LEGAL
Parties that outsource their ediscovery must take control of the ediscovery management process early. However, doing so presents some interesting and unique challenges for companies that do not have experience or expertise in dealing with the variety of ediscovery vendors. Managing ediscovery comes with a number of challenges:
This article will discuss how organizations should approach the vendor selection process when outsourcing their ediscovery projects to a third party. Key in this process is choosing the right pricing model, as they vary widely from vendor to vendor and can substantially affect overall ediscovery costs.
Choosing a vendor with the pricing model that delivers cost certainty and control for your situation is also a critical part of the vetting process when outsourcing ediscovery projects. There are four basic ediscovery pricing models prevalent in the market today: traditional, fixed fee per matter or per custodian, fixed fee upon ingestion, fixed fee for reviewable content. The following lists what you should know about the structure of each model and their primary advantages.
The traditional model – and the one that continues to dominate – consists primarily of line-item pricing for each activity provided by an ediscovery vendor: collection, processing, predictive coding, hosting for review, and production. Vendors will charge an ingestion fee for potentially discoverable content and processing fees for the various operations performed on this data, such as optical character recognition (OCR), image conversion, deduplication, filtering on search terms, culling, and hosting. OCR has been framed as a major cost element because of the CPU-heavy nature of the process and the length of time that it requires.
Using the traditional model, vendors will charge a hosting fee for each month an ediscovery “job” is being hosted on the vendor’s servers, and often a per-seat fee for each individual that must have access to the processed data. Most traditional models include a production fee, while others charge a premium for predictive coding, although this process is more commonplace and is becoming a standardized offering from many vendors.
The advantage of the traditional model is that each part has been directly connected to a stage in the process, which can be useful for cost-recovery purposes and justifying the costs to a court. The costs, however, will tend to be higher and extremely front-loaded, making this a bad model for smaller matters or matters that are shorter in duration.
The fixed fee per matter or per custodian model charges a fixed fee per matter and/or per data custodian. In a typical case, each custodian requires varying levels of data volume and process complexity, so this model may not provide good value. If there are few custodians, each with a substantial amount of data to process, this model provides reasonable value. If there are a larger number of custodians, each with a small amount of data, this model is worse off.
The advantage of the fixed fee per matter or per custodian model is cost predictability, since there is a fixed fee that is negotiated up front, and the number of custodians is often well known, but there is a good chance that a customer will be paying too much, including for some data that will not be used in a case. Costs can be reduced in this approach by carefully monitoring the number of custodians targeted.
In the fixed fee upon ingestions model, an organization gives a vendor all of its data for a case and the vendor provides a flat cost per gigabyte ingested to manage the data through to the end of the case. Although offering high predictability, this model assesses charges before much of the data may be culled. If a customer will be handing off a substantial amount of “noise” that has not been culled from the data, the customer will be paying for data that will never make it past the initial stage of processing. Moreover, potential buyers have to investigate what “all inclusive” pricing really includes, because there may still be additional charges for production or user fees, introducing new variables – namely, the volume of productions (which can add up in a multi-party case) and the number of users (which can add up based on the scale and size of the review team).
The advantage of the fixed fee upon ingestions model is some of its predictability: Customers know up front how much the case will cost them with respect to overall volume. In-house counsel often prefer predictability over cost savings, and so this model might appeal to that segment of the market.
The last pricing model, fixed fee for reviewable content, charges a fixed fee per document or gigabyte of content, for the duration that the data is actively used in the case.
Important advantages of the fixed fee for reviewable content model include pricing is proportional to only the data in review and only when it is being reviewed. Customers pay only for what will be useful in ediscovery, and they are not required to pay for data that will never be relevant. All of the ediscovery services provided by the vendor are included in a single fee, not charged piecemeal as in some other pricing models, thus introducing more simplicity. Unlike the fixed per custodian or per matter pricing models, this fixed-fee approach bundles easily with the purchase of review, as a price per document is also a unit by which managed document review services is purchased. When technology and review is priced at the same basic unit – the document reviewed – it enables easier end-to-end aggregation and estimation of total cost. Because this model asserts costs proportional to data post-culling, there is a little less predictability in the early stages. As soon as culling operations are completed and the data in review is known, then the cost is much more predictable.
A potential downside to the fixed fee for reviewable content model is that the per-document pricing inherent in this model is typically priced by vendors to include the risk of large and complex documents, and so pricing could be high for cases in which documents are fairly simple and/or small. Also, since this model charges a fixed monthly fee per document or gigabyte, long-running cases (e.g., those longer than 24-36 months) will be subject to higher costs than with other models, although risk can be mitigated by eliminating some data from the system. Most of these risks are mitigated by a highly competitive price point.
Customers should seek to minimize the amount spent on data that is irrelevant and allocate most of their costs toward the information of highest value. In most cases, this will mean that only relevant documents should be reviewed, and managed only for the length of time they are in review. Ediscovery pricing remains apples to oranges. Some potential buyers have reported exponential differences in project cost estimates, making procuring these services a confusing and highly burdensome task, with no real measure for success. It is imperative to carefully weigh the various models, based on the information known to the buyer, and choose the model that best leverages that information.
From a strictly value-based perspective, the best approach to ediscovery project management and scoping – and the approach that should be selected when choosing a vendor that offers the optimum pricing model – will most frequently be one that focuses on content in review during a case.
Ed Sohn spent over two years at Thomson Reuters’ largest office in New Delhi, India, training and managing hundreds of Indian attorneys, technologists, and business professionals. He now focuses on developing and delivering integrated technology and outsourcing solutions for Thomson Reuters legal managed services. Ed frequently speaks and writes on best practices in ediscovery processes and technology. Prior to joining Thomson Reuters, Ed was an associate at King & Spalding LLP.
Reprinted with permission from the Association of Corporate Counsel 2016 All Rights Reserved.
www.acc.com
http://www.acc.com/legalresources/publications/topten/ediscovery-pricing-models.cfm
-->
