Personalization: How Jiff Does It

Marketers have long understood the power of personalization. People respond better to messages that speak to their unique situation, which is why segmentation and market research flourished in the second half of the 20th century.

Since then, tech companies such as Netflix and Google have revolutionized the field, zooming in from gross and clumsy segments into the needs and desires of specific individuals.

Today, the need for personalization in employee health has become increasingly apparent. Each employee has a unique set of health goals, motivations, and habits. At the same time, employers have invested in scores of apps, tools, and other point solutions that target specific healthcare problems.


The result is a complex and highly inefficient market that fails to match consumer needs with the supply of available point solutions. In this broken market, employees often fail to connect with the solutions that are right for them, and when they do, they are drowned in disparate and conflicting messages.

There are two key opportunities for personalization in healthcare. First, to drive the right employees to the right solution at the right time. Second, to deploy targeted content that keeps employees actively engaged. By seizing both these opportunities, employers can get the most out of their health benefits investments, increasing utilization of health programs by the people who need them most.

How Jiff Does Personalization

Jiff is unmatched in its ability to create a personalized experience for employees. Our distinctive capability in this area stems from three core elements — our data foundation, recommendation engine, and engagement optimization system.

1) The Data Foundation

Any tech company can tell you that most critical element to personalization is a strong data foundation. It takes an enormous amount of high-quality, relevant data to understand the behaviors, motivations, and desires of any one individual.

Jiff’s ability to personalize is unparalleled due to both the volume and variety of data we collect on employees and their health. We aggregate various types of employee data, all in a manner that adheres to the highest standards of privacy and security. Several of these data sources are outlined below:

  • HR / Benefits / Payroll. Data from employers’ such as benefits eligibility, plan design, coverage tier, family composition.
  • In-app behavior. Data on how employees interact with the Jiff app, including clicks, shares, likes, participation in events, redemption of rewards, support calls, and program utilization.
  • Digital exhaust. Jiff’s platform partners share data with Jiff on user behavior across their apps, devices, and services.
  • Real-world behavior. Data on employees’ activities outside the Jiff platform and point-solution ecosystem, either self-reported or captured from their mobile device. For example, location patterns.
  • Attitudes, Goals, and Demographics. Employee data such as age, gender, income, and geography are collected from surveys and HRIS.
  • Personality. Psychographic data on users, based on survey analysis as well as Jiff’s proprietary  algorithms that synthesize data from all of the above inputs.

No single point-solution vendor, benefits consultant, or even payor can claim to have line of sight into all of the above data sources. By combining different types of data, we gain insights into longitudinal consumer behavior that would otherwise be inaccessible. Jiff’s unique role as a platform technology affords us this capability, and is ultimately what sets us apart.

2) Recommendation Engine

Smart recommendations provide employers with greater leverage from their existing health programs, helping to deploy solutions in a more targeted manner. That’s what we mean by “right solution, right employee, right time.” For example, a 35-year-old athletic mother-to-be has much different needs than 55-year-old smoker with heart failure, and accordingly, should participate in different types of employee health programs.

Furthermore, the so-called “right” programs are dynamic and situationally dependent. For example, three years later, that same mother-to-be is now 38 and a mother of two. The “right” programs for her have changed, as have the actions she should take to stay healthy. Jiff’s recommendation engine continually adapts to these changes, adjusting its recommendations to best match employees’ current situations.

Finally, Jiff’s recommendations carefully balance employees’ desires with their needs. We know that to achieve health improvement, we must engage users in both. For example, the 55-year-old smoker may want to join a mental health / mindfulness program, but what he also needs — and what will deliver the greatest bang from an employer cost-standpoint — is smoking cessation. Jiff’s recommendation engine utilizes sophisticated algorithms to take both perspectives into account.

3) Engagement Optimization System

Even the best data in the world or the most precise recommendations will make little impact if consumers don’t engage. That’s why the Jiff app was built with engagement in mind, incorporating core principles learned from the worlds of game design and behavioral economics. But Jiff augments these further by utilizing our data foundation to personalize the experience through a system of engagement optimization.

Think about Facebook’s newsfeed. No one who uses Facebook sees the same newsfeed. It’s customized specially to you, based on your friends, your groups, and your interests. Furthermore, the newsfeed is always learning. It learns about you by tracking what you click on and engage with; it learns from your friends, based on their actions; and it learns from the rest of the world. Billions, if not trillions of A/B tests go into optimizing your newsfeed to make it fun and addictive, and every day it gets better at serving you content you want to see.

Jiff’s engagement optimization system operates much like Facebook’s newsfeed but as a force for positive behavior change for your workforce (rather than for catching up on friends’ baby photos). And soon, we will go one step further.

Not only will we target content, alerts, and events to specific individuals, we will personalize incentives and rewards. The engagement optimization system will keep track of what it takes to motivate a behavior, say, the use of telemedicine. Then, it will personalize the incentive type and amount to each individual, based on what we predict will yield the desired outcome.

For example, based on our understanding of ‘Mark’ (and people like him), we predict he’ll try out telemedicine for 100 points (with a $5 redemption value). In contrast, we know for Sally (and people like her) it will take 150 points. Jiff’s system will allow HR leaders to optimize their limited incentives budget to achieve the greatest impact, all within the bounds of EEOC regulations around appropriate use of employer health incentives.

Personalization is a powerful tool that has penetrated most consumer industries, but has yet to achieve wide adoption in employee health. Jiff is ready to be your company’s partner on your journey towards better health at lower costs.

For more information, guides or additional content visit Jiff’s resource page. Or, to see the enterprise health benefits platform in action, contact us.