Online tutor service Paper Education faces growing pains as layoffs follow questions about efficacy

Open this photo in gallery:

Online tutoring provider Paper Education Company Inc., a Canadian tech startup that flourished during the pandemic, is experiencing growing pains.

Last week the Montreal company laid off 105 people – 20 per cent of its corporate staff. It was Paper’s second cull this year, after it cut 81 corporate workers in April. (Paper also employs about 2,000 mostly part-time tutors.) Tech firms such as Shopify Inc. SHOP-T and Clearco have similarly trimmed staff since the tech downturn started in late 2021.

Revenues and bookings lagged expectations in the first half, chief executive Philip Cutler said in an interview to explain the cuts. Buying patterns “are different than they were a year or two ago.”

The company faces several challenges. The U.S. government provided tens of billions of dollars to school districts during the pandemic earmarked for online tutoring services. That ignited demand for platforms and services such as Paper. But that funding is set to end next year; Florida’s Miami-Dade County and Brevard County school districts have already suspended plans to buy such services.

The end of the funding has also prompted some customers to drop Paper, which gives students 24/7, on-demand access through its online portal to live tutors, who help with subjects such as math or English or review essays via text. “The funding cliff is very real and unfortunately has forced us to cut some amazing products like Paper,” said Julie Young, an educational technology specialist with the Tustin Unified School District in California, in an e-mail.

Jeff Burke, superintendent of the Splendora Independent School District, praised Paper for a strong rollout last year at the six Houston-area schools he oversees and said he was satisfied with the product. But Splendora dropped the service anyway: It faces a US$2-billion-plus deficit this year as it struggles to pay for essentials such as salary increases and enhanced safety measures after the Uvalde school shooting. “We had to cut a lot of things we were using with the intention of going back when our financial situation improves,” Mr. Burke said.

With so many spending priorities, how much U.S. schools can afford, without aid, to spend on new technologies “is a good question,” Mr. Cutler said. “It’s unclear how things will shake out.”

Another issue is perceived value for money. While many school officials have praised Paper for providing their students greater access to tutoring, low utilization and awareness are chronic issues. In February, New Mexico cancelled a US$3.3-million contract with Paper after three months. Interim secretary of education Mariana Padilla stated in a letter to the company obtained by education news site Chalkbeat that it was “not providing the results in terms of engagement, support, or delivery of service to the state’s students.”

Boston Public Schools will also discontinue its use of the service after signing up in 2021. “Paper has not been worth it,” said district board spokesperson Max Baker in an e-mail.

In Ohio, Columbus City Schools signed up for a year of Paper for US$913,000 in 2021. The following June, staff asked the school board for a two-year, US$2.26-million extension, to be funded with pandemic money.

Instead, at a public meeting, board members questioned Paper’s effectiveness and complained about the surfeit of digital platforms teachers had to deal with. Chair Jennifer Adair said her daughter’s experience with Paper “was frustrating and annoying and she didn’t want to use it again.”

The board postponed voting until its members could review the usage data, which showed that 7 per cent of some 27,000 students at 44 of its schools used Paper at least once. Less than half of those used it a second time. Columbus decided last September not to renew.

Low utilization has surfaced as an issue “more than we’d like and is certainly something we need to improve on,” said Peter Misek, a Paper board member and managing partner with Framework Venture Partners, an investor. He said that’s difficult when the job of promoting the service is left to overburdened teachers and school districts.

So Paper has invested in raising awareness and engagement locally. This past year it sent a customer service manager to Florida’s Hillsborough County to train teachers and speak to students about the service, put up promotional banners and booths and deliver prizes such as cash, AirPods and pizza parties to the most frequent users. Hillsborough student utilization of Paper more than doubled to 82,000 “learning activities” in the 2022-23 school year. “Our product takes time to be adopted, and it’s great when it is,” Mr. Cutler said.

Paper has also faced tutor discontent. Posts on Reddit reflect a general dissatisfaction with the service, with some tutors complaining about the irregularity of shifts, the lack of job security and the hurried nature of the job, which diminishes quality. On Glassdoor, which tracks employee sentiment, Mr. Cutler has a 40-per-cent approval rating, which he blamed on reviews mostly “from tutors who haven’t had great experiences.”

Chalkbeat has reported that Paper’s tutors often feel overwhelmed having to handle as many as five students in different subjects and grades at a time – all while being timed by the company.

Diya Katra, who joined Paper at the end of 2019 as a part-time tutor while attending university in Mississauga, Ont., said she enjoyed working with students and felt supported by the company. But as it signed up more schools, the workload could become overwhelming. Dealing with as many as six students with varying skill levels learning different subjects was “extremely difficult,” she said. “I don’t think it was fair to the students. When you’re trying to respond immediately when you’re being timed, you’re not always giving the best responses.” She quit in early 2021 after finding it hard to manage an unpredictable schedule while attending school.

Mr. Cutler said a lot of what tutors have shared in public forums “is valid. We’re constantly working to improve that.” He noted that last year tutors spent 10 per cent of their working hours juggling three students at a time and 1.4 per cent with four. Paper’s goal is to have no tutor work with more than three students – even though rival services have told Chalkbeat they won’t let tutors work with more than two at a time.

“The vast majority of our tutors are very satisfied with their work,” he said.

Mr. Cutler, 34, grew up in privilege in Westmount, Que., and started a tutoring business 13 years ago while earning an education degree at McGill University. He was struck by the inequities in education, observing that wealthier students had better access to tutoring than those who needed it most. He started Paper in 2014, a year after graduating, with chief technology officer Roberto Cipriani.

While sessions were one-on-one, tutors would be able to handle multiple sessions at once. It would be less expensive and more scalable than conventional tutoring, delivering benefits to students who could otherwise not afford a tutoring service.

By early 2020, Paper had 30 employees, 100 tutors and annual revenue of less than $2-million from schools in the U.S. and Canada. Then the pandemic hit. Educators and political leaders fretted about the impact of “learning loss” from studying online, particularly among students who were socioeconomically disadvantaged. The U.S. government committed vast sums of money to provide free tutoring. Paper had the right product at the right time and signed up some of America’s largest school districts.

Investors such as SoftBank SFTBY piled in as Paper raised US$380-million across two financings in 2021 and 2022. It now serves more than three million students in some 450 school districts, almost entirely in the U.S. The company was generating about US$70-million in annual revenue as of March.

School districts are tricky customers: Contracts usually require approval by elected board officials; yearly renewals are far from certain given the system’s persistent funding challenges.

For example, the Hillsborough district in Tampa agreed to pay US$2.675-million, or US$26.75 a student, for the 2021-22 school year. When utilization was lower than anticipated, the district negotiated a price cut, to US$12.40 a student, for 2022-23.

Still, the board’s discussion last November about extending the contract at the lower price was no rubber-stamp affair. Two members said kids in their families had never heard of Paper. One questioned the ambitious goals to increase utilization.

“If we’re going to make an investment like this, we want to see students leveraging this as much as possible, and we know we have to do better,” Superintendent Addison Davis previously told the board. He said that if usage didn’t improve markedly, staff would not request another renewal. The board approved the deal 6-1.

Both Mr. Davis and his deputy have since left the school district. Their replacements are reviewing the data and have not indicated whether the district will renew.

One question is how Hillsborough will perceive Paper’s success in improving learning outcomes. Some critics think Paper and other digital tutoring services are no replacement for the real thing, which typically involves a student working with the same tutor several times a week. “If you’re looking for tutoring, Paper isn’t tutoring. If you think of it as a 24/7 on-demand homework help, it is that. And maybe there’s value in that,” said Amanda Neitzel, an assistant professor at Johns Hopkins University’s School of Education.

Paper has contracted research firm LearnPlatform to prove its efficacy in improving learning outcomes by surveying students who have used the service. “That’s really important,” Mr. Cutler said. “It’s critical that students are actually learning when they use Paper.” Several studies shared by the company show it met the applicable “promising evidence” standard under the U.S. Every Student Succeeds Act (ESSA).

But the results were mixed. LearnPlatform’s Hillsborough study determined there was no overall statistically significant relationship between Paper’s math or English tutoring sessions and student achievements. LearnPlatform did find that some of those tutored by Paper had higher math achievement and that a subset of students benefited in English, enough to meet ESSA standards. A study in Arizona’s Yuma Union High School District found a “significant positive correlation between usage of Paper and academic outcomes,” particularly among “English learners“ in English scores.

Mr. Cutler said the surveys prove Paper is effective and benefits students academically. “That means it’s a solution that appeals to our customer base.”

The company has made three acquisitions and enhanced the product, adding voice recognition technology to support reading practice by students. Mr. Cutler expects sales to pick up and said cost-cutting will get Paper to profitability without needing to raise more capital.

Mr. Misek, the Paper board member, said he isn’t concerned that demand for online tutoring will dry up along with pandemic funding. “There’s no scenario, unless society is willing to walk away from this generation, that I see demand being lower in the future.”

Editor’s note: A study by LearnPlatform into usage of Paper’s online tutoring service did not show average usage fell in the 2022-23 school year from the year before; rather, LearnPlattform used a lower sample size for its survey in the second year. In addition, Paper added voice recognition features to its platform to support reading practice by students. The story has been updated to reflect the corrections.