Normalizing Failure with Online Dispute Resolution

Courts were working on ways for litigants to engage remotely before COVID-19, though it may be that we just took notice because of COVID-19. It may also be that before COVID-19, court conversations on remote engagement were insular. One of the most celebrated implementations of remote engagement is online dispute resolution (ODR), a technique widely used in the private sector, particularly among online retailers.1

Courts around the country have implemented or are considering to implement ODR. Unfortunately, there is a lack of rigorous evidence to say one way or the other if it works. Perhaps as a product of the excitement about remote engagement, there seems to be no appetite for developing a body of evidence surrounding it. If evaluation suggests ODR does not work now, would it work if it underwent refinement and adjustment? Can the justice system become comfortable with a system that may need some later iteration to be useful?

ODR implementation should be accompanied by rigorous evaluation. ODR is not free, either to start up or to maintain, and when resources are stretched thin, courts should know what works as they decide whether to invest in or maintain ODR systems. Stepping back, ODR is the use of technology to automate some or all steps in a resolution process.2 Because it can reduce communication costs for parties who have access to high-speed Internet, it could be useful for any phase of adjudication. The Hague Institute for Innovation of Law (HiiL) suggests that ODR may be particularly useful for at least three phases of resolving a legal problem: diagnosis, conciliation, and decision.3 In the United States, courts thus far have shown less interest in engaging with the diagnosis phase through ODR.

There is a general reluctance among courts to engage with disputes or parties pre-filing. So that leaves conciliation (settlement) or decision as areas of focus. For conciliation, parties can engage in settlement negotiations as they please, but ODR may offer a low-cost method of communication. Presumably, that’s why the private sector relies on it so heavily. For some high-volume, low-level, and factually simple disputes, like traffic cases,4 possibilities related to the decision made may be the focus.

ODR can be flexible. One promising aspect of that flexibility is asynchronicity, that is, the possibility that litigants and the court need not meet (even virtually) at the same time. The asynchronous nature of ODR has the potential to improve access to justice by removing barriers that come with mandatory physical presence at a court facility. These barriers are the usual—transportation and childcare challenges, missing work for an unpredictable amount of time, or concerns about public health both for individuals who are in high-risk categories and those who may be unwittingly contagious.5 Also, proponents of ODR suggest that remote interactions with the justice system and opposing parties are less stressful than in-person interactions. The theory suggests increased resolution of cases by those who otherwise would be unlikely to proceed among stressful or anxiety-laced case events.6 When parties are able to distance themselves from overwhelming and immediate emotions in their cases, they may be able to use more caution in composing language, which could facilitate settlement.7

As true of most innovations, there is no consensus that ODR works, nor that asynchronicity is positive. Some have argued that the asynchronous model results in response delays, prolonged waiting periods, and sabotaged conversations, all of which can heighten emotion.8 While some suggest power imbalances among parties may be neutralized by ODR,9 others fear they will be exacerbated.10

Courts must sometimes balance fairness and efficiency. ODR has the potential to create efficiencies, especially around high-volume dockets, with little reimagining of the process.11 Is it wise to replicate the in-person process in an online setting, and is it a concern when most ODR systems as deployed in courts do little to reimagine the existing process? Moreover, if courts envision ODR as a default or mandatory process for certain types of cases—so as to create the efficiencies of which we spoke—what will they do to address the 21 million people in the United States who lack adequate Internet access, or with the 44 percent of households with incomes below $30,000 that do not have broadband Internet?12 Addressing these challenges might require public-access terminals or computers in public spaces and community centers.13 It remains unclear whether such substantial costs would be outweighed by the speculative benefits of moving online.14 Once again, rigorous evaluation seems in order.

There are a handful of empirical studies that investigated user experiences in ODR processes, with particular focus on gender,15 hierarchical dynamics,16 availability of information,17 and small-claims dispute resolution in construction cases.18 Many of these research efforts concern platforms operating outside the United States. The body of research we have seen in the United States relies mostly on self-reported pre-ODR and post-ODR comparisons, with data mostly compiled by courts and private platforms—which may be unconfirmed by independent research.

The pre- versus post-comparison is a weak research design, meaning these studies tell us little. Some might think that if we did not have ODR before and we now do, it makes sense that whatever differences we see must be a result of ODR. But the world changes around us all the time. We do not know whether any difference between pre- and post-ODR worlds are due to ODR. Some courts, for example, likely adopted ODR in early 2020. Should one attribute to ODR the huge drop in case-processing volume that many courts experienced around this time? We would not do so, but that is exactly what a pre- versus post-comparison study would do.

Generally, if we implement ODR this year and see an increase in settlements from last year before we implemented ODR, it could be a direct result of the ODR platform. It could also be fewer high-stakes cases being filed, a general increase in filings, more amicable parties, better lawyering, better self-help materials unveiled during that same year, turnover in mediators, and any number of other things. Studies that do not take the whole picture into account rarely produce credible results.

As in-person justice may become less and less the norm, courts should perhaps consider adding remote engagement opportunities. Public safety demands nothing less. But they should also evaluate these systems rigorously. And courts should not be afraid of failure. We can embrace a culture of celebrating failures because those failures allow us to recognize success. There’s reason for optimism and there’s reason for doubt. It’s time to evaluate which is an accurate representation of what ODR can do for courts and the public they serve.


Renee Danser is associate director of research and strategic partnerships, Access to Justice Lab, at Harvard Law School.

Elaine Zhong is a J.D. candidate, 2022, at Harvard Law School.

D. James Greiner is the Honorable S. William Green Professor of Public Law, and faculty director, Access to Justice Lab at Harvard Law School.

  1. See generally, Ayelet Sela, “The Effect of Online Technologies on Dispute Resolution System Design: Antecedents, Current Trends, and Future Directions,” Lewis and Clark Law Review 21 (2017): 635; see also Ethan Katsh and Leah Wing, “Ten Years of Online Dispute Resolution (ODR): Looking at the Past and Constructing the Future,” University of Toledo Law Review 38 (2006): 19, 41.
  2. Joint Technology Committee, “ODR for Courts,” JTC Resource Bulletin (2017): 2.
  3. HiiL, “ODR and the Courts: The Promise of 100% Access to Justice?,” HiiL Trend Report 4 (2016): 37.
  4. Id.
  5. See, supra, note 2 at 7.
  6. Id. at 6.
  7. See, Zoe I. Barsness and Anita D. Bhappu, “At the Crossroads of Culture and Technology: Social Influence and Information-Sharing Processes During Negotiation,” in Michele J. Gelfand and Jeanne M. Brett (eds.), Handbook of Negotiation and Culture (Stanford, CA: Stanford University Press, 2004), pp. 350, 354-57 (discussing that parties’ ability to thoughtfully respond rather than reacting immediately limits conflict escalation in the context of e-negotiation); See also, Dafna Lavi, “No More Click-Click in Here: E-Mediation in Divorce Disputes—The Reality and the Desirable,” Cardozo Journal of Conflict Resolution 16 (2015): 479, 498.
  8. Lavi, supra, note 7 at 502.
  9. Abraham Kuhl, “Family Law Online: The Impact of the Internet,” Journal of the American Academy of Matrimonial Lawyers 21 (2008): 225, 242.
  10. See, e.g., Amy J. Cohen, “Dispute Systems Design, Neoliberalism, and the Problem of Scale,” Harvard Negotiation Law Review 14 (2009) (emphasizing that dispute system design choices should take into account contextual and intuitional variables and particularly power imbalances in the process).
  11. Orna Rabinovich-Einy and Ethan Katsh, “A New Relationship between Public and Private Dispute Resolution: Lessons from Online Dispute Resolution,” Ohio State Journal on Dispute Resolution 32 (2017): 695, 720.
  12. Joyce Winslow, “America’s Digital Divide,” Trust Magazine, The Pew Charitable Trusts (July 26, 2019).
  13. If courts do choose to do this, the authors would enjoy seeing them explore public spaces beyond just libraries, like grocery stores, post offices, and community recreation centers.
  14. Amy Schmitz, “Expanding Access to Remedies Through E-Court Initiatives,” Buffalo Law Review 67 (2019): 89, 160.
  15. See, Martin A. Gramatikov and Laura Klaming, “Getting Divorced Online: Procedural Outcome Justice in Online Divorce Mediation,” Journal of Law and Family Studies 14 (2017): 97 (finding high levels of satisfaction with online divorce procedures and quality of outcomes of both male and female divorcees in the Netherlands, although the former focused more on monetary and time costs while the latter focused on negative emotions).
  16. See, Katalien Bollen and Martin Euwema, “E-Supported Mediation: What Do We Learn from the Field?,” paper presented at the National Mediation Conference: Emerging Dynamics in Mediation—New Thinking, New Practices, New Relationships, September 10-13, 2012 (finding that a hybrid process combining online intake with face-to-face mediation had an equalizing effect in hierarchical labor settings on parties’ fairness and satisfaction perceptions).
  17. See, Laura Klaming, Jelle van Veenen, and Ronald Leenes, “I Want the Opposite of What You Want: Reducing Fixed-Pie Perceptions in Online Negotiations,” Journal of Dispute Resolution (2009): 139 (finding that providing negotiators with incentives independent from the resources that have to be divided, as well as providing them with information about the opponent’s preferences, led to more agreements in online negotiations).
  18. See, Udechukwu Ojiako, Maxwell Chipulu, Alasdair Marshall, and Terry Williams, “An Examination of the ‘Rule of Law’ and ‘Justice’ Implications in Online Dispute Resolution in Construction Projects,” International Journal of Project Management 36 (2018): 301, 311 (finding that the ODR process does not affect parties’ satisfaction with the “rule of law” or “justice” in small-claims ODR in construction projects, while suggesting further research on the cultural contexts around these concepts).