“Wild ducks make a lot of noise, but they also have the sense to benefit from occasionally flying in formation.”
Team building requires an optimum mix of skills and capabilities. With several hundred so-called e-discovery vendors littering the landscape, how should one go about selecting the right one?
Though they sound alike, e-discovery vendors are as different as a bicycle shop, a car dealer and an aviation technician, all of whom can claim to be in the “transportation” industry. At the earliest stages of a case, it is important for client and counsel to honestly assess gaps in their own skills inventories and determine what kind of assistance will be needed to pursue a successful course of action.
Large and complex discovery projects have always benefitted from a team approach. Why then is a “new” team required for electronic discovery? In the past, litigation support vendors provided services that became commoditized – for example, scanning at a per-page price, and coding at a per document price. If electronic discovery is regarded in the same way – “What is your best price per Gb?” – then many of the best practices set out in the Sedona Canada Principles will have been violated. E-discovery requires expertise from across many disciplines, an overall strategic understanding of the objectives of the matter, and a uniquely collaborative approach when it comes to opposing counsel, investigators, expert witnesses and judges.
In Canada today we do not have a common legal definition of what constitutes “electronic discovery”. Some lawyers and judges may confuse the inspection of a hard drive for production, and many lawyers believe that “native file review” means reading e-mails one by one in Outlook. In such an environment it is no wonder that comparing one vendor with another is so challenging. A comparative pricing grid may assign cells to “de-duplication” or “culling” but how each of these processes is actually done can be as different as commercial fishing is from buying fillet at the supermarket. Though the result of both activities is “fish,” the costs and timing of each are very different indeed.
To make a good decision on vendors, one must start with an accurate definition of the project objectives (“halibut dinner”) and a clearly stated specification of what output is required (“four dinners with salad and dessert, no wine.”) Fretting about price per gigabyte (or kilogram!) is far less important than understanding the nature of the deliverable. To risk mixing metaphors – consider the following RFP: “We own 20 acres of waterfront and would like quotes on the design and construction of a cottage. Please indicate your hourly rates for each subtrade, and the cost of materials. We would like a big and beautiful cottage. We need the cottage finished in a month but we are very cost conscious. We will base our decision mostly on hourly rates.”
Obviously, in this situation the hourly rates are not very important compared to factors such as: what size of cottage will be built? What will the design be like? Will it be built according to code? What will the monthly heating cost be once it is built? Is there a five or ten year guarantee of quality materials and workmanship?
At least in construction there IS a building code. Not so in electronic discovery, where every law clerk with Summation iBlaze can do their own “e-discovery” by importing PST files. In the absence of a common e-discovery code, the onus is on the buyer of services to determine what level of design suitability, quality and safety are required. This can often be a matter of negotiation with the vendor. One of the downsides of an adversarial-style RFP process is that the ability to negotiate with a vendor – as you would negotiate with an architect or a builder – may be minimized or even lost.
1. Do you need help up front with planning, strategic or specialized legal advice?
If so, you need an experienced electronic discovery legal consultant. Electronic discovery legal consultants may practice law, may offer their services as independent consultants, or may be employed by full service electronic discovery service bureaus. As in any situation where retaining an expert is involved, make sure the consultant you retain has the knowledge, experience and maturity to make a meaningful contribution to the team effort.
2. Do you need help with preservation and capture of electronically stored information?
Many costly mistakes are made at this stage of a litigation or investigation. Data capture can be a stand-alone requirement, but more often, data is being captured for the purpose of early case assessment, culling and review. There are individuals and companies that offer forensic capture services, but by retaining such a company you may have prematurely decided that forensic imaging is the best approach to preservation. Bear in mind that servers are almost never captured forensically, and back-up tapes are not forensic copies of server data. Therefore forensic imaging of 100 hard drives may in many cases be too intrusive, costly and slow.
Another issue to consider is that specialized forensic companies offering capture services use applications developed for the criminal investigation market that may not be compatible or desirable in the later stages of the civil litigation discovery process. For example, searches performed in EnCase and reported in HTML format cannot simply be exported into a review tool. FTK images and EnCase images may use different hash code algorithms, making de-duplication impossible or unnecessarily difficult. Finally, companies that specialize in forensic imaging may not be able to offer other capture technologies even if these would be appropriate, defensible and cost effective. A full-service electronic discovery vendor can provide different kinds of capture services depending on the circumstances of the case, and all data captured by a full service company will be seamlessly integrated into its further processes.
3. Do you need help with culling out irrelevant documents before review?
Over the last few years several US start-ups have offered concept and clustering software to the electronic discovery marketplace in an effort to reduce the time required for culling and review, and to improve the accuracy of Boolean and proximity search methods. Some of these companies were funded by venture capitalists and were established from the beginning with the goal of being sold as quickly as possible for as much money as possible. Several of these high profile companies have already disappeared because their products never delivered on the overblown promises of their marketing departments.
Others have had modest success but all of them are in the developmental phase. Often the search algorithms implemented in the software are hidden to view and some doubt has been expressed as to their defensibility. Finally, these systems are generally hosted in the USA, making them risky or illegal for most Canadian based litigation and prices can be in the thousands of dollars per gigabyte.
Another issue to consider is that for some of the leading concept search engines, all data must first be processed. While this may reduce the time for review, it will also increase the time and cost for processing. In the right case, of course, the right tool can do wonders-but not every culling tool is right for every case. Rather than paying directly for the service, it is best to work with a vendor who has access to multiple software tools. It is usually best to pilot these tools with a large sample of data and to determine how compatible the culling process will be with review and production workflows.
4. Do you need help with review?
Some sort of processing is required to perform review, and a review platform is also needed. Processing is a very challenging, complex set of procedures. It is not comparable, say, to importing email files into the Summation e-console. Processing by a service bureau may include:
- Anti-virus checking and quarantine
- File type exclusion
- De-duplication and near duplication
- Extraction of full text
- OCR of image files
- Extraction of metadata into structured fields
- Bates numbering
- Enhancement of metadata (normalization)
- Conversion (for example Netscape or other legacy systems)
- Export to review platform or litigation support software
- Tracking of source for each record
- Quality control and validation
- Detailed file-level reporting
“Native file review” does not mean opening email messages one at a time in Outlook. It means using a litigation support software tool to search, sort and tag documents that have been processed by not converted into a fixed image format such as TIFF or PDF. That stage, if required, is also a service that can be provided either within the review platform (“TIFF on the fly”) or separately by the service bureau, as part of the production process.
Moving large amounts of data from one process to another requires attention to detail, large amounts of storage and the skill and expertise to manage the data properly. It is not advisable to switch vendors midstream, which is why e-discovery vendors should ideally be viewed as part of the team, in place for the long haul.
5. Are you concerned about privacy and security?
If you are not, you should be. All electronically stored information must be handled with the utmost care and safeguarded against loss or intrusion. Of the hundreds of vendors out there, only a handful can claim to have invested in the kinds of IT security infrastructure that would be acceptable to a financial institution, large business of any kind, or even a top-tier law firm. Security must include physical, personnel and IT, and is implemented through a combination of policies, procedures and audits, as well as hardware and software tools.
The privacy issue is becoming one of the hottest topics in Canadian electronic discovery because all businesses here (including law firms) are subject to PIPEDA. Given the various risks of cross-border data flow, including search and seizure by customs agents, the intrusive provisions of the Patriot Act, and the fact that the US is not compliant with the European Union Directive[1] (as is Canada), it is critical that Canadian ESI stay in Canada in most cases. Unfortunately, this does limit one’s choices because there are only a handful of top-tier service providers who keep all data in Canada at every stage of the project.
6. Do you care about quality?
Lawyers who would never send out a document with a single typo, and firms whose precedents are drafted word-by-painstaking word by practice group committees, sometimes forget that quality is also an essential part of the e-discovery vendor deliverable. In e-discovery, quality should refer not only to the work product itself – the review database – but to every aspect of the process including chain of custody, project management, exception reporting, processing, dealing with anomalies, file conversions, data integration, culling, near-de-duplication, creation of load files, and web hosting. Quality means a validated and defensible work product. It means a high level of comfort that money has been spent on a work product that can be reused after the litigation is over.
7. Do you have paper?
In our zeal to handle e-discovery we sometimes forget to ask about the humble paper collection. In their scramble to enter the electronic discovery market, most new vendors have no way to handle paper, or they partner with a scanning company. That could work if the scanning company is oriented to litigation and the partnership is well-established. But if not, your best bet is a company that combines experience with paper and electronic evidence, so the evidence can be de-duplicated and structured in a single database.
Checklist for Vendor Selection
For a helpful analysis and a tool for drafting a thorough RFI, reference should be made to “Navigating the Vendor Proposal Process: Best Practices for the Selection of Electronic Discovery Vendors.”[2]
When selecting a vendor in Canada, though technical capabilities and price are important, there are other factors that are critical to reducing cost and risk, and improving the chances of success for any project. Risks are managed not only during the vendor selection process, but in the establishment of a commercial relationship, through effective contracting and the clear communication of roles and expectations.
For example:
- Is the vendor merely expected to provide a service or should it be expected (and able) to contribute meaningfully to the team effort?
- Does the vendor understand the legal and business risks of sending data outside of Canada for culling, processing or hosting?
- Does the vendor understand the sanctity of solicitor-client and work-product privilege (and in the case of government, cabinet secrecy) or is all data simply treated as data?
- Is the vendor’s work product compliant with all emerging rules and standards in every applicable jurisdiction?
- Does the vendor have documented procedures at every stage of the e-discovery process so that its work can be validated and defended if necessary?
- Is the vendor oriented to American or Canadian-style litigation practice?
- Does the vendor have a policy about conflict of interest?
- Can the vendor demonstrate an enterprise-wide commitment to quality processes?
- What happens to your data when the case is over? Does the vendor have its own client document retention/destruction policies?
- Does the vendor have general business and also professional liability (errors and omissions) insurance coverage?
- Has the vendor invested in an IT infrastructure that can meet or exceed the security requirements of the most discerning organization?
- Is the vendor offering a one-size-fits-all software solution or a range of suitable options depending on the circumstances?
- If the vendor’s web-hosting system goes down on the eve of production, how quickly can it be restored? And how much data will be lost?
- When exchanging productions, can the vendor accurately handle conversions from and to multiple data formats?
[1] “Directive 95/46/EC on the protection of individuals with regard to the processing of personal data and on the free movement of such data.”
[2] Prepared by the RFP+ Sub-Group of The Sedona Conference Working Group on Electronic Document Retention and Production.
