Introduction to Discovery (6/7): The key to controlling costs lies in the quote check (Part 1)

It is common to get a "phase quote" in every business situation, but in reality, there are not many cases in the discovery industry that lead to a phase quote.

The reason is that different vendors have different quote formats, they don't know how to read the quote, and they can't compare.It's not that there aren't many competitors.Rather, the discovery market continues to expand in proportion to the amount of data held by companies, and new discovery support vendors and vendors are being created one after another.This seems to be the number one reason why companies can't control the cost of discovery well.

Is it a daily occurrence that there is a difference between the estimated amount and the billed amount?

In general business, whether it is a manufacturing or sales business or a service business, it is rare that the estimated price and the actual price are significantly different.If the actual billing amount is significantly higher than the estimate, it will cause trouble for the person in charge of ordering, and the contractor may be pointed out that the planning at the estimation stage is not easy, so it is a little. If it is over, it will be stored at the estimated amount.Even if it is expected that the estimated amount will be exceeded due to unavoidable circumstances, it is normal for the ordering side to speak to the ordering side and obtain permission.

However, in the discovery industry, it often happens that the quoted amount and the actual billed amount are very different.This indicates that the quote is only a "reference price".In many cases, we compared the prices of multiple vendors at the estimation stage and chose a cheaper vendor, but when we finally looked at the billing amount, it was more expensive than other vendors.

Moreover, it is difficult for companies to even judge whether the amount billed in this way is appropriate.The fact that the discovery fee is charged through the intervening law firm instead of being charged directly by the vendor is another factor that makes it more difficult to check the contents.

Even if you have doubts about the amount billed, you cannot ask a vendor other than the vendor who outsourced the business after the discovery, "Is this amount really reasonable as a third party?"However, it is also unrealistic to ask the Legal Department of another company that has experienced the proceedings, "How much did your company cost for discovery?"As a result, discovery costs often start and end in ambiguity.

Key points to get a correct quote

When the other party receives the complaint and decides to respond, the lawyer will probably ask you to "take measures against discovery."Upon receiving a request, the legal department of a company will request a quote from a discovery support vendor, and will select a vendor based on the amount, content, achievements, etc.

Vendor selection is done by the lawyer himself (not exactly the lawyer, but the internal department of the law firm, Litigation Support, to which the lawyer belongs).However, it should be noted here that it is very risky to leave the selection of this discovery vendor to an American law firm or a US subsidiary.

The reason is that for companies that have no experience in discovery, it is difficult to understand the breakdown of the quotation and the company that selects the vendor, so it is difficult to judge, so the case where the law firm selects the vendor and contracts at the asking price There are many cases.It would be nice if the vendor's work of choice was of high quality, but if it wasn't, the deal would be higher than the market price and the discovery would be of poor quality, which would have a disadvantage in the proceedings. I can't even see it.

In the following, we will introduce a "view of estimates that will not fail" in order to prevent such failures.

High quality discovery vendors have fine quotes

Example 1 is a quote actually used by a discovery comprehensive support vendor.For legal professionals who have never experienced discovery, this may be the first time you've seen it.

We'll go into more detail later, so it's a good idea to start by choosing your own discovery vendor and take note of your own findings and questions.Make a note of your own interpretations and questions.

After that, if you look at the explanation below, it will be easier to understand what kind of mistakes you are likely to make.

Items such as "Data Collection", "Data Process / Analysis", and "Data Hosting" indicate each process of discovery.If the cost is calculated for each process in this way, it is easy to understand the cost ratio of each work to the total amount, even if it is an approximation.

Keep in mind that the "review cost" is the most expensive of all processes.This review is the most important process in discovery, and if you are familiar with it, you can judge that "If you reduce this review cost, the overall cost will change significantly."

The discovery process collects documents, analyzes them, finds out if they are evidence, and submits them.In other words, most of the objects are "documents".Therefore, it is common for quotations to always be calculated on a "unit price basis".

However, when it comes to calculation per document, it is difficult to find out the exact number, so in reality, we count it as "per computer" or "per hard disk".Alternatively, because the capacity of each data is different, it may be calculated as "per GB".

Vendors with a proven track record of discovery know how much work to do depending on the size of the company.The "estimated amount" is calculated there.After all, the total cost of the discovery estimate is calculated by "unit price x estimated amount".

Many vendors do not show this estimated amount and offer a quotation with only "unit price". When asked "Why isn't the total cost calculated?", You'll usually get the answer "I can't calculate it because I don't know how much data I'm going to discover."

If you receive this answer at face value, you can understand that "the less data, the less the total amount, and conversely, the more data, the more."

When that happens, the first thing legal professionals think of as a way to reduce costs is to reduce the amount of data.However, if you inadvertently try to reduce the amount of data and delete what you originally had to submit as evidence, you may be punished as a concealment of evidence, so you have to be careful about that. Don't.

In addition, the answer "I don't know the expected amount of data" can be regarded as "there is little discovery record" or "I don't want to disclose it as an estimate" from a disappointing point of view.The rougher the calculation method of the estimate, the more excuses can be made even if there is a gap between the estimated amount and the actual billed amount.

In particular, you must be careful about items that are calculated per hour.For example, when estimating the review amount, vendors with the data "Review 1 documents per hour" are conscientious.However, if the day and evening are not clear, you can even "intentionally raise the amount over time."

In addition, in the case of vendors with low work quality or vendors who do not support Japanese, garbled characters may occur before the review, and it may be necessary to go back to the stage of the data process and redo the work. I will.Many vendors openly charge the ordering party for redoing work due to such "vendor's own mistakes".

That is why it is a prerequisite that the estimate is calculated by "unit price x estimated amount".If you are a vendor with a proven track record in discovery, it is not difficult to estimate the estimated amount from the size of the company.

In the estimate given at the beginning, the item of expenses incurred separately is described as "(Note)".This is also a good general estimate, but some vendors don't give it at all.

For example, there are options that should be explained in advance, such as notes such as "Documents created in Japanese will be recalculated by a separate estimate" and "Analysis of emails sent and received via mailers will be charged separately". I often hear that the amount of money was added from one to the next, resulting in a large amount of money far from the estimate originally presented.

In many cases, the estimate does not include tasks such as hosting costs, load file creation, and upload costs, which will be described later.In addition, additional work to deal with special file format processing is often not included in the cost, so companies dealing with such files are advised to check carefully.

How to read the estimation by work process of discovery

So far, I have briefly explained the basics and prerequisites for viewing a quote, especially how to compare when making a phase quote with another company.From here, I will explain the details according to the estimation example given earlier.

(1) Data collection

Data collection is, in a narrow sense, the reproduction and collection of data that can be evidence.This example estimate summarizes the process from preparation to duplication and collection.The preparatory process is the identification and maintenance work of the day and evening.

• Hearings will be held as the first step of identification and maintenance of the day and evening.Starting with identifying the departments and parties involved in the proceedings, it is necessary to ask the person in charge of the information system not only about the system and file format, but also about the data storage rules and regulations.
• It is necessary to discuss the maintenance procedure and obtain the consent of the parties concerned so that the data will not be overwritten, lost, or tampered with.
• It is also necessary to create a "data map" so that you can immediately check which person has what kind of science and what kind of data is stored in which media in later discovery work.

To do this data collection, the vendor needs to send an engineer to the company.The work often takes several days, in which case transportation and accommodation costs will be added separately.In addition, the "estimated amount" in the estimate is the labor cost for the target device, and the hard disk for storing the duplicated data and the hard disk for backing it up may be charged separately.

(2) Data process

The data process is the "data processing" work that analyzes the collected data, extracts the necessary information from the electronic data, and creates a database.

The amount of data preserved and collected in the previous process is enormous, and it is impossible for the plaintiff and the defendant to view all of it.Of course, it also contains a lot of data that is not related to the proceedings, and we need to quickly screen out those unnecessary documents.

Data process tasks include seven types of tasks related to data culling and filtering and their peripheral tasks (preparatory tasks, etc.).

(XNUMX) to (XNUMX) shown below are preparatory work for (XNUMX) and subsequent data analysis.Culling is the process of excluding data that has nothing to do with it in advance, and the easiest way to understand it is to remove the program configuration file and OS configuration file.

Since the contents of the compressed file cannot be checked as it is, it should be expanded and checked so that the same file can be checked by multiple parties or stored on multiple recording media. Also excludes them.After narrowing down to some extent, text information and metadata information are extracted, and organized and stored for each index created.

Currently, most of the information disclosed in Discovery is electronic data.Including data stored on multiple PCs, servers, mobile terminals, and archived e-mail data creates a huge amount of information.As mentioned above, if you print out the data for one computer, it will be the document for about four 1-ton trucks, so if there are 2 custodians (data holders), the document is purely for 4 trucks. Will be the target.

It is virtually impossible for lawyers and staff lawyers to look at all of these documents and determine whether to use them as evidence.Even if it were possible, paying a lawyer would be a huge expense, plus a daunting amount of time.

It is also clear that disclosing data that is not related to proceedings is disadvantageous in terms of corporate strategy.Information about new products should not be published before it is released.

That is why we use advanced IT technology to shake off documents that are clearly unrelated to the proceedings.This sorting work is the "data process", and the accuracy of this work is very important because it affects the accuracy of the subsequent discovery process.

(3) Data analysis

Data analysis analyzes the data prepared in the process so far and prepares to review the evidence data.Even with detailed data process work, non-litigation-related data still accounts for a large proportion.Therefore, companies and lawyers select keywords, and discovery support vendors cooperate to provide technical advice.

After that, by performing a more advanced keyword search to identify and extract the target data, we will narrow down to only the necessary data.This is a task called "analysis".

There are three main tasks that occur here.

If the analysis is performed properly and the necessary data can be narrowed down, the accuracy and efficiency of the review, which is the next process, will increase, and the discovery itself should proceed smoothly.

What you have to pay attention to here is whether the vendor is processing for Asian languages ​​as in (XNUMX).

If it does not support Japanese, problems such as "documents cannot be seen as garbled characters" or "documents are not sufficiently narrowed down" occur in the review process, and problems such as the review not ending as planned occur. Is also possible.

With an environment where you can read Japanese
The environment that "can support Japanese" is different

It is also important that it supports Japan's unique application culture.For example, it is thought that many companies use Outlook made by Microsoft for mail software (mailer), but in Japan, "Becky!" Made by Rim Arts and "Shuriken" made by Just SYSTEMS are used. There are many companies.

Although e-mail information, which is a communication tool with the outside, has a very important meaning in discovery, if it does not correspond to such a mailer unique to Japan, the accuracy of process work will be significantly reduced. I will end up.

In many cases, the estimate does not mention that, and it is still conscientious if it says "Japanese files are charged separately", but it is possible to support Japanese files in the first place. For example, it is better to confirm whether it is within the range of the estimate and whether it also supports applications unique to Japan.

(4) Data hosting

Data hosting is managing the data used in the review on a designated server so that it can be viewed by the review tool.

In the past, the vendor with the lowest unit price at the phase estimation stage had about three times as much hosting as other vendors.The amount of money varies greatly depending on what kind of data you are hosting at what stage.

Unfortunately, the discovery industry does not yet have a global standard for discovery process control.The general idea is that no matter what stage you host, it's okay if you can disclose the evidence as a result.Therefore, the "standard quotation" will not appear forever.Each vendor has a different basis for calculating costs.

Introduction to Discovery (1/7): Don't leave the choice of discovery vendor to others!　As mentioned in the above, the calculation method differs depending on the vendor, whether to charge the data capacity before decompression or the data capacity after decompression, and even if the unit price is cheap, the data capacity will increase several times after decompression. The swelling project is a prime example.

Does the legal counsel requesting the discovery have such awareness?It is dangerous to compare only the unit price if it is simply written as "hosting".Because it is an industry without clear guidelines, it is necessary to look at the total cost, not the unit price.

Recently, an increasing number of companies are hosting the collected data as it is even after the discovery is completed.This is because if a proceeding or an investigation / investigation develops into a cross-border case, the same data may be used in the discovery of another proceeding or investigation / investigation.Therefore, companies that are likely to host continuously should also ask the vendor for the hosting fee that will be charged after the discovery is completed.

(5) Project management

For discovery cases in most international proceedings, it is advisable to have about two project managers in charge of the project.The role of the project manager is to think about search keyword setting with the legal staff of the company, create a document batch for review, and support smooth project progress by intervening between the company and the law firm. I will.

Based on FRONTEO's experience, project managers spend about 1 hours a month on a single project.However, it is necessary to note that transportation expenses, business trip expenses, etc. that are incurred in connection with this will be charged separately, so if you request an overseas vendor, these expenses will be incurred.

Example 2) Project management quotation

(6) Review fee

It is no exaggeration to say that "discovery cost = review cost", and review cost accounts for a large proportion of the cost.As you can see from the quote, most of the review costs are labor costs.

Therefore, recently, an increasing number of companies are introducing automatic computer analysis called "predictive coding" in their reviews.

FRONTEO developed predictive coding that can analyze Asian languages ​​using artificial intelligence, and when it was used in actual projects, it was 3 minutes of the assumption of other companies compared to the case where Americans conducted review work. I was able to process it in one period.In addition, since there is no labor cost, the cost was reduced to one-fifth, and the customer's evaluation was high.

Whether or not to use predictive coding is at the discretion of the vendor and varies in accuracy.When checking the quote, it is highly recommended to check the precision and recall rates of predictive coding, as well as whether or not you are using predictive coding.

By the way, many discovery vendors should be reluctant to offer this precision rate, recall rate.If you have data with a track record of discovery, you should have a sample, but it is also true that there are many cases where you do not give out the excuse such as "I can not give a rate because Japanese is difficult".This is because the rate is directly linked to the review cost.

(7) Production

Production is the work of submitting evidence.Once the review has selected evidence documents, the attorney-at-law will decide which evidence file to submit to the other party.The vendor processes the data specified by the lawyer into a reliable file format (Tiff format) as evidence, and creates an English translation file if necessary.

Depending on the case, about 3% of the files to be reviewed will be produced.If there are 9 review files, 2,700 of them will be produced.

Previous article:Introduction to Discovery (5/7): Technology of Discovery Vendors As Important as Discovery Know-how

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