Top 5 demographic data fields fintech startups need to collect
Published on April 15, 2020
Project: Catalyst Fund
Most fintech startups are focused on finding product market fit, hiring the right talent, and fundraising. They are often not worried about collecting granular demographic data on users, since such data will likely go unused and could introduce friction into the sign-up process. As a result, these companies rarely have the data they need to understand their customers and the usage/transaction data they already have.
However, fintech startups should realize that demographic data can inform better decisions about product market fit, talent, fundraising, and every aspect of business development and growth. For example, demographic data can help fintech startups understand:
- Why are people signing up and/or using our initial product offerings?
- What types of people showed interest, but dropped off early in the funnel?
- Which customers are exhibiting unexpected or unintuitive usage behavior?
However, note that decisions around data collection should always comply with KYC requirements, first and foremost.
Here are top five demographic data fields fintech startups should collect from day one, as well as an illustration of the utility of that data. Once startups master this level of data collection, there is a world of possibilities they can pursue (see BFA Global’s AI Readiness Toolkit for more information).
Life decisions can vary wildly by gender, particularly when combined with other data fields such as age, education, etc. Knowing a user’s gender can reveal differences in spending behavior, the effects of marriage or childbirth, gendered income patterns within various occupations, and preferences for one product over another. Unless a company foresees additional specific gender identities as part of their user base, “Male, Female, Other, and Unspecified” are sufficient answer options.
Use case: Your new data-driven marketing analyst is trying to determine how to allocate scarce advertising resources across various social media platforms, which have very different gender usage distributions. A simple breakdown of users by gender and age can help you decide which platform makes most sense without spending resources to test which will provide the most value.
Age allows fintech startups to know how product usage varies by life stage. Those in their mid-twenties and below likely earn less, but also have few expenses. In contrast, those in their thirties and forties have higher incomes, but manage expenses ranging from child costs to purchase of large assets like houses or cars.
A user’s situation in terms of income and expenses (as indicated by age) can inform product marketing decisions. For example, it makes little sense to market a long-term savings product to an eighteen year-old, or a digital wallet to a 65 year-old who is looking to retire. In many cases, users may feel sensitive about giving their exact birth date since it is used to verify identity in other situations, so merely asking the birth year will suffice. When running analytics, data scientists and analysts will usually group people into age brackets (<18, 18 to 25, etc), so simply knowing the birth year can reduce sign-up friction and seeming intrusiveness, while serving a very useful purpose.
Use case: After a year of data collection, you find that a vast majority (>80%) of investment accounts are used by those above the age of 50. This allows you to better calculate the ROI on your Instagram marketing campaigns, as very few users in that age group use that particular social media platform. You find that shifting those campaigns to Facebook greatly decreases your marketing costs, while resulting in an increase of investment account uptake.
3. Region / Location
For digital businesses, users can be concentrated in a particular city, or can be dispersed across the entire globe. However, fintech startups will likely find that usage and needs vary considerably by location, which can indicate something about urban/rural environments and even socio-economic conditions. Companies should not allow users to self-report location in a text box, even though it may seem like the simplest option at the time, as such data tends to be difficult to aggregate and analyze. Instead, they should spend time breaking down their user base into locational categories that they find distinctive to their particular case and offer those as a dropdown menu. Those that expect a more global user base can find a service that automatically generates such breakdowns by country or state.
Use case: You operate in a city that has six distinct neighborhoods and need to make a decision about where to open up additional ATMs/branches/offices/agent locations. You’ve noticed that one of your competitors is opening up branches in two neighborhoods in which your user base is fairly weak, so you make the informed decision to open branches in your two strongest neighborhoods, making sure to concentrate them near the border of the competitive regions as a disincentive for your users to switch providers.
For fintech startups, education can be a valuable data point, as usage of both financial services as well as tech products tends to vary by education. Startups targeting underserved populations will also find that education is a good indicator for validating that they are reaching their target users. Furthermore, knowing the education level of your users can help maximize the accessibility of marketing campaigns or product tutorials.
Education is generally segmented by highest level achieved or initiated, and could include: a) some primary, b) some secondary, c) some college, d) some vocational degree, d) completed college degree, f) completed vocational degree, and g) completed post-graduate education. One thing this does not capture is whether or not a user is currently in school. Rather than a simple checkbox, it’s helpful to include a field that asks “If in school, what year do you expect to graduate”. This way, data scientists later on can identify student vs. non student behavior with more accuracy.
Use case: A new government initiative is issuing grants to companies that provide financial assistance to university students. If you can target current students, you can offer them special products and establish a great partnership with the government, thereby amplifying the value you provide to your users.
Knowing users’ occupation allows fintech startups to better segment customers, which can lead to unique product offerings, better targeted marketing campaigns, and a slew of other revenue-generating results. This is another field that suffers greatly from allowing users to self-report, resulting in up to hundreds of thousands of different occupations that are both redundant and unhelpful. Conversely, companies sometimes use high-level, equally as unhelpful, sector classifications, such as Construction, Financial Services, Academia, etc., that do not aid in the company’s data collection end goals. Instead, startups need to develop data fields that reflect roles and can differentiate their customers by income frequency, seasonality, amount and type.
Use case: During your next round of funding, your most promising investors are deeply interested in how gig-workers are using your product, as they are a burgeoning industry in your region of operation. Over 50% of your user-base has selected “app-based service industry” as their occupational field, so you can distinguish gig-workers from shop employees, street artisans, etc and understand how customers are using your product.
In general, fintech startups that master data collection and analysis will be much better positioned to understand their customers and make informed strategic decisions. While many focus only on transaction and usage data, insights from such data will be supercharged by the addition of a few demographic fields. Read more about how Catalyst Fund’s inclusive fintech startups are using data for agility and for user insights.