Introduction to Discovery (7/7): Questions to ask in order to determine the vendor's ability to respond to discovery

If companies cannot take the initiative in advancing discovery, they cannot control costs.As mentioned above, it is desirable for Japanese companies to contract directly with a discovery vendor.

Therefore, I have briefly summarized what you should always ask when selecting a vendor and negotiating, the ideal answer to it, an example of an answer to be noted, and the reason.

The eDiscovery process is divided into information governance, identification, maintenance / collection, processing / viewing / analysis, creation, and submission processes, as specified in the EDRM model.When choosing a vendor, you need to know where and how far these processes can be performed.

Most large vendors will say, "I'm working on all the processes."But don't take this for granted.You should also be aware of the possibility of subcontracting to a subcontractor vendor. When checking the "coverage of discovery", it is also a good idea to check whether you are subcontracting or outsourcing to a subcontractor vendor.It is recommended that you include the sentence "Be sure to obtain the written consent of the ordering party when subcontracting to the outside" in the contract.

Vendors who answered "2. Only part of the work" in the example answer to note may only use general-purpose tools for even that "partial work" or throw it to other companies.Changing vendors depending on the process is prone to mistakes and there is a risk of information leakage, so you should choose a "discovery comprehensive support vendor" that can handle all the work consistently.

When outsourcing to an American vendor, the question is where to do the process and review work.In most cases, it will be done in the United States, but doing so will require more translation costs and time to do so.If you do the process and review in Japan, you don't need that cost and time.

You should also pay attention to the data management perspective.It is the same as giving up data management when you take it overseas.No matter how much you have an NDA with a contract vendor, if you are subcontracted, you will not know who is viewing the data and it will be a risk for the company.

This tooIntroduction to Discovery (6/7): The key to controlling costs is the estimate check (Part 2)As explained in detail in, if Japanese is not supported, data will be garbled, process accuracy will be reduced, and wasteful translation costs and time will be incurred, resulting in inefficiency.

Please note that even if the company "supports Japanese", the discovery work may not support it.

For example, if you have an office in Japan, or if you have a staff member who can understand Japanese in the United States, you will be answered that you can handle it.However, what is important is not whether you can speak Japanese, but whether you have a high track record and abundant experience in discovering Japanese companies.

It is no exaggeration to say that "discovery" = "review", and the ratio of review cost to the total discovery cost is large, and in order to control the cost well, it is important to make this process as efficient as possible.Predictive coding plays an active role there.

Predictive coding is computer-processed pre-review with near-human accuracy.By introducing this technology, it is possible to significantly reduce the cost and time of review work, which used to rely entirely on the human eye.

Predictive coding is becoming so essential to the efficiency of reviews that there is no reason not to use it.Vendors who nevertheless don't use it either don't have the technology or deliberately do a "human eye review" to increase profitability and choose a cost-prone method.Vendors you don't know about like this should be excluded from the options.

Demonstrations are the most effective way to know the accuracy of predictive coding.

If you have sample data for one PC, you can demonstrate and check the accuracy of work in one day.The vendor's "tuning" power is also important.In the demonstration, the results may vary depending on the sample data collection method.In such a situation, I would like to find out how to perform high-precision work while communicating with the person in charge of proceedings on the company side, and how to respond to it.

In addition, whether or not the recall rate, precision rate, etc. can be submitted immediately before the demonstration is also a point to determine the vendor's responsiveness.This is because it is impossible to make an accurate estimate unless the supporting vendor or the person in charge of the vendor understands and understands the numerical supporting data.

It is a waste of money and time to translate when the evidence documents have not been narrowed down.Some vendors want to translate all of the collected material before the process is done, but most ideally, only the evidence to be submitted is translated.

Even if that isn't possible, it's a good idea to know when to translate.You can trust any vendor who says they will translate after the review.

(* PDF image files and paper materials may be subject to OCR)

If you convert the document to Tiff and apply OCR, you can prevent the data from being garbled, so it is a method that the just-named "Japanese-speaking vendor" wants to take.However, since the accuracy of OCR is limited, there is no doubt about extracting reliable evidence, and if you are a vendor with high-precision Japanese analysis technology, you do not need to use OCR. Can also analyze the data correctly.

Many vendors may not be able to answer clearly when asked for review speed.As for the reason, be careful if you get an ambiguous response such as "I don't know how long it will take because Japanese is complicated" or "I don't know because it depends on the document".

This is because they often want to spend as much time as possible and incur extra costs.Since it is directly linked to the review cost, which accounts for most of the sales for them, if you show a clear number, you will know the necessary cost, and it will be difficult to charge a lot of cost over a long period of time, so that is Their true intention is to be as blind as possible.

Even vendors who provide decent numbers should check the number of files processed per hour.Vendors who have accumulated know-how to carry out reviews efficiently may be able to review 60 files per hour even for complex materials.In some cases, it was possible to increase the number of files to 120.

American lawyers and law firms often want data.At various stages of the process, people say "I want you to send the data to the United States," but you don't have to follow this.If you can store it in a data center in Japan, you can proceed there. (As mentioned above, sending data to a place that your company cannot control increases the risk of information leakage.)

Asking lawyers why, he says, putting the data in the United States improves the impression of courts and the Justice Department.By presenting the documentary evidence (although there are other data), the aim is to impress the Japanese companies that they are supportive of the investigation.

This is not a mistake.The fact that the data is under the jurisdiction of the court and the Ministry of Justice creates a situation where you can investigate at any time.

But what if American and European companies are in a proceeding? The EU's "Data Protection Directive" rule stipulates that moving data to a third country that does not have a sufficient level of data protection is NG.Since there is no comprehensive law for the protection of personal information in the United States, the movement of data is basically prohibited.If an American lawyer asks for it, the lawyer himself may be punished.This is often called out at seminars, so American lawyers know it in common sense.

Nevertheless, if the reason is that "bringing data to the United States will favor the proceedings," European companies are always at a disadvantage.This doesn't make sense.

Also, in cross-border cases, the same data may be used as evidence in multiple proceedings.It is easy to imagine that one file is used in both the proceedings A and the proceedings B.

Regardless of the scope of the proceedings, a huge amount of data is now collected for each discovery.It is wasteful in terms of cost and time to discover the same document for each case.That is why "cross-matter management" is efficient because it is managed in a data center and can be taken out for each case.

Sending data to the United States or hosting data on an American server is the exact opposite of "cross-matter management," and I have to say that it goes against the times.

Previous article:Introduction to Discovery (6/7): The key to controlling costs is the estimate check (Part 2)