top of page
  • Writer's pictureKarsten Schmidt

Achieving Precise Customer Insights to Improve Journey Quality and Resulting Experience

To orchestrate omnichannel customer engagements, Life Sciences companies must rely on in-field commercial and medical teams to provide valuable insights on customer behaviours and preferences. We have asked Nikhil Gupta, Founding Partner of Xcellen, about latest trends and challenges in areas from customer profiling and segmentation to targeting.

Karsten: Thank you for taking time for this interview, Nikhil. To get started, what are some of the latest trends that you are observing in omnichannel customer engagement for Life Sciences companies and the related challenges for customer profiling?

Nikhil: Thanks for this interview, Karsten, I’m glad to share my learnings with you and the audience.

Coming back to your question: As we all know, the spread of Covid lately resulted in a decreased access to HCPs. Consequently, pharma companies embarked on a “Digital Journey” relying on digital channels (instead of F2F calls) to interact with HCPs. However, most of these efforts were random and knee jerk responses. Many companies and their affiliates were adopting digital channels, as they thought right, with a single objective of connecting with HCPs.

Today however, things are very different, most pharma companies meanwhile have some experience with digital channels. The focus now is to create an integrated omnichannel strategy which involves:

1. integrating all channels to create a seamless customer experience across all of them, including email, social media, chat, voice, and in-person interactions,

2. creating and distributing relevant, consistent, and personalised content that suits customer preferences,

3. leveraging data analytics, customer insights, and AI-driven technologies to provide personalized content and communication across channels.

The companies are also very keen on ensuring that their omnichannel strategies are consistent across markets, if not the same. In a recent project, we helped a multinational nutrition company to create a regional omnichannel framework to be adopted by all countries in the Asia Pacific region. There are many benefits to having such regional / global frameworks as it facilitates

➤ the setting of standardised KPIs

➤ monitoring and comparing performances across the region

➤ driving company’s vision

➤ hardcoding certain aspects of the strategy e.g., not using a particular channel due to compliance reasons

➤ while still allowing customisation at local level

A key challenge to creating such detailed personalised strategy is that you need more and accurate data about your customers’ preferences, their ways of working and engaging.

The companies therefore need to look at multiple data sources rather than just one. They also need to enhance their technological capabilities to capture data in a way that can be used for analytics.

The figure below helps to understand different sources of data that life-sciences companies can use to create an integrated omnichannel strategy

Karsten: What are important aspects to ensure that field-based customer profiling is accurate and properly validated for omnichannel cycle planning?

Nikhil: Field based profiling has always been and will remain a key to collect data about your customers. It could either be your first resort to fill data gaps or your last resort since there is no other data available to complete your customer profiles.

When a Life Sciences company wants to collect profiling data, two aspects must be kept in mind:

1. Nirvana is not possible: You won’t be able to obtain 100% accurate data. There will be biases, and you must live with them.

2. Do not worry, because if you are not achieving Nirvana, no one else will. All your competitors are probably sailing in the same boat.

Now, what separates a company from its counter parts is the way in which they manage field profiling. There are few key aspects to it:

➤ Creating a good quality questionnaire through a cross functional workshop including stakeholders for example from commercial excellence, sales, marketing, digital, medical and compliance functions to ensure that it is aligned with brand strategy.

➤ Ensure that the questionnaire is not too long easily understandable and executable.

➤ Make your sales managers accountable for profiling. Give them access to profiling data in real time to

➢ check progress regularly making sure that the profiling effort is consistent, and on time,

➢ check data quality through insightful dashboard, so that inconsistencies can be identified in real time.

Most companies however, struggle with these aspects especially because the questionnaires are usually just planned by one function and in absence of a robust platform (or using excel sheets or CRM for this purpose). This is where custom built cloud platforms such as Power Customer Targeting are helpful. Xpower offers real time, easy to read data analysis built for sales managers. The interactive dashboard allows them to analyse and challenge the inconsistencies of data in real time.

Karsten: How can Machine Learning (ML) algorithms help to predict customer personas or segments?

Nikhil: Machine learning algorithms can be powerful tools for creating or predicting customer personas or segments. These are some ways in which ML can be used for this purpose:

➤ Clustering: This is a technique of machine learning that merges similar data points. It can be used to segment customers based on their behaviours or preferences. For example, clustering could group customers based on their purchase history, demographics, or interactions with a company's website or social media channels. An interesting contrast to usual customer segmentation is that you don’t have to decide which variables are important for segmentation and weight them. Instead, the algorithms will help you to do that.

➤ Classification / Prediction: Classification can be used to predict customer personas based on their attributes, such as age, gender, location, or interests or any other information available. You can extrapolate customer personas based on sample data out of the rest of the universe with a certain degree of accuracy.

➤ Regression: This is a technique in machine learning that predicts a numerical value based on other variables. For example, we have recently used regression approaches to split the brick potential value across HCPs, a significant data point when segmenting the customers.

➤ Collaborative filtering: This is a technique of prediction based on customers/personas’ historical behaviour or preferences.

By using these techniques, companies can better understand their customers and tailor their marketing, sales, and customer service strategies accordingly.

Karsten: Could you share an example in which Machine Learning helped to build personas and what impact this approach had?

Nikhil: We recently helped a MNC company to develop HCP personas based on their channel and content preferences, using unsupervised clustering algorithms and allowing Marketing team to develop targeted content and channel strategy in accordance with these personas. The team was able to achieve a 40% increase in their webinar engagement from earlier 15-20% to increased >60% rates.

The company had the following data set available:

➢ CRM Data

➢ Headquarter activities

➢ Vendor provided data

All data except vendor provided data was also tagged to product messages and content delivered type.

Step 1 – create micro clusters

Step 2 – creating customer personas by overlapping micro clusters

Karsten: Once customer personas have been determined, what are implications for building segment-based customer engagement journeys?

Nikhil: Creating customer personas is usually just a starting point in defining engagement journeys. A further stepwise process can help to establish your customer engagement journeys (CEJ).

Step 1: Channel Mix Optimisation – Create a channel mix plan or, in other words, determine channelwise investments over a period of 6 to 12 months for each persona using a regression model to determine this split or just base it on your team’s experience or a mix of both.

Step 2: Cycle Plan Creation – Once you have an overall investment level determined over a long period of ~12 months, it needs to be split into smaller implementable cycle plans for 1 to 3 months depending on your promotional cycle. This will allow you to plan shorter targeted campaigns during each cycle

Step 3: Create customer engagement journeys (CEJ) for the cycle – Machine Learning algorithms can be used to determine an optimal customer journey predicting the best possible sequence by analysing the historical engagement data. The clients can also create CEJ manually using the following A-B-C approach:

Aim - Each interaction point should have a clearly defined purpose

Balance - Number of interactions and type of interactions should be well spread across the cycle

Coordination - Each interaction should connect to the next one seamlessly across both the rep and HQ driven channels

Finally, it is also important to evaluate the availability of content and therefore determining the content gap.

Karsten: Thank you so much for taking the time to share your insights and expertise with us, Nikhil. Your valuable input will certainly help our audience to better understand the latest trends and challenges in omnichannel customer engagement for Life Sciences companies.

In summary, we can say that the Life Sciences companies are nowadays focusing much more on omnichannel strategies. These aim at integrating all channels to create a seamless customer experience by distributing personalized content. Leveraging data analytics, customer insights, and AI-driven technologies can help adjust communication across different channels. Regional/global frameworks are beneficial as they facilitate monitoring and comparing performances across the region, setting standardized KPIs, and driving company vision. The key challenge to creating a detailed personalized strategy is that companies need more and accurate data about their customers' preferences and ways of engaging. Field-based profiling remains a keyway to collect data about customers, and machine learning algorithms can be powerful tools for predicting customer personas or segments.

In your experience, what are some of the biggest roadblocks that Life Sciences companies face when trying to implement an effective omnichannel strategy, and how can these roadblocks be overcome?

For further reference please contact or


Xeleratio Consulting Ltd.

We help Life Sciences executives improve sales performance with innovative best-in-class Business Excellence tools and methodologies . Expertise in Business Excellence has been gained with over 12 years of working in different global and regional roles in the Life Sciences industry.


Please feel free to reach out and request a free strategy session whenever convenient. See all articles on Xeleratio Consulting Blog


bottom of page