In this paper, we present the results of an analysis of Jumia and Konga, (the two biggest E-Commerce stores in Africa) to highlight personalization techniques they implemented using framework for E-Commerce personalization. We also compared how personalized experiences are uniquely provided to each customer using the traces of user's purchase history, browsing history, user preferences, on-site behaviour, and personal data. Results show that Jumia and Konga employ various personalization techniques to boost loyalty among African audiences, improve conversion rates, and increase sales. Our findings could guide designers and relevant stakeholders on how to personalize ECommerce experiences and other related sites to appeal to African audience.
Many E-Commerce sites personalize customer experiences to aid and motivate them to purchase goods and services from their platforms.
There are numerous benefits of applying personalization techniques to
ECommerce systems. For instance, such system could learn from a customer and
recommends personalized products which the customer may need [
The Jumia and Konga E-Commerce platforms were chosen for this research study
because they are easily the biggest and top ranking E-Commerce sites, as well as two
of the top 10 most visited sites in Nigeria and Africa, as seen on Alexa rankings [
Specifically, our studies show that personalization methods can be strategically deployed based on the user’s previous behaviors to boost loyalty among African audience, improve conversion rate and increase sales. Our findings could guide designers and relevant stakeholders on how to personalize E-Commerce experiences and other related sites to appeal to African audience. We limited our research to the customer behaviour requirements as they can be deduced from the system. In future, we plan to evaluate technology requirements; technology implemented by the E-Commerce site in order to tailor contents to specific users.
Jumia, previously known as Africa Internet Group (AIG), is a system of E-Commerce
marketplace and classified websites and applications [
Launched by Sim Shagayain in July 2012 [
In order to assess personalization in Jumia and Konga E-Commerce sites, we
employed the framework for E-Commerce personalization developed by Kaptein and
Parvinen [
The three requirements regarding customer behaviour are 1) the personalized content presented to a user must have an effect on the outcome of the business. 2) The effect should be different for each customer – it should be heterogeneous. 3) The effect should be stable to a large extent. On the other hand, the requirement regarding technology consists of the technology implemented by an e-business in order to tailor contents to specific users. These requirements are: 1) ability to measure the effect of personalization. 2) Ability to manipulate content, 3) ability to scale the algorithm used for personalization. This study focused only on the first category: requirement regarding customer behaviour, since our aim is to reveal the personalization methods used by both E-Commerce sites; Jumia and Konga online marketplaces, and how they are implemented to boost loyalty, drive sales and increase conversion.
To collect data for this study, we used the survey technique and the structured
questionnaire taken in December 2017. The questionnaire was based on a pre-existing tool
developed by Venkatesh [
X = ∑fx / N Where X = the mean score, f = the frequency of each questionnaire item, x = the rating scale point, and N = Total number of respondents on each questionnaire. Findings, essentially shows that respondents are aware and appreciated the facts that Jumia and Konga E-Commerce sites use personalized ads and recommendations, surveys, reviews and ratings, email notifications and many other personalization techniques to provide user-specific contents to customers. 3.1
Using the framework for E-Commerce personalization discussed in [
Close observation show that Jumia and Konga sites provide user-specific contents to their customers. They modify product display shown to the user on the home page using purchase history and browser history of the users, as well as product information that the user checked out or purchased on the site, in the recent past. Additionally, Jumia and Konga sites send personalized email notifications to users using data from purchase and browsing history.
Specifically, Jumia and Konga personalize customer experience using purchase history, browser history, user preferences, on-site behavior, and personal data as compared in this section.
Purchase History: Using user’s purchase history which the system manages, we found out that both Jumia and Konga E-Commerce sites automatically changes the product content on the home page that is shown to each user. The products shown to each user are similar to the one that the user purchased in the recent past from the site. Similarly, products searched for or purchased by others (from same geographical location or age as the current user) are suggested.
For instance, when a user buys a wrist watch on both E-Commerce sites; Jumia and Konga, on coming back to the site, the system shows the customer many other personalized ads of wrist watches offered at discount, especially those that are related to the one that users checked out during their last visit to the E-Commerce site. This type of personalization could lead to increase in sales and conversion rate. Browsing History: Going by the privacy policy statements, it is obvious that Jumia and Konga sites track and collect information about the browsing pattern of users and uses such information to modify what products are served to the users. They analyze user’s previous browsing behaviors and recommend products that they predict the user might want.
On both E-Commerce sites, customer’s browsing history helps the system to keep track of user movement around the site such as what products the user checked out, what products user wanted to pay for, pages the user visited, etc. Using such information, E-Commerce sites presents customers with a personalized reminder of those pages containing products previously checked or visited, so you don’t miss any new thing. For instance, this is why these E-Commerce sites display “Privacy Policy” and “Terms of Use” information that reads as follows: “We use cookies to deliver targeted contents, analyze trends, administer the site, track users’ movements around the site and gather demographic information about user base as a whole. By continuing to browse our website, you accept that we are using cookies…” This mode of personalization could lead to increase in product sales and conversion rate.
User Preference: Jumia and Konga E-Commerce sites readily rely on data generated through surveys, reviews and ratings, to promote personalized ads to their users. This is seen from the buyer trust and safety section in both sites. They show that data generated via surveys, reviews and ratings are used to determine user preferences and subsequently present personalized products to these users. It compares data gathered from a user via surveys against other subscribers and then with the help of the support personnel and system recommendation and personalization algorithms, it promotes personalized products to users.
Again, both E-Commerce sites also use customer reviews and ratings to determine user preferences and subsequently present personalized products to their customers. The system uses how users have rated different products in the past to figure out what they might like and make recommendations. This is particularly visible in Jumia more than in Konga as Jumia has an extensive review and rating feature that seamlessly provides needed user data for personalization purposes.
On-site Behaviour: We observed that both E-Commerce sites: Jumia and Konga employ the on-site behaviour strategy to capture both known and unknown behaviour data from site visitors. This is a type of strategy used by the system to capture product detail pages viewed, product categories viewed, search details for products or category, number of times a product was viewed (aggregated), product(s) added to the shopping cart, etc. This valuable information is used to promote personalized ads and campaigns to users.
Once captured, these on-site behaviors or activities are allocated to a customer’s or user’s profile. With this, E-Commerce site is able to figure out what the customer needs, they are able to show personalized ads to specific users when user comes back or send personalized ads to user’s email address which was supplied during user profile creation. This method of personalization could lead to increase in sales and conversion rate.
Personal Data: These are personal details of users who visit a site. Personal information includes names, email address, gender, marital status, age; address etc; they are usually captured during signup, email campaigns or subscription processes and stored for login purposes as well as for the future needs of the site.
Looking by the privacy policy of both E-Commerce site: Jumia and Konga, it is evident data generated during signup, email campaigns or subscription processes are used to discover what type of user one is and in which category a user belongs. That information assists the system to show personalized ads to specific users when user logs into the site or send personalized ads that are relevant to the user’s interests, unto their email accounts.
For instance, on some weekends, Jumia would use personal data that users supplied to determine prospective ready-for-marriage users and forward personalized wedding brand ads or promote/recommend some other items that it feels such category of users might like. On the other hand, Konga uses data generated during signups, email campaigns ad subscription processes to send personalized ads to customers’ email address reminding them of the exact product page(s) they visited last and/or notifying customers about new products. This way, relevant personalized ads are served to the customer as at when required. This could lead to general improvements in customer engagement, increase in customer satisfaction, sales and conversion rate.
One big difference between the Jumia and Konga is that Jumia has built and integrated a ChatBot known as JumiaBot which assist users to make informed decisions about what to buy and how to buy products in real-time with respect to their cultural and economic status. On the other hand, this Chatbot helps the E-Commerce administrators to quickly track and understand user preferences real-time. This way, personalized ads are served to customers as at when required. This could ensure that customer’s perceived needs are met, sales driven and conversion rate increased. faaUosnsrsdemisshteetduhlpsedeeJsruycsismtstoieioammnCsatrkhoaeanctliBkinno-uetsetor auUnssedersrpaortfienfsegursrevanelcoyenss,e.rteovtireawcsk preferences in real-time.
and unknown onsite be- unknown onsite behaviour haviour and uses them to and uses them to provide provide personalized con- personalized contents. tents.
Uses personal data generated during signup, email campaigns and subscription processes to know user category and likely contents user will need at some time.
Uses personal data generated during signup, email campaigns and subscription processes to know user category and likely contents user will need at some time.
In summary, E-Commerce shop will generally struggle to improve site engagement,
raise sales dynamically and increase conversion if it fails to personalize experiences
to its customers [
This paper contributes to research in E-Commerce personalization by analyzing Jumia and Konga, which are two of the biggest E-Commerce stores in Africa, to discover what personalization techniques they employ, using the framework for E-Commerce personalization. We also compared how personalized experiences are uniquely provided to each E-Commerce user of the platforms using data about user’s purchase history, browse history, user preferences, on-site behaviour and personal data.
Results show that Jumia and Konga apply various personalization techniques to boost loyalty among African audience, improve conversion rates and increase sales. Findings from this research could guide designers in building personalized experiences into E-Commerce shops and other related sites that are specifically targeted at African audiences.
In this study, we limited our research to the customer behaviour requirements as they can be deduced from the system. In future, we plan to evaluate technology requirements; technology implemented by an E-Commerce site in order to tailor contents to specific users.