Increasing donations to Grant Circle, an employee relief fund

Background

Many working families are one financial shock away from disaster. Natural disasters, reduced work hours because of the COVID-19 pandemic, or an unanticipated medical issue are among the types of emergencies that could set a household back on utility or rent payments. Missed payments for workers who already live paycheck to paycheck can quickly snowball into cycles of debt, and even eviction.

Employee relief funds are one mechanism to help individuals who experience a disaster-related or personal financial hardship. The way these funds typically work is that an employer and its employees contribute into a pool of collective savings, and when an employee experiences an emergency that leads to financial hardship, they can apply for a cash grant to help them stay afloat. Cash grants are a valuable tool to ease the financial burden of an individual or family who has experienced an emergency, filling in the gaps to provide a safety net and prevent major personal financial crises.

We partnered with Canary, a mission-driven organization that created an emergency relief fund platform called Grant Circle, which facilitates the provision of timely, small-dollar financial assistance in moments of need. Employee donations to the collective savings pool are critical for employer-sponsored relief programs to be sustainable, so we worked with Canary to see if we could increase the number of employees who donate to Grant Circle.

Key Insights

To inform the design of the Grant Circle’s donation webpage, we turned to the wealth of behavioral science literature on charitable giving and donations, and built upon the following learnings:

• Researchers Shang and Croson find that mentioning previous donor contribution levels can increase donation amounts. So as to not anchor the folks who’d otherwise give larger amounts of money onto smaller donation amounts, we included a range of the amounts of previous donations: e.g., “People donate between $2 - $2,500.

• Researchers List and Lucking-Reiley find that seed money increases donations, likely because potential donors view the size of the seed donation as a reliable signal of the cause’s worthiness. On Grant Circle’s donation webpage, we included information about the seed donation(s) that started the Grant Circle.

• Brown, Meer, and Williams discovered that individuals are more generous if they feel they are earning directly for charity rather than donating income they have already earned. On the donation webpage, individuals could choose to donate via a payroll deduction or credit/debit card. In order to encourage the mental model that donors are earning directly for charity, we set the payroll deduction donation option as the default.

After redesigning elements of the Grant Circle donation webpage, we started to think through when the best time would be to encourage employees to donate to their organization’s Grant Circle fund, specifically in relation to an employee’s payday. Would employees be more likely to donate if they were prompted on the same day as their payday or if they were asked a few days before payday? To gather insight, we turned to previous similar research and found some conflicting evidence:

• When people get paid, they tend to feel a sense of abundance and are therefore less likely to consider trade-offs within their budget. This is the idea of opportunity cost neglect, as explained by Frederick et al.

• In a previous experiment CCL conducted, as seen on page 27 of our 2018 Annual Report, there were no statistically significant differences in whether or not people saved if they were reminded to save on payday vs. a few days before payday. However, people who received reminders a few days before payday saved about $30 more than those receiving reminders on payday.

When prompted a few days before payday, it’s possible that people feel close to their next paycheck, so they are willing to set aside more for donations. Conversely, on the day of receiving their paycheck, people might already be planning for how to spend their paycheck, and therefore be more conservative in the amount they are willing to set aside. Applying these findings to encouraging donations to Grant Circle, we hypothesized that more people would donate when prompted on payday, but the average size of a donation would be larger if prompted a few days before payday.

Experiment

To test our hypothesis, we worked with Canary and two of their employer clients who have Grant Circle relief funds. Employees at both organizations were sent an email encouraging them to donate to their respective Grant Circles. All employees were randomized into one of two conditions:

1. Half of employees received the email the day of their payday (on a Friday)

2. The other half of employees received the email the Monday before their Friday payday.

Every employee received the same email, which was designed with many of the same behavioral science insights applied to the Grant Circle donation page. If they opened and clicked the main call to action within the email, participants were redirected to the Grant Circle donation page. In the five days following receipt of the email, we tracked the number and size of all donations.

Results

This experiment concluded active data collection in October of 2020. The email was sent to a total of 1,515 employees across two employers with Grant Circle relief funds. Due to data limitations, we only had access to email open and click rates for one employer. Interestingly, we found that the employees who received the email on payday (Friday) were significantly more likely to open the email, compared to those who received the email on Monday: 33.7% vs 13.3%. While we didn’t have data on when employees opened emails, we wonder whether they were opened on Friday or over the weekend, since many of the employees in our sample are not office workers who spend their days in front of a computer.

In total, 16 employees (or 1% of our sample) donated within five days of receiving the email. Because few donations were made, we were unable to draw any conclusions around our hypothesis. It’s possible so few donations were made for several reasons, including:

1. Employees who would normally be able to donate were struggling themselves due to COVID-19;

2. Workers might generally have email and donation solicitation fatigue;

3. The email wasn’t connected to a particular milestone in time (such as Giving Tuesday or receiving a bonus), so it’s possible the timing felt arbitrary for potential donors.

While we cannot know for certain what the cause of the low response was, we do know that:

1. Email open and click rates are generally low;

2. We did receive a handful of donations after the five-day window, suggesting that our window may have been too small; and

3. A subsequent Giving Tuesday campaign generated higher donation volume.