In our previous blog, we described the consumer data value chain, the position of data marketplaces within it, as well as the four modes of data collection that have evolved along with the evolution of digital technologies. This blog will expand on these topics by exploring the impact of these data collection modes, particularly in their ability to collect different data types from consumers (data generators) as well as their potential in helping consumer product developers (data users) offer a high degree of personalization and timeliness in their solutions or recommendations.
The different types of data collected on consumers include:
- Personal identifier data, which describes a person's identity, such as their name, full date of birth, or biometric information (like facial features, iris, or fingerprint), address, and social security number
- Demographic data, which describes a person's characteristics, such as age, gender, race, marital status, education, and profession
- Financial data, which includes information on personal assets (like securities and funds), liabilities (like loans and mortgages), incomes, expenses, and credit history
- Location data, which describes a person's location at any given time
- Physical data, which describes a person's physical attributes, such as hair color, eye color, weight, height, and gait, as well as health data, such as physical activity, hydration level, and general state of well-being
- Emotional data, which describes a person's current mood and state of mind
- Social data, which describes a person's interpersonal ties and social networks, such as family members and friends
- Environmental data, which describes a person's preferred environmental settings (like temperature, humidity, and lighting) as well as the environmental conditions in which a person exists at any moment
- Opinion data, which shows a person's preferences (such as likes or dislikes) as well as opinions on a given topic
These different types of data allow consumer data developers to derive varying levels of insights and correspondingly react to consumer needs at different speeds and offer varying levels of personalization in their solutions and recommendations. The degree of personalization and timeliness can be evaluated via several measures, which include:
- Identifying who the solutions or recommendations should be offered to. This largely depends on personal identifier data.
- Identifying what solutions or recommendations should be offered to the consumer. This mostly depends on the consumer's demographics, financial capabilities, physical and emotional needs, personal opinions, social networks, and even environmental data.
- Identifying when a solution or recommendation should be offered to a consumer. This mostly depends on the consumer's financial capabilities and location and environmental data.
- Identifying where a solution or recommendation should be offered. This mainly depends on consumers' location data.
- Identifying how should a solution or recommendation should be served. This depends on the consumer's location data and personal opinions.
- Timeliness, which is how quickly a consumer product developer can respond with solutions or recommendations depending on the consumer's needs. This depends less on the type of data and more on the mode of data collection and measures if the product developer can address the consumers' demands on an annual, monthly, daily, or even real-time basis.
Below in Figure 1, we use a heat map to show how the four modes of data collection impact the type and breadth of data collection, the level of personalization and timeliness in solutions or recommendations offered, and the timeliness with which the product developers can respond.
Figure 1: The four data collection modes and their ability to collect different types of data and enable different degrees of personalization and timeliness in solutions or recommendations
Several interesting conclusions emerge from this heat map:
- Traditional channels are very good at collecting personal identifier and demographic data because they get access to such data during surveys or while registering consumers for membership accounts. However, beyond these data types, traditional channels do not have the ability to access other types of data. As a result, the level of personalization that can be offered using data from traditional channels is quite low. The slow data gathering and analysis process also ensures that product developers cannot react to consumer demands in a timely fashion.
- Traditional devices are good at collecting more than half of the data types identified above, given that consumers pursue the majority of their work and life activities online these days. While traditional devices have strong capabilities in identifying what to serve and whom to serve, they lag in determining when, where, and how to serve solutions and recommendations as well as the timeliness with which they react to consumer needs. This is because traditional devices like laptops are still bulky and not something that consumers carry everywhere and use all the time.
- The arrival of smartphones became a game changer. Compared to traditional devices, smartphones and other similar devices (including apps on these devices) are very good at collecting data across most of the data types defined above. This has been made possible because of the variety of sensors that smartphones have, the variety of apps that tap into all of this sensor data and allow consumers to conduct their everyday activities on the smartphone, and the fact that the small aspect ratio and user friendliness of these devices means that consumers carry smartphones with them all the time wherever they go. As a result, data from smartphones can allow product developers to offer a high degree of personalization in their recommendations and solutions and react to consumer demands in a very timely manner.
- Interestingly, emerging devices like wearables and smart home devices, which are built using the latest technology stack, do not have the ability to collect as wide a swath of data as traditional devices or smartphones. This is because emerging devices focus on very specific applications, have an underdeveloped app ecosystem, and suffer from a highly fragmented marketplace. However, emerging devices are superior to smartphones when it comes to collecting biometric data as well as information on the physical and emotional state of users – a virtual goldmine when it comes to personalizing attributes like when and where to serve recommendations and solutions.
To summarize, those interested in monetizing data generated via emerging devices should carefully analyze the heat map to determine how and where they can extract value from the data value chain. It is worth noting that increasing focus around privacy from both regulatory agencies and consumers is likely to pose a challenge to monetizing data generated via emerging devices – a topic we will discuss in our final blog of this series.