Separating the People from the Problem

From The Practice July/August 2020
Colin Rule and the rise of online dispute resolution

When the Apple II was released in 1977, it was among the first computers marketed and mass-produced for businesses and individuals alike. Apple would later adopt the slogan “The computer for the rest of us,” hinting at its technology’s broad appeal among a nonexpert consumer base. It is fitting, then, that as a grade school student in 1980, Colin Rule first dabbled in the world of online dispute resolution (ODR) by running a bulletin board out of his bedroom in North Texas with the help of his Apple II Plus.

At that time, bulletin board systems were digital networks for exchanging messages and files via modems connected by phone lines—precursors to today’s chatrooms, comment threads, and social media of the internet. As heated, often offensive exchanges—so-called flame wars— became an increasingly common problem on bulletin board discussions, Rule developed an interest in conflict resolution in digital spaces. “This was before I was ever trained in dispute resolution,” Rule says of his early foray into ODR. “There were guidebooks of sorts about how to resolve flame wars, but these were early days.”

Rule realized his dual passions for dispute resolution and technology could come together to create something special.

Rule’s interest in peacemaking would only grow. By the late 1990s, Rule had earned a bachelor’s degree in peace studies, a certificate in dispute resolution, and a master’s degree in public policy in conflict resolution, all in addition to experience as a mediator at the National Institute for Dispute Resolution and the Peace Corps. “I left for the Peace Corps in ’95 when the internet was largely just a figment of people’s imagination,” he recalls. “I came back in ’97 and everybody had email addresses. Everybody was talking about the internet. It was amazing to me how quickly things had changed.”

Rule realized his dual passions for dispute resolution and technology could come together to create something special, and he embarked on a career in ODR. After working for, an online database of mediators not unlike 1-800-Dentist, he founded his own company, Online Resolution, in 1999 in an attempt to translate the nascent field of ODR into a platform that resolved real disputes. Rule’s snowballing expertise would subsequently lead him to key ODR-related roles at eBay, PayPal, Modria, Tyler Technologies, and then back to, where he was recently named president and CEO.

In this article, we shine a light on ODR and its evolution using Rule’s career as our guide. In building ODR systems for the world’s largest online marketplace and for court systems across the country, Rule’s career offers a window through which to observe and understand the larger ODR movement—a movement that is all the more important as the world grapples with the continued impacts of COVID-19, such as how to administer justice and resolve disputes in an increasingly distanced society.


If one were to trace the development and underpinnings of ODR back far enough, eventually they might notice another, very similar term slip into the discussion—alternative dispute resolution (ADR)—and it is worth noting the distinction between the two. ADR is, generally speaking, the broader concept, with some defining it as “[a]ny method of resolving disputes without litigation.” Or, as Rule puts it, “ODR is simply ADR plus technology.”

As one would expect, the history of ADR does indeed go back much further than that of ODR. As Rule describes, forms of ADR have been around since disgruntled customers thought to chisel their complaints onto clay tablets, from consumer disputes over copper ingots in ancient Mesopotamia to arbitration agreements in England’s Middle Ages. Much more recently, ADR has been used in the United States throughout the 20th century in instances such as collective bargaining and the use of other mediation and arbitration methods to resolve labor disputes.

In the 1970s, Rule adds, ADR began to take form as a field unto itself with the rise of programs like Harvard Law School’s Program on Negotiation, which at the time was home to ADR pioneer Frank Sander. “Sander gave a talk in 1976 where he really introduced the concept of the multidoor courthouse,” says Rule. “He argued that maybe we should create different pathways for resolution for different types of cases. Sander was really one of the fathers of the ADR movement.” Around that time, community dispute resolution centers were popping up to resolve certain types of issues—those that were worth resolving but perhaps did not rise to the level of a full courtroom experience, like complaints about barking dogs or property line disputes. As we see below, ODR became a natural extension of these efforts as the tools and capabilities of the internet were introduced and continuously expanded.

Building a system for 60 million disputes: eBay and PayPal

In 2002, Rule published a book, Online Dispute Resolution for Business, distilling his experiences gained from founding Online Resolution. Not long after, he got a call from a senior vice president at eBay who had read his book and wanted to fly him out to talk about it. Rule spent about a week consulting for eBay. A month later, they asked him to sign on as a director in eBay’s trust and safety division to help build up their ODR system. “I was like, ‘Well, I have a company. My whole family’s back in Massachusetts,’” recalls Rule. “Then they said, ‘Here’s how much we’re going to pay you.’ And I said, ‘Oh, OK. So, I guess I work for eBay now.’”

In 2003, Rule moved to California and became eBay’s first-ever director of online dispute resolution. At that time, the company had just acquired PayPal to streamline its buying experience and was already—as it remains today—one of the largest online marketplaces in the world with roughly 86 million registered users. By the time Rule left the company eight years later, the customer base had tripled to 250 million users navigating the platform using more than a dozen different languages. “When I was there, eBay was doing more transactions on a daily basis than NASDAQ,” emphasizes Rule. With such a massive volume of deals, there were inevitably problems. The majority of those problems, he explains, were misunderstandings—maybe the seller mixed up the packages for two separate customers—rather than bad actors. As Rule argues, the relatively low rate of nefarious problems is built into the way eBay functions as a marketplace. Indeed, he jokes that his wife did not believe that a peer-to-peer system could operate at that scale without a greater share of bad actors:

In 99.9 percent of the transactions people do what they’re supposed to do. The buyer pays, the seller sends an item, and the buyer’s happy with it.

Colin Rule, president and CEO of

I remember when I talked to my wife about eBay for the first time, she said, “So, let me get this straight. You’re sending money to some stranger you’ve never met before on the hope that they’re going to send you what you wanted.” I said, “Yeah, that’s pretty much it.” And she said, “That sounds pretty stupid to me.” But it turns out, eBay works. In 99.9 percent of the transactions people do what they’re supposed to do. The buyer pays, the seller sends an item, and the buyer’s happy with it. Of course, now eBay is small compared to websites like Amazon or Alibaba, which are huge e-commerce websites, but those are mainly professional businesses selling to individuals. eBay was really a lot of person-to-person transactions, at least in the early days.

Then again, problems—nefarious or otherwise—are still problems. If a package gets lost, whether there was a mix-up at a shipping carrier facility or it was delivered next door, set aside, and forgotten, it would be difficult to reasonably accuse the seller of malicious intent. Nevertheless, the purchased item never reached the buyer, which is still an issue for the integrity of the platform if left unresolved. As Rule explains, between 1 and 3 percent of transactions generate some type of issue that requires resolution. When the vast majority of that 1 to 3 percent relates to instances where no one was acting criminally, you still have a problem on your hands—or, more accurately, you have around 60 million different problems on your hands every year. “With 60 million disputes per year, we knew pretty much every kind of problem that you could have,” Rule says with a laugh. His task was to build an ODR system that would effectively handle these cases.

Rule is careful to note that for the small fraction of those issues that are the result of foul play, there are systems in place at eBay to address them appropriately. For example, if a buyer purchases a laptop computer on eBay and receives a picture of a laptop computer in the mail instead, this would be fraud and not a case for mediation through ODR. “If a buyer told us their seller had tried this trick, then we’d flip the case over to our fraud investigations team, who’d look into it and maybe call local law enforcement in the seller’s area to file formal charges,” says Rule. After all, he adds, “you don’t want to create mediations or negotiations between criminals and their victims.”

One lesson that Rule and his team learned early on was that framing—which is to say how the platform outwardly labels and describes its ODR process to those involved—matters. He recalls a time later in his career when he worked with tax assessors to help them with an ODR project. “I remember asking the tax assessors, ‘How many disputes do you have a year?’ And they said, ‘Oh, we don’t have any of those.’ After all, these were elected officials and they don’t want to have disputes with their citizens,” Rule explains. “I said, ‘Alright, what about appeals?’ And they said, ‘Well, sure, we have 15,000 appeals a year.’ So I’m thinking, OK, you don’t like the word dispute, but you’re OK with the word appeal. From my perspective, same thing.”

You don’t want to create mediations or negotiations between criminals and their victims.

Colin Rule

Back at eBay, it was equally important to get the wording just right if the ODR system was to succeed, though the literal task before them was deceptively simple. In short, they needed to determine the right text for the hyperlinked button that would lead users to the ODR portal. At first they went with “dispute console,” but like the tax assessors, Rule and his team realized through experience studies that eBay users were not keen to start dialogues about “disputes”—they wanted resolutions. When they considered using “mediation” instead, it confused more than it clarified. “Something like 90 percent of people thought that we were saying ‘meditation,’” says Rule. “They were like, ‘Why would I meditate about my dispute?’” Finally, they changed the button to read, “If you have a problem, click here.” This way, users more or less understood and were prepared for the process. “The mediation, negotiation, and arbitration framing—even the dispute and conflict framing—works well for academics and practitioners because it’s very specific and precise, and it helps us talk about what we’re dealing with,” says Rule. “But when you work with actual people who have these challenges, that framing is often not particularly helpful.” In other words, the guiding principle of the ODR system needs to be about meeting users where they are by addressing their concerns and helping them get on with their lives.

ODR at eBay: The process

Imagine you are an eBay user and some aspect of your most recent purchase did not unfold as promised or otherwise expected. Maybe you paid for a new item and it arrived with clear signs of wear and tear. Perhaps it did not arrive at all. You see the “If you have a problem, click here” button, and you click there. What happens next?

You have begun eBay’s ODR process. The first thing you notice is that you are not immediately directed to a person to whom you might explain your problem. Instead, you are prompted to select the nature of your problem from a lengthy list of possible issues. “Very rarely did somebody come in and say, ‘Yeah, I’ve got a problem that’s not on your list,’ because we had added everything in over time,” says Rule. You select option number six: “I bought an item, it arrived, and I’m not happy with it.” You are then asked to provide details. You write up a few quick sentences: “I ordered this vintage jersey from the seller, and it arrived with cat hair all over it. From the pictures and description, I was led to believe it was in like-new condition. I would not have paid as much if I knew it would be covered in cat hair.” Once you enter your account of the issue, the system then lists some potential solutions and asks what you would accept, such as, “Do you want to return the item for a full refund?” or “Do you want to ask for a partial refund?” The list goes on. You look down the list of solutions and indicate that you would be willing to accept options B, E, and G. The system then thanks you for providing these details and informs you that it will now contact the seller about the issue.

To this point, no human but you (and possibly the seller by now) has interacted with this dispute. The moment you file the dispute—for any reason—the money in the seller’s account is automatically frozen. The system then reaches out to the seller to inform them that the buyer who purchased the vintage jersey has this problem and asks if they would be willing to accept any of the listed solutions. The seller might say, yes, option B of a partial refund is acceptable. They might say, no, none of these solutions works for me. From here, the ODR system would continue to facilitate an asynchronous negotiation between the two sides. Half of the 60 million disputes—30 million—end here, with the two sides coming together in mutual agreement using the ODR system. Apart from the buyer and seller, no human from eBay or otherwise has entered the equation.

In fact, Rule explains, 90 percent of the 60 million annual disputes will be resolved with technology alone. “In all those cases, there will be no human neutral,” he says. He elaborates:

A customer service rep won’t have to touch those cases. It will just be the software that we built working with the parties to find resolutions. After all, we have 60 million disputes with only 25,000 employees worldwide. If all we ever did was work at disputes with our 25,000 people, we wouldn’t even get through a third of that in a year. And that was why eBay hired me—they just couldn’t afford to hire endless people to manually mediate all these cases.

Instead, eBay’s limited number of customer service agents handle only that remaining 10 percent of cases, typically via email and/or phone. And yet, while these elevated cases are being addressed with real employees, Rule emphasizes that even then the process remains streamlined. “The technology makes it very obvious what questions need to be answered in order to reach a resolution at that point,” he says. “An agent could sit down and resolve 20, 30 cases in an hour just going through and asking simple questions and clarifying what information has been submitted.” Information like a shipping tracking number, for example, might settle certain ambiguities or disagreements. The customer service representative would collect as much information as they could and, if no agreement could be reached, make a decision.

90 percent of the 60 million annual disputes will be resolved with technology alone.

It is worth noting the constructive positive-feedback loop in this ODR system—ambiguities at any stage in the process are often grounds for adding more detail or adjusting policy earlier in the process. For example, when ambiguity around culpability for undelivered items arose with some consistency, eBay established a policy whereby the onus was on sellers to document proof that items were not just sent but actually delivered. A seller might opt not to pay the extra cost associated with obtaining this tracking information from the shipping company, but they would do so at the risk of losing in any potential dispute should the buyer raise an issue with the delivery.

As Rule notes, there is always resistance to these types of system changes. “When we announced that they would lose any dispute without paying extra for that type of proof, at first the sellers said, ‘But that’s crazy. I shipped it. I have a receipt from USPS that says I shipped the item,’” says Rule. He continues:

The sellers didn’t like that at the beginning, but eventually what they did like is it became consistent. We didn’t have to do investigations because it was all automated. We can look it all up once they give us the tracking number. We could then write a computer program that would go out to the USPS website and say, is there a delivery confirmation? Yes. OK, seller wins. Is there a delivery confirmation? No. OK, seller loses. It’s completely transparent. Everybody knows what the rules are before you even engage in a transaction.

And, given the small-dollar nature of the disputes, Rule argues that speediness really was a crucial component of making the process fair. “The average value of these purchases is low, maybe $75,” he says. “What we found is that if a resolution took longer than 12 days, the buyer was more frustrated than if they had just lost on day one. If they just got a note on the first day that said, ‘Sorry, you can’t get your money back,’ they were less frustrated than if we made them go into a long process that ended differently.” Again, the focus was on the user experience, recognizing that their interest lies in the dispute being resolved not just fairly but efficiently.

Taking ODR to court: Modria and Tyler Technologies

In 2011, after eight years at eBay, Rule took his show on the road. Applying the experience and expertise gained (as well as the license to some of the technology he and his team had built) during his time at eBay, Rule founded Modria, an ODR service provider. At Modria, Rule created ODR systems to resolve e-commerce disputes over vacation rentals, ridesharing, home repairs, freelance work, and online reviews, each with its own idiosyncrasies depending on the specifics of the services offered. Modria also worked with entirely new types of cases that didn’t intersect with e-commerce in obvious ways, including property tax assessment appeals, health care disputes, insurance reimbursements, and even consumer protection for some transactions that occurred offline. “For any company that had a big pool of cases that they needed to manage, the sorts of tools that we offered at Modria were usually a good fit,” says Rule.

One exception, he offers as an example, is Craigslist. “People would always tell me that I should resolve disputes at Craigslist because they have so many disputes there,” Rule says. There are, however, some crucial differences in how Craigslist functions as a marketplace that make ODR tricky to implement, which help elucidate its limitations. He explains:

The primary problem with Craigslist is they don’t move the money. Most of the transactions are face to face and with cash. If you buy a laptop from somebody at a Starbucks and you take it home and it blows up, too bad. You probably don’t even know the name and address of the other side. You only have an email-forwarding address through Craigslist. So, in that system, I might be able to build a resolution process that will deliver what I think are fair outcomes, but the trouble is, I can’t enforce them because it’s cash. I can’t take the cash out of the other person’s pocket. They could be in another state or another country by the time you file a dispute. There’s a question we ask a lot in dispute systems design: What characteristics of the disputes that you’re targeting make ODR a good fit? And with Craigslist, the characteristics just don’t lend themselves to ODR. You need to change the marketplace quite a bit in order to get it to work because you have to be able to enforce your outcomes.

In 2017, Modria was acquired by Tyler Technologies, one of the largest providers of court technology in the world. To put Tyler Technologies’ reach in context, by Rule’s estimation, around half of all U.S. citizens live in an area where the company helps run some aspect of the court system. With the acquisition of Modria—which brought on Rule as vice president of ODR—Tyler Technologies was bringing ODR to the courts.

There’s a question we ask a lot in dispute systems design. What characteristics of the disputes that you’re targeting make ODR a good fit?

Colin Rule

Rule would often joke at Tyler Technologies that as a mediator without a JD he never wanted to work with the courts. “I wanted to compete with the courts,” he says with a laugh. “I wanted to buy billboards outside the courthouses that said, ‘Don’t go in there! Come to instead!’” After all, he argues, court is often not an efficient option for debtors and others with small-dollar cases (for more on how others are addressing this access-to-justice issue, see “Drawn to Action” and the larger “Access to Justice Lab” issue of The Practice). And yet, when he joined Tyler Technologies, he found himself working with a flood of courts asking for help adapting his ODR solutions to their processes.

For Rule, the goal in working with the justice system was not to facilitate more people “having their day” in court but rather to facilitate agreeable outcomes and solutions to their problems. “What Tyler has done is it has integrated Modria into the court filing flow,” he explains. When a case is filed in a court’s e-filing system, if it is a case type that is considered resolvable by ODR, Modria will automatically invite the parties into a type of facilitated online chatroom aimed at working out the dispute. The chatroom need not include any human interlocutors apart from the two parties, although they have the option to invite a mediator to assist them if they wish. If the parties are able to come to an agreement, then Modria automatically generates a settlement agreement, e-files the agreement back to the case, and the court hearing can be avoided altogether—saving everyone time and money. If the parties cannot reach an agreement, then they proceed to a traditional court hearing.

As Rule explains, the types of cases that tend to land in these ODR chatrooms usually include many high-volume, small-dollar cases. “You see a lot of small claims, a lot of landlord–tenant cases, debt collection, and family cases,” he says. And there are clear reasons some types of cases never interact with Modria while others do. He continues:

With a debt collection case, it’s often pretty straightforward. It’s not that hard to build a software program to handle a debt collection case because they tend to look quite similar. Do you owe this money? Do you have the ability to pay this money? Do you want to change your payment plan? Whereas, if it’s an intellectual property violation or if it’s a contract that went awry, that’s a little bit tougher to build a rules-based, algorithmic ODR process. That’s not to say you can’t—there’s obviously a lot of commercial arbitrators and mediators that focus on those high-dollar-value cases—but that probably is where you’re going to rely on the expertise of a human neutral as opposed to just an algorithmic process.

With the onset of the COVID-19 crisis, courts are increasingly turning to solutions like ODR to provide parties with acceptable solutions to disputes, such as the ones highlighted above, in the face of unprecedented obstacles to normal court processes. Rule, however, cautions that while the ODR revolution is underway, it has not arrived in force just yet. “Right now, most of what they’re doing is the same thing they did offline—they’re just doing it over Zoom,” he says. “It’s a step forward in the direction of ODR, but there’s so much more that you can do with the use of technology.” Rule hopes that, as Zoom exhaustion sets in, judges and court administrators might consider the wider array of tools that ODR could bring to bear on the justice system.

The future of ODR and lawyers

Throughout his career, Rule has spoken to a lot of lawyers and bar associations about the implications of ODR on the legal profession. He notes a lot of anxiety among practicing lawyers and a great deal of confusion among law students who, just as they are learning various ways the law has worked in the past, are faced with a very different outlook for the future. “It’s like what Richard Susskind wrote in The End of Lawyers,” Rule adds. “If you’ve got less than five years left in your practice, you don’t have to read this book. But if you’re going to be practicing for more than five years, you better read this book because you have to know what’s coming.” (For more from Richard Susskind on legal services in online spaces, see “The Future of Courts.”)

Rule likens what he anticipates will be the legal profession’s transition to broader adoption of ODR with the types of changes other industries have already experienced. The financial industry, in his view, provides one useful example. He explains:

It used to be that there were people whose job it was to stand on the floor of the stock exchange with pieces of paper, shouting, “I want to buy 500 shares of IBM. Who wants to sell to me?” That was a whole thing. That was their job. Now, all of those jobs are gone, right? Computers are doing those trades in milliseconds. And yet there are more people working in finance than ever before. Technology didn’t kill the financial industry. It eliminated some jobs and created many more jobs. Now what happens is all these people in the financial industry are helping to program the computers that are going out and doing those trades. Versions of that happened in medicine and in entertainment. Those industries have been completely transformed with technology. Well, now that’s happening in the law.

Rather than extinction, however, Rule sees tremendous opportunity for lawyers in the possible futures of ODR embedded in the legal landscape. Like finance, he expects there will be more, not fewer, people working in the law 10 years from now—their jobs will just look different. Rather than the faceless, monolithic, automated system of justice that some fear with the introduction of technology-enabled systems like ODR, Rule expects a vibrant diversity of systems and approaches that are tailored to handle all the specifics of a given case. This, he adds, will require a lot of effort from experts—largely lawyers—who can inform the programming to ensure it is implemented as intended. Lawyers are necessary in building this future, argues Rule. They should also be aware that, especially in light of the unfolding pandemic, increasing integration of technology in legal systems is not a fad they can wait out.

In many ways, the hearts-and-minds battle within the legal profession is beginning to tilt in ODR’s favor. Indeed, in addition to increasing dissemination of detailed explanations pushing ODR further into the legal profession’s Overton window, influential jurists are beginning to signal support for ODR in ways that are either expected or already underway (such supreme court justices in states like Utah and New Mexico as well as in countries like India). Where exactly ODR goes next remains unclear. One thing is clear: there is no going back to a fully analog system of justice.


Leadership in Law Firms

Executive Education
April 28, 2024 - May 3, 2024