The blind men and the elephant is a story many of us have heard since childhood. But in real work, I would say most readers have either done something similar or watched others do it.
KOL Mapping is one example where we often see companies "touching only one part of the elephant." When choosing KOLs, everyone knows they should select experts with strong influence. But how should "influence" be defined? How should the expert list be built? That is where the large-scale blind-men-and-elephant scene begins.
First, look at how client-side teams often do it -
1. The veteran approach
They have worked in the therapeutic area for years, carry their own knowledge and "lists," and know many experts.
2. The collective-wisdom approach
Come on, everyone, who do you know? Let's pool the names.
3. The academic approach
Check the societies, check the guidelines, check the top journals...
Then look at how some vendors do it -
1. By department
2. By medical society
Before anything else, we want to ask: what is the core of KOL Mapping? To borrow the phrase made famous by a certain post-Jobs tech leader, what is the first principle of this work?
Two words: ranking. And what is the core of ranking? Quantification and comprehensive coverage.
Start with the easier one: coverage. Without comprehensive coverage, ranking is impossible from the beginning.
For example, if a recent list of academicians did not include Tu Youyou, then any attempt to select the strongest academician would never choose Tu Youyou. If the physician database is incomplete, experts will inevitably be missed.
Now look back at the methods above. Which one is comprehensive? None of them.
There are people you know and people you do not know, people you are familiar with and people you are not familiar with. Add cross-department diseases and experts, and the number grows even larger.
Next comes quantification. Without quantification, there is no way to form a ranking that people can agree on.
For example, the best painting in each person's mind is bound to differ. Some may choose the Mona Lisa, others Starry Night.
But the most expensive painting can have consensus, because "price" is a quantitative standard.
So how do we accurately build an objective and quantifiable evaluation system for physicians and experts? This remains a major challenge. Let us return to the blind-men-and-elephant scene:
- Some people evaluate KOLs based on experts' positions in medical societies.
- Some evaluate KOLs based on published papers.
- Some evaluate KOLs based on clinical trials they lead.
- Some evaluate KOLs based on academic exchanges in the relevant field.
In the end, even within the same field, the KOL lists produced by different methods overlap surprisingly little. Everyone believes their own logic makes the most sense, yet no authoritative standard emerges.
People quickly realize that evaluating KOLs this way is biased - if using only one data source creates bias, then surely making the evaluation dimensions more comprehensive will improve accuracy, right?
The ideal is full and beautiful. Reality is another story...
A product manager at a large pharmaceutical company really tried this. It took one month to produce a KOL list with 10 physicians. The hair lost over that list could probably have been used to string an erhu.
Then, in front of him, we spent 30 seconds ranking 100 KOLs in his field. After he personally recognized the accuracy, this product manager... what happened next was too strange to describe in detail. After all, he is now a regular client.
So how does MeDomino evaluate KOL mapping?
From the beginning, MeDomino's KOL mapping product was designed around two core elements: comprehensive coverage and quantification, covering people, data, and dimensions.
We have collected data on more than 4 million physicians across China, with dimensions covering every type of academic activity you can imagine. Our analysis of each dimension can also go deeper into publication time, participants, impact scale, and more, ensuring that the data is completely objective, real, and reliable.
Of course, our large and comprehensive physician database was not created by sacrificing an employee's hair. It is built on our strong technical capabilities. In the era of big data, we can not only obtain information from as many public channels as possible, but also use AI to structure that information and store it by category across databases. Whatever data direction we need, we can retrieve it accurately and quickly at any time.
At the same time, based on this data, we have built a standardized KOL evaluation system to assess KOLs scientifically and objectively, minimizing the differences between lists produced by different evaluators. Automated data ingestion, automated splitting, and structured standardization allow us to retrieve the right data from the database.
This is an original MeDomino technology and the foundation of our HCP/KOL management product line. Contact us to request a trial.
