platform in Russia. For the analysis, Python has been chosen as an instrument. Random forest method will be used to identify how factors differ in the importance. Kernel Density Estimation method will be used to identify how conversions are distributed depending on the value of a parameter being analyzed.
The results will be useful both for marketing managers, entrepreneurs and academia. New knowledge can be used by marketing specialists to enhance the results of messenger marketing campaigns, maximize ROI and LTV of their customers by adjusting time, length, emoji count, channel and eventually making their messenger mailings more efficient in terms of response rate. As for the academia, these results on messenger mailing campaigns optimization will lay the foundation for further research in the messenger marketing field.
Ayodele, T., Zhou, S., Khusainov, R. (2009), Email Reply Prediction: A Machine Learning Approach. Conference Paper, 2-4
Babanov, P. (2019), Automated sales funnels and sales funnels: is there any difference? Cossa,1-7
Babanov, P. (2019), Messenger-marketing: what is it, and what role do sales funnels play here? Cossa,1-3
Balasudarsun, N. L., Sathish, M., Gowtham, K. (2018), Optimal ways for companies to use Facebook Messenger Chatbot as a Marketing Communication Channel. Asian Journal of Business Research,8,6-15
Baturin, A. (2018), The secrets of efficient messenger marketing. WebEvolution,3-4
Borcov, A. (2019), How you can use messengers in your business: automated mailings to
Chat2Desk (2019), Enhancing sales funnel with messengers. Spark,1-4
Coussement, K., Harrigan, P., Benoit, D. (2015), Improving direct mail targeting through customer response modelling. Elsevier,11,7-9
Duijst, D. (2017), Can we Improve the User Experience of Chatbots with Personalisation? University of Amsterdam,11-18
Feld, S., Frenzen, H., Krafft, M., Peters, K., Verhoef, P. (2012), The effects of mailing design characteristics on direct mail campaign performance. Intern. J. of Research in Marketing, IJRM-00957,4-8
Motivation. International Conference on Conversational User Interfaces (CUI 2019), 5-8
Følstad, A., Bjørkli, C. (2018), What Makes Users Trust a Chatbot for Customer Service? An
Exploratory Interview Study. SINTEF, Oslo, Norway, University of Oslo, Oslo,10-13
Hildebrand, C. Bergner, A. (2019), AI-Driven Sales Automation: Using Chatbots to Boost Sales. Sciendo,10,39-41.
Jain, M., Kumar P., Kota R., Patel S. (2018), Evaluating and Informing the Design of Chatbots. IBM Research, India, Computer Science & Engineering, University of Washington, 18, 3-11
Karionov, E. (2018), How to increase sales with messenger marketing. Medium,5-10
Kazakov, A. (2019), Sales funnels in messengers - must have for online sales growth. Kazakov - Sales from A to Z,1-5
Kemp, S. (2018), DIGITAL IN 2018: WORLD’S INTERNET USERS PASS THE 4 BILLION MARK, Global Digital Report,1-8
Kluyko, G. (2018), Chatbots, messengers and email marketing: the practice and theory of
holistic use. ESputnik,12,3-8
Kobzar, E. (2019), Chatbots in Messengers: expectations and reality. WebPromo Experts Students Day Conference,1-10
Komarov, V. (2018), How to use WhatsApp bot - efficient way of quality lead generation through messengers. Avic Agency,1-5
Korepanov, K. (2019), How to build automated sales funnels in messengers - manuals and
Markozashvili, D. (2018), The Development of Chatbots and Automated Sales Funnels
through Messengers. VentureCapital,1-5
Pereira, J., D ́iaz,O. (2018), A quality analysis of Facebook Messenger’s most popular
chatbots. University of the Basque Country(UPV/EHU), 2143-2147
Research and Markets. (2019), Global Chatbot Market Anticipated to Reach $9.4 Billion by
2024 - Robust Opportunities to Arise in Retail & eCommerce, Markets Insider,1
Rettie, R. (2003), An evaluation of e-mail marketing and factors affecting response, Journal of Targeting Measurement and Analysis for Marketing,6-10
Rieke, T. (2018), The relationship between motives for using a Chatbot and satisfaction with Chatbot characteristics in the Portuguese Millennial population: an exploratory study. U.PORTO, 5, 25-35
Riy, V. (2019), Sales funnels in messengers. WebPromoExperts,4-5
Ryzhkov, A. (2019), 9 main advantages of messenger marketing. VentureCapital,1-2
Serebryakov, A. (2020), Everything about mass mailings in messengers. Netology Media,1-5
Serebryakov, A. (2019), The trends of messenger marketing in 2019. TextBack Blog,6-7
Serebryakov, A. (2018), What is messenger marketing, and why you should use it. Cossa, 1-10x
Sopina, D. (2018), Sales funnels in messengers. Postium,1-8
Spajic, D. (2020), Text, Don’t Call: Messaging Apps Statistics for 2020, Kommando Tech, 1-4
Spychalska, D. (2019), How chatbots influence marketing. Sciendo,23,258-265.
Tikhonova, E. (2020), Messenger marketing: mass mailings, chatbots and sales, Netology, 1-2
WhatsHelp.io (2018), Why automated sales funnels in messengers are needed, and why they
Yakovlev, A. (2020), How chatbots and sales funnels in messengers will automate your sales
in 2020. VentureCapital,1-5
Zaroual, B., Broeck, E., Walrave, M., Poels, K. (2018), Predicting Consumer Responses to a
Chatbot on Facebook. Cyberpsychology, Behavior, and Social Networking,10,6-7
Zherinova, Y. (2019), All you should know about messengers if you are a marketer. SEOnews,1-10
Zherinova, Y. (2019), Chatbots and automated sales funnels in messengers: what should you start with? PPC.World,7,1-10
Zumstein, D., Hundertmark, S. (2017), Chatbots – An Interactive Technology for Personalized Communication, Transactions and Services, IADIS International Journal on WWW/InternetVol. 15, No. 1, 96-109