By Jennifer Lee, Jennifer Lui, and Francesco Polizzi

As the world continues to manage the developing Coronavirus (COVID-19) crisis, cities and countries around the world have announced closures of educational institutions at both the local and national levels. Naturally, this is highly disruptive for education and Photomath is beginning to see the consequences of these closures through student studying patterns in affected areas.

Used by ~150-million students, Photomath is the #1 math education app in the world with students solving over 1.5 billion math problems per month in our app. In addition to our scale, the global distribution of our learners allows Photomath usage to be a proxy for global math studying.

We have analyzed Photomath usage patterns in several countries to examine how the COVID-19 crisis has affected students’ studying patterns. Not surprisingly, once an outbreak begins, we see a steep drop in Photomath app usage as life becomes highly disrupted. For example, in Italy, Photomath saw >100% drop in growth following the outbreak of the coronavirus on February 21.

Once schools close and students start studying from home, many seek resources to support their learning outside the classroom. On March 4, Italian officials announced the closure of all schools and universities through March 15 due to COVID-19. Subsequently, Photomath saw an uptick of ~100% growth in usage, an increase of 90% above normal seasonal patterns.

There have been a plethora of news articles about how COVID-19 is interrupting children’s education and how teachers are trying to address the situation with digital edtech solutions and conferencing tools. Photomath, in particular, is a supplemental edtech tool that students use to get “unstuck” on math assignments and to review problems and concepts from class. The increased Photomath usage in Italy following national school closures shows that students have returned to their pre-outbreak levels of math studying and then some. Moreover, the massive increase in usage, above and beyond the norm this time of year, indicates an increase in self-directed learning by Italian students at home.

In Japan, a similar pattern emerged. As early as February 4th, a cruise ship was confirmed to have 10 cases of coronavirus. It was quarantined for two weeks and finally docked in Japan on February 12th. As the new coronavirus spread across the country, schools throughout Japan closed on March 2nd.

As with the Italian case, the increase in Photomath usage in Japan post school closures (>100% growth) suggests that students are studying at least as much math, if not more, than pre-outbreak levels. Again, like Italy, the major increase in usage we’re seeing these past few weeks suggests Japanese students are also engaging in self-directed learning.

Although the COVID-19 pandemic is clearly very disruptive to all aspects of people’s lives, Photomath usage suggests that students in both Italy and Japan have continued to study and learn during this difficult period.

In the US, which lags behind Italy in terms of COVID-19 by about ten days, Photomath is beginning to notice a similar pattern.

On March 13, the White House declared a national emergency after the World Health Organization categorized COVID-19 a pandemic. Across the country, school closures are being announced at both the local and state level, with potential for extended closure. Some states, like Kansas, have gone so far as to close all K-12 schools for the remainder of the school year.

As with Japan and Italy, we have seen a dramatic 100% drop in American students studying math since the declaration of national state of emergency.

As of March 20, app usage in the US started to increase, but we have yet to see the same degree of uptick as seen in Japan and Italy. This suggests that American students have not resumed studying at home following school closures. However, many US school closures only began the week of March 16th so it is unclear if the school year is over or if students can continue learning via digital tools like in Japan and Italy.

Many education experts have noted that the inequality of infrastructure and access to resources such as high speed internet across the US is a major challenge to digital distance learning. Unlike other online learning tools though, Photomath’s mobile app works without an internet connection, which makes it ideal for distance learning. As students continue to be assigned math homework, Photomath can support their continued learning regardless of connectivity. Therefore, we would expect any extended low engagement in the US to reflect the school year simply ending early for the majority of US students rather than it reflecting a difference in internet accessibility between countries.

As we continue to track the data, a couple of questions remain:

1. Will enough American students be able to adopt digital education tools to continue their learning like in other countries?
2. How will COVID-19 affect learning the rest of the year?

Photomath’s response

In the face of this extraordinary and unprecedented event, Photomath is rolling out free access to Photomath Plus, our premium paid product, through April 30, 2020. Accessible at https://photomath.net/covid19-support.

Nothing will truly replace live face-to-face schooling, but we believe everyone, whether you are a parent, student or educator, should have access to all the tools possible to cope and help promote independent learning.

Photomath Plus is currently only available in English. Free access to Photomath Plus is available in app versions 6.9 on iOS and 6.5 on Android and later.

ABOUT PHOTOMATH

Photomath is the world’s most used math education app with over 150-million learners globally. Powered by advanced AI technology, the app instantly scans, accurately solves, and intuitively explains math problems to users through step-by-step explanations, ultimately promoting comprehension of fundamental concepts. Based in San Mateo, the company is funded by investors including Goodwater Capital, Learn Capital and GSV Ventures.

Three months ago, I was having a discussion over lunch with friend and Mozilla co-founder, Asa Dotzler. Asa lives in the Santa Cruz Mountains where he has a 1mb/s internet connection. Browsing with worse-than-3G speeds, some pages take him more than half a minute to load. As he explained, this makes entire portions of the web inaccessible, especially those with heavy advertisements. Although many modern ads don’t block webpages from loading, they often require non-trivial computing resources; eating your battery quicker, taking more memory, and inevitably putting page content in front of the user slower. Turns out, this problem is not isolated to the homes of the Santa Cruz Mountains an hour outside of Silicon Valley. Much of the world is negatively impacted by poor practices in digital advertising. These practices perpetuate the technical divide, gentrify the internet, and inevitably harm the Open Web.

Let’s rewind to 2001 at the apex of the dotcom crash. At the time, digital advertisers were compensating for loss of revenue by embracing popups; a tactic which quickly increased ad clicks and severely degraded user experience. After becoming the first web browser to adopt popup blocking out-of-the-box, Firefox signaled the beginning of the end of popups. An elegant solution to a complex issue, popup blockers uniquely addressed the problem with digital advertising at the time.

The health of digital advertising is declining once again, and with it, the health of the web.

Fast forward to today; the modern remedy to invasive advertisements has been ad blocking. Browser extensions like Adblock Plus provide immediate utility to users like Asa; an easy “opt-out” for those who would like to skip ads entirely. Meanwhile, behind the scenes, an elaborate game of cat and mouse is unfolding. Independent ad supported websites are forced to add heavier advertisements while being starved of revenue. Ad Blockers attempt to identify ads while Ad Networks attempt to avoid detection. Enter stage left: full page ad takeovers, ad-blocker blockers, first party ads, and the list goes on. Failing to uniquely address invasive advertisements, blockers serve as a band-aid on a bullet wound; a temporary solution to a systemic problem. The health of digital advertising is declining once again, and with it, the health of the web.

The problem of invasive advertisements isn’t inherent to the essence of the web. There exist ad networks on the internet who are committed to providing responsible ads; many of whom go unnoticed by content creators. At the crux of their invisibility is the ease of content creation in the modern age. Tools like Wordpress and Blogger don’t only make creating websites simple. With one-click plugin installations, placing ads without considering which ad network to use has become the norm. Choice exists for creators, however, the appearance of choice and the tools to choose wisely do not.

Today we are excited to announce the launch of SiteSonar, a browser extension aimed at putting an end to bad ads. More

SiteSonar is a browser extension aimed at putting an end to bad ads through crowd-sourcing performance of advertisements on the web

The SiteSonar browser extension (Firefox/ Chrome) is at the center of our mission to empower creators on the web. Through automatically monitoring ad performance as users browse, it empowers users by shedding light on how ads affect their everyday browsing. Through crowd-sourcing that performance data for display on our public dashboard, it serves an important role in empowering developers to make educated decisions when choosing an ad network.

Additionally, the performance tool allows users to profile specific sites for ad performance. You can use this to profile your favorite sites, or if you are a developer, sites that you maintain yourself. Asa and I both used this to profile our personal websites with some surprising results. Check it out to make sure you’re not asleep at the wheel as a content creator!

Interested in doing your own data analysis on the benchmarks that SiteSonar creates? We’ve got an export tool that allows you to do just that. All data you collect is available via the export tool in JSON format.

Lastly, SiteSonar collects a subset (roughly 10%) of your browsing data to do its job unless you choose to opt-out (check “settings”). We believe that the Open Web should mean equal access; something that can be attained through crowd-sourced data to empower content creators. As such, we care about your privacy and will never attempt to personally identify you. Check out our privacy policy and the data we collect for more information.

The SiteSonar dashboard is a tool for creators to view the crowd-sourced data that we collect. Aggregating benchmarks by ad network, anyone can view a breakdown of the average TTFB and File Size per ad. Additionally, users can drill down into ad networks to view the most common type of asset each ad network loads, along with the number of ads we’ve benchmarked for the network, and the distribution for each benchmark.

When browsing on slow and metered connections, we realize that every byte and millisecond matters. For that reason, we also aggregate ad benchmarks by individual website in order to give users the ability to make conscious choices for where to browse.

By crowd-sourcing ad performance from users like you, we hope to give power back to creators, and put an end to bad ads once and for all. Join us:

Download for Firefox or for Chrome.

& view the crowd-sourced data we’ve collected at site-sonar.com/dashboard.

Thanks from the SiteSonar Team: Francesco, Asa, and Purush. This project wouldn’t have been possible without substantive additions from awesome people like Justin, Luke, and Harald.