Christian is a computational linguist and philosopher by training and has been working at excentos for the past twelve years. His focus is to develop techniques around natural languages with special emphasis on methods that enable computer systems to unlock the knowledge contained in large amounts of textual data. He is an active contributor to the research field and utilizes his findings to further develop the Machine Learning technology used in excentos’ Guided Selling solutions.

Christian: Artificial Intelligence (AI) research aims to create systems that are able to solve non-trivial tasks in an intelligent way. In the context of cognitive science, it tries to mimic functions of the human mind such as problem-solving.

The field of AI emerged early on with the development of modern electronic computers in the 1950s. Almost from the beginning, there have been two competing paradigms that approached the recreation of intelligence from different principles. The first is the so-called 'Symbolic AI' which is based on large handcrafted systems of deductive rules and the descriptive power of formal logic. The second approach, originating from 'Computational Cognitive Science', is inspired on how a human brain works on a neural level. These were competing paradigms in extreme places on the map of AI, and they work in different ways.

Blurring its original meaning and usage, the term 'AI' is now used almost synonymously for Machine Learning (ML), albeit being associated, is actually much more restricted than the general concept of AI. That being said, ML currently is without a doubt the most important concept in regard to the practical relevance of AI-related techniques. ML enables computers to learn from experience and in that sense perform human-like tasks by processing large amounts of data and recognizing data patterns.

Christian: Perhaps the most prominent applications of AI-related technologies in Ecommerce are search and recommender systems which are almost omnipresent in online shops and websites in general. From a broader perspective the uses of AI in Ecommerce can be considered regarding to two major aspects:

User-Centric Functionalities: In the context of online shops or web presentations for brands, AI enables consumers to explore and navigate voluminous product assortments by search and recommender functionalities as well as by applying techniques for data clustering, which means to come up with automatically created groups of product items based on their underlying data representation. In this way it is possible for a fashion website e.g., to present a range of thousands of different dresses and in terms of a manageable number, let’s say ten automatically and smartly determined product clusters based on available image and text data.

With knowledge-based recommender systems AI also enables the possibility to provide highly personalized recommendation services by combining insights from data-driven consumer behavior analysis with specified user requirements acquired by web questionnaires or conversational interfaces. By these means AI enables consumers to access complex product domains geared to their actual needs and guided by expert knowledge.

Schematic representation of excentos analytics tool

Business-Centric Functionalities: Complementing user-centric applications, like the ones mentioned so far, AI is also used to contribute business-centric functionalities such as optimizing warehouse logistics in terms of demand forecasting or providing valuable consumer insights from collected usage data. Thereby, the latter is a typical use case in which user and business-centric functionalities go hand-in-hand, as customer intents gathered by user-interactions via conversational interfaces can provide a reliable data basis for AI-based consumer insight analytics, as well as product requirements research.

Christian: Today, small teams or even single developers can use various resources to solve specific problems. While this may have been difficult in the past, there is a lot that has changed during the last decade. Despite the tremendous increase of computation and storage capacities, one of the main influences that drove this development was a cultural change that took place in the AI research community, which was essentially taking the contributional work to the open source community mainstream. Therefore, there is a huge amount of high-quality open source software, data collections and research papers freely available to everyone.

The Deep Learning approach to Machine Learning recently led to impressive performance leaps for several tasks like voice recognition, computer vision or natural language processing. Although in principle available to everyone in terms of open source software, applying Deep Learning methods successfully from scratch requires massive amounts of data and computation which heavily limits its accessibility in practice. Even if the costs involved are affordable for a company there simply won’t be sufficient amount of data for many business relevant use cases.

To tackle these limiting aspects of Deep Learning a particularly useful technique called Transfer Learning is in place. Transfer Learning benefits from the fact that the research community and especially the big players in field like Google or Facebook use their vast computational resources along with their huge reserves of data to train state-of-the-art models and make them freely available. Transfer Learning exploits the implicit domain-independent knowledge represented in pre-trained models and transfers it to target domains with limited amount of available data. Smaller businesses that possess the technical knowledge can use these kind of pre-trained models for their domain-specific problems and operate successfully with limited data and resources. But at the same time, for businesses that don’t have any expertise or have even fewer resources, there are several service providers that supply the required expertise for AI-based solutions.

Christian: AI technology has always been at the core of our Guided Selling solutions as we utilize knowledge representation, recommender engines and nautral language processing. We’ve been working on Guided Selling software since 2007, and over the years we have developed and heightened our personal expertise in this field. In the early days, we realised that there was a greater demand for technical expertise to just create a concept for a Product Guide. Hence, self-sufficiency has always been one of our goals and a major change has been to our project delivery process.



With the launch of the excentos Workbench platform, we were able to provide our customers with an easy-to-use tool, while abstracting away from the technically demanding details. Now they can create a Product Guide on their own in 5 simple steps. To get there and to keep going in this direction we are constantly pushing and adopting state-of-the-art results from relevant areas of the field.

Christian: What is surprising at the moment is that a system can go quite far by using language as a stream of patterns. It was thought that it would require a deeper level of understanding to be able to distinguish between a positive statement from a negative one, but this was not necessarily true and it was surprising to see that a system could do this by just using language as a stream of patterns.

Hence, there is a lot of potential to get smarter with recommendations when it comes to creating a system that can extract meaning from what it processes. One example would be a book recommender that goes through the text of a book and states the kind of prose or writing style, the quality of the writing itself, etc.

Extending the scope of knowledge extraction could be very powerful and helpful to us. Imagine that we have some kind of semantic representation that can go through thousands of medical research papers pertaining to cancer and detects a valuable pattern, insight or a finding. This is the kind of AI-powered assistance we need because the key insight is that it complements human endeavor and doesn’t hinder it.

Every website receives some form of user engagement, for e.g., a visitor clicking a link or a button or simply browsing a website or purchasing something online. Web analytics are used to measure, track and analyze such usage metrics to optimize and improve the performance of a website. They provide data that reports on how your business is performing versus its competition; insightful information about its key demographics; user behaviors and insights; the reach and impact of advertising campaigns; awareness of seasonal traffic behavior etc. to name a few. Analytics continually track such varied events and are hence capable of reporting on specific analyses across a vast range of data.

Regular analytics will give you insights, but excentos analytics offer customer insights for specific and defined aspects. Click the button to skip the introduction to web analytics and read directly about excentos analytics.

Businesses differ on principles, objectives and plans, which are also translated to their respective websites. Websites themselves could have an objective other than just selling or even multiple different ones. Goals are used to define different parameters or steps to drive strategies based on set objectives. Analytics provide a way to set customized goals according to your website’s competencies. For e.g. A website can track how many responses they receive for their newsletter subscription or their query form. Such a goal-based approach is quite effective and methodical for optimization.

Web analytics and tracking tools of course keep track of the number of visitors, as well as detailed information about the visits themselves. A user-centric overview is provided about the average time spent by visitors; how many pages were visited per user session; how many visitors are returning to the website and how often. These show the average navigation path taken by website visitors and indicates reasons a visitor might bounce. Other insightful information about demographics such as location, device, real-time reports are all essential to deliver good website functionality as well as to create a data-driven marketing strategy.

While improvements and changes are expected within a website, it is also crucial to know more about how traffic is driven to said website itself. Sources are where the users were browsing before they arrive at a particular website and medium is the method they chose to get there. For e.g. a user is browsing an external source website, sees an ad, clicks on it and navigates to a landing page of the linked website. This makes the first website the source and the cost-per-click ad a medium. This helps to know how much traffic is being generated from organic vs. paid channels and is a good indicator of how much time, effort and money should be spent on each. While search engine optimization provides organic visibility, it is also an important performance indicator as it shows the quality of a website quality and its trust ranking. When it comes to paid advertising, such insights directly impact strategies related to a websites' rate on investment.

Pages on a website host different information and it is important to maintain good and relevant information to ensure quality and to keep the visitor engaged. Analytics track which pages on a website are more popular with visitors and which ones aren’t. As mentioned before, this helps maintain quality which retains visitors and is also used to measure content effectiveness. Behavior is also measured by actions taken on a website and it is valuable to see whether a site’s design is user-friendly. Analytics throw light on which pages cause the maximum bounces and contribute to a higher percentage of the bounce rate.

Ecommerce websites have a transactional or sales-focused functionality and analytical tools are configured to the core purpose of selling and revenue. Hence, tracking customer journeys can give better insights and substantially improve sales margins. More knowledge is always powerful as timely analyses and reporting enable continuous learning from customer behavior and purchases. There are also various tools that report on how a particular product or category is performing. Therefore, with customer and product-centric viewpoints, one can view an online business from different perspectives and can recognize patterns and work towards establishing a good customer relationship.

A product guide leads a customer through a sales funnel that is based on insightful consumer behavior. It engages with them while asking relevant questions specific to their needs. Such an optimization tool is particularly useful for products which answer niche needs, product categories that are complicated and even large product assortments. It also targets navigation and functionality problems such as visitors browsing through countless product pages or manually setting filters to choose products. Although useful in certain circumstances, filters primarily have a functionality akin to settings. Hence, they provide a general output lacking valid reasoning and at times don't even provide results at all. A product guide on the other hand takes the consumer through a series of questions and answers, providing a valid reason at every step of the process and as a consequence, will always display a recommendation. It is also structured to reduce wasteful browsing, which stops visitors from bouncing of your website. For this reason it is an on-site optimization solution.

Schematic repersentation of excentos Analytics

This shows you that out of all the traffic that clicked-through, which category of a consumer was most interested in the bikes. Furthermore, you can narrow down on segments to find out particular choices. For e.g., you could find out the average height of adults who come from a specific country and if there is any correlation with a type of bike or some other related feature.

The image used to illustrate the example gives a schematic view of excentos analytics and shows individual events such as user journeys, bounces and exit points. You can track single journeys through the funnel or get a collective view of all. Data visulizations can be changed to sift through and draw appropriate conclusions.

It shows you which products are performing better than others and their individual click-through rates. You can evaluate product items that are not working in a particular market or segment and plan your inventory accordingly.

An easy overview allows you to know if your products are getting traction in other countries, as well as comparisons between number of visitors, revenue, clicks, popularity and other such parameters.

As it learns from customer data, excentos analytics provides helpful suggestions and steps you can use to potentially improve your business.

Special metrics can be tracked and measured, if they are an integral part of the product guide and customer journey. To give an example, this can apply to products which have a customization feature, and hence can track which color, pattern, template, etc. is more popular with customers.

This also shows us that analytics and tracking don’t have contextual usage barriers. They can be used for general performance monitoring or can be customized to optimize and measure frequently visited parts of a website. Regular web analytics are essential to track, maintain and improve websites, but online businesses can benefit even more by integrating a unique tool such as a product guide. excentos analytics show you exactly how your products are doing with customers from across segments and provide you with even more customized options for optimization.

As physical mobility has been deterred, customers are shopping online and are reacting to the fear of running out. According to NielsenIQ, consumers are focused on products that support health maintenance as well as focusing on finding everyday commodities, in these days of lockdowns. This would appear to be the new normal for the weeks to come, and it is important for businesses to realize consumer priorities. This also means an increase in adoption of digital channels by consumers, yet they will always expect intuitive and apt services and this provides an opportunity for businesses.

A report by ContentSquare says that, “These purchases clearly correspond to the logic of the Maslow pyramid: in order to satisfy needs at the base of the pyramid, consumers are stocking up mostly on food and health necessities, ensuring they have the basic equipment to stay at home (TV, PCs, headsets, fridges…) and taking care of their finances. At the same time, any purchase that falls into the top tier of the Maslow pyramid of needs, such as non-necessary items and of course, any outdoor activities, are dropping significantly.”

The following graph by ContentSquare gives an industry-wise overview:

'Local governments and merchants are distributing vouchers to boost consumer spending in the short-term. As compared to the U.S. and U.K. which are handing out cash to citizens, vouchers are a way of directing consumer behavior purely towards the purpose of spending. “The eastern Chinese city of Hangzhou announced a plan to issue e-vouchers via Alibaba's online-payment service Alipay Thursday. By 4 p.m. Sunday, the resulting consumer spending had amounted to 453 million yuan (about 64 million U.S. dollars), 15 times as much as the total of the vouchers given out, with the redeemed government subsidies reaching 28.93 million yuan, data from local commerce bureau showed,” said one xinhuanet.com article.

Taobao is a e-commerce marketplace owned by Alibaba Group Holding. Recently they launched an initiative to help transform traditional factories in China. Hai Wang, general manager of Taobao’s C2M business unit said that, “Stepping up our made-to-order strategy is an ongoing Taobao initiative to diversify product supplies across the ecosystem to meet demand from our consumers and help manufacturers lagging in the digital race to use technology to transform their processes. Manufacturers will benefit from increased efficiency and reduced costs and will be able to generate higher revenues.”
Taobao expects that the C2M initiative will bring 10 billion new orders to factories from across the country over the next three years, while transforming 1,000 manufacturers into “super factories” with output exceeding 100 million yuan (US$14 million) each, driving productivity at 10 factory clusters in China to 10 billion yuan or more. (Source: The South China Morning Post)

Huawei was set to launch its P40 series in France, which had a total of 1412 cases at the time. Due to rising concerns for public safety, the event was cancelled and Huawei decided to launch the new series through an online event instead.

In order to quickly change tack and adapt, Cosmo Lady, the largest underwear and lingerie company in China, enlisted employees to promote and sell through their social circles via WeChat. They created a sales ranking to drive motivation and participation.

1. Prioritize communicating through digital media by shifting your marketing and sales activities online.


2. Your website should host a comprehensive FAQ section to address all the changes made to your business activities in light of the crisis.


3. Work more on communicating with current customers but ensure you do not overwhelm them.


4. Ensure your website is optimized and working efficiently for customers and potential leads.


5. Keep track of current consumer behavior as certain aspects might become habitual in the foreseeable future.

The economic impact of COVID-19 will be potentially grave. Businesses need to survive through the current slowdown while simultaneously planning around scenarios for a recovery period, with most saying that this period is likely to occur in the last quarter of the year. As the pandemic might be prolonged, transforming offline activities to some form of a digital channel is a necessity given the current situation around the world. Hence, businesses should focus on digital channels along with building a good customer experience.

As more people are working and staying at home, providing a good website experience is essential. 360 leads says, “If there was ever a time for lead nurturing and acquisition, it’s now. Most of us have found ourselves with extra time, working from home, online for the better part of the day, and, with social distancing being the new normal, longing for human connection and reassurance.”

For B2Bs especially, having an online presence and a solid customer experience strategy will not only benefit them in the present moment, but also in the long run. With regard to customer experience, the chart below from hotjar.com, gives us an idea about the major frustrations customers face according to customer experience professionals. Before the ‘new normal,’ we can note that the top three problems were longer response times, not understanding customer needs, and unresolved issues.

hotjar CX graphic

Although shops are opening, the nature of interacting with a customer has changed, as seen with contactless deliveries, curbside pick-ups, and even with in-store staff. At the present moment, it is difficult to foresee when things will go back to the way they were and one way or the other, some offline activities can be routed to online mediums. Hence, employee interaction with customers can take an online form (e.g. video, email or chat) or has been solved by simply providing relevant advice and information on websites. E.g. during the lockdown in China, retail selling has taken the form of chat, and several brands have adopted this as a new and additional sales channel.

A Product Advisor is an ideal tool as it provides advice based on the needs of the customer, in other words, it can use your in-store knowledge to strengthen your website experience. By utilizing this knowledge, personalized recommendations are able to deliver an informative experience.

Jungheinrich Forklift Guide

An example of this is how Jungheinrich — a leading manufacturer of logistics products, uses a Product Advisor to display product recommendations to its website visitors. By using the excentos Product Advisor, Jungheinrich is able to recommend the best forklifts to hire. A visitor is guided through a series of questions to ascertain which kind of forklift is required, and as this could prove to be quite technical, they have also provided an option to seek personal advice directly. Once the visitor gets a recommendation through the advisor, they can generate a PDF with all the product details and reasoning ready at hand. In this way, Jungheinrich uses every interaction point as a lead generation opportunity throughout the experience.

Jungheinrich PDF generation

The PDF itself can also contain outbound links that could e.g. bring the customer back to the product page or connect them to any other step in the sales process. According to an Accenture report, "91% of consumers also say that they are more likely to shop with brands that provide offers and recommendations that are relevant to them." Product Guides and Advisors are an asset for websites as they not only help to assist visitors, but are able to generate leads to ensure that no opportunities are lost.

Email marketing is an important component of lead generation, which more than often is mismanaged. 70% of millennials say that they are frustrated with brands sending irrelevant emails. Brands can stay ‘top of mind’ by sending product recommendations through personalized emails. A lot of people still prefer receiving product information via emails, so ensure that you have a dedicated space on your website with the right solution that encourages sign-ups. The visitor should be curious and interested enough with the website content to voluntarily sign-up to receive personal recommendations, thereby also establishing a higher lead quality. Occasionally, you may also need to send cold emails and although it is an obvious point, it bears repeating that it is essential to have a well-thought-out email template.

'From' name

The ‘From' name is an element that is most prominent when one receives an email. To add a human touch to the email, you can send emails from a sales representative or manager who is assigned to the lead in question.

Subject of the Email

Always personalize the subject of your email as it is one of the first things the subscriber will read. A good example is to use the subscriber's name in the subject line, while also addressing the request they have signed up for.

*example text

Opening Text

The same goes for the opening text of the email, it is important to make the subscriber feel that you are addressing them on a one-to-one level. You should work on the tone as per the intended context, as the same would carry forward through the body of the email. Check out this Sephora example:

*example text

Product Attributes and Images

When sending emails related to your product, always remember to systematically display product attributes to ensure that the subscriber has all the information at hand and that it is clear and easy to understand. Sephora often uses great personalized emails to recommend products that are perfect for the individual customer. Visuals speak louder than words, so ensure that you use the latest images of your products and that they are of good quality.

*Sephora personalized email

Outbound links

Remember to always have a few outbound links to relevant website pages such as product details page, shopping cart, contact information, etc. Always double check whether each link leads to the intended page and whether there are any errors. Asics uses several outbound links in this example:

Asics reminder email

Call-to-Action Button

Create a visible and benefit-focused call-to-action button that will prompt the subscriber to click it. The design and placement of the button are equally important, make sure its position matches the overall reading flow of the email. The same Asics example has excellent CTA buttons at the bottom of the email, that prompt the user to act.

Asics reminder email

Lead generation needs a pragmatic approach and one way to achieve this is to develop your own lead generation strategy:

1. Conduct a Marketing Channel Audit

By conducting an audit you will get a better understanding of which channels might work better to generate leads. Email marketing will probably remain a priority in this list, but also consider the impact of social media, paid advertising, search engine optimization, affiliate marketing, guest blogging, etc. on your lead generation strategy.

2. Raise Awareness and Motivate Interest with the Right Content

Once users click-through to your website, ensure you have created the required landing pages and have left no stone unturned to produce content that gets you results. Keep the aim of such pages in mind—is it to inform the visitor? Or is it a page that should encourage them to sign-up? Guided Selling tools like Product Advisors can be particularly useful as they are interactive in nature and recommend products and solutions based on the customers' needs. Create such relevant lead magnets and ensure they are available at the appropriate stages in a customer's sales journey.

3. Analyze and Adjust

Whichever medium you use, measurement is the surest way to decide what is working for you, hence analytics should unquestionably be a part of your lead generation strategy. You can track visitor behavior, click-through-rates, bounce rates, performance of landing pages, and conversion by using analytics platforms such as Google Analytics, Matomo, Cyfe, etc. There are many more free tools and services available as well. Track email marketing metrics such as click-through-rates, open rate, bounces, sign-ups, unsubscribes, etc. to help create better email templates.

Lastly, remember that relevant communication and presentation of information play a key role in lead generation.