Jason Harrop is a founder and CTO of SpeedLegal. SpeedLegal's core offering is an XML-centric document automation solution which is designed to cater for complex variable content, such as is found in contracts and other legal documents. Prior to forming SpeedLegal in 1999, Jason was an e-business consultant with IBM Global Services Australia, and prior to that, a technology/IP lawyer at Australian law firm Corrs Chambers Westgarth. Jason serves on the leadership team of the OASIS LegalXML eContracts Technical Committee, and the Victorian Society for Computers and the Law.
As the marketplace for legal services becomes more competitive, a key challenge for law firms today is what, where and how to invest in order to improve their business. One of the areas where a law firm can make substantial gains is in document production efficiency, in which precedents/forms/model documents play a critical role. This paper presents a way of thinking about precedents, clients, and the bill they receive, which demonstrates that a carefully targeted investment in document automation can yield substantial payoffs in firm performance via increased profits (and profitability as a percentage of revenue), as well as client satisfaction, provided the firm bills for the value of the precedent. The approach advocated supports targeted point investment with particular partners who wish to advance document preparation efficiency in their own practice group, which is a more agile approach than changing the whole firm at once.
It is hoped that this paper will generate an ongoing dialogue around precedent economics, including planning investments in precedent improvement and client billing. A mailing list to support this community of practice has been set up at [email protected]. You may join this list by sending an email to [email protected]
In transaction oriented matters, the key concrete deliverable is usually a document or set of documents which meet the goals of the client.
The document could be as simple as a confidentiality agreement, or as complex as major capital works project documentation or finance documentation.
Matters involving such documents are the subject of this paper.
Where a firm is able to perform the work at an hourly rate, and bill and recover all the time spent on the matter, it ought to be profitable.
Unfortunately, this does not always happen:
- A partner looks at the bill before sending it to the client, and writes off time so that the total is more palatable.
- Or the client complains about the cost, and the bill is reduced.
- Or some time is not recorded in the first place.
- Perhaps the matter was performed for a fixed fee, which worked out at less than the actual billable hours.
Even if the client silently wears the bill, it may have adverse effects on the client relationship.
Then there is the other side of the coin. The work is billed at an hourly rate, and recovered, but since the lawyer started with a good precedent, the bill is smaller than it could have been.
It is likely that the first client will be asked to pay too much (since there is no precedent), and conceivable that subsequent ones will pay too little (unless they are charged for the use of the precedent).
In spite of these difficulties with the hourly rate model, it is often said that because of the hourly rate, efficiency does not matter in law firms.
Nothing could be further from the truth.
The Australian Business Review Weekly magazine recently published revenue, estimated profit, and staff numbers for Australia's top 20 law firms.
So far as estimated profit was concerned, they estimated the profit in 2 bands:
- 35-40% of revenue
- 30-35% of revenue
As BRW quotes a managing partner as saying, "It's all about relative profitability with your competitors – attracting and retaining the best people."
Profit = Amount billed and recovered – cost
Profit is increased by revenue growth, and/or decreasing costs. Profitability (of a given piece of work) is increased by reducing costs (ie increasing efficiency). As BRW put it in a follow-up article:
“Without booming revenue growth, the big law firms have only two ways to increase the all-important partner profit-share – cut costs and/or reduce the number of partners who share the pie.”
Profitability always matters. To paraphrase the managing partner quoted by BRW, you want to be more profitable than your competitors. So efficiency matters.
Efficiency always matters, but it is especially important if your margins are squeezed.
If your margins are squeezed, that is, the amount a client is prepared to pay is going down, then profitability can only be maintained by a commensurate reduction in costs. In a competitive marketplace, you must do all you can to ensure that each transaction remains profitable. As will be seen from the remainder of this paper, an increase in volume or marketshare is a bonus which is likely to follow from an improvement in efficiency.
The question we grapple with is what costs to cut. Too often, costs are cut by reducing head count in a shared service such as IT or KM. While this can deliver an immediate reduction in costs, it is likely to affect internal service levels, and reduce the capacity to take on new projects.
This paper identifies costs which can be cut without impacting on service delivery. Indeed, it identifies ways to cut costs while simultaneously improving service delivery. As we will see, achieving this requires more focused application of existing resources, and is likely to justify additional investment.
This paper approaches the problem of which costs to cut, by looking at costs from the perspective of the costs involved in producing a document.
Here are three related reasons why it is imperative to understand what it costs to produce a document:
- So we can know how much profit we made on a given matter, or, put another way, the minimum we must recover on it in order not to make a loss
- So we can look for opportunities to increase our profitability, by reducing those costs
- So we can price documents rationally, better understand the marketplace, and justify our prices to our clients
The cost components involved in producing a document for a client are as follows:
- Precedent cost
- Labor to deliver first draft
- Labor to deliver final result
- Use of document processing software
- Training costs
- Miscellaneous (rent, electricity, infrastructure including computer hardware/software, LEXIS/NEXIS etc)
In this paper, Precedent means a document which is intended to be re-used as as a solid starting point for the documentation required in a particular class of transactions. Precedents are sometimes also known as "model documents", "forms", "templates", or glossaries.
Precedent Cost means the lifetime costs involved in creating and maintaining a precedent/form, divided by the total expected uses of that precedent over its lifetime.
Generally, it is the actual cost of the firm's investment in the precedent plus some proportion of the overhead of running the precedent department/function.
The Precedent Cost is a production/input cost. Note that the amount (if any) actually billed to the client is something else – namely what I call the “Precedent Usage Fee” (discussed later).
Precedent Cost includes:
- Initial investment in development of the precedent
- Periodic maintenance and review costs
- Labor (by fee earners) to feedback possible improvements to the precedent devised during this matter (since this time shouldn't be charged to the client)
- Part of the overhead of running the precedent department/function.
In this paper, "Labor (or Time) to Deliver First Draft" means the salary costs of the time invested in preparing the First Draft.
“First Draft” means the first draft of a document which the firm sends to the client.
The First Draft is not the first version saved in the firm's document management system, nor the first draft back from a secretary or wordprocessor (WP) operator, nor the first draft a partner looks at. The First Draft is unlikely to be perfect; it may include questions, clarifications, or missing sections.
Labor to Deliver First Draft includes:
- Receiving instructions
- Time to find precedent(s)
- Time to perform research
- Drafting effort
- Quality control (supervision)
The significance of defining the First Draft as being that which the client receives in response to their instructions is that receiving the First Draft back is a milestone which the client can and does measure ("that was a quick turnaround", or "I haven't heard from the firm for weeks...", as well as “this looks good” or “there’s still a lot to do”).
In this paper, "Labor (or Time) to Deliver Final Result" means the salary costs of the time invested in getting from the First Draft to a document delivered to the client which it can sign.
Labor to Deliver Final Result includes:
- further drafting
- the time spent in negotiations
- receiving further instructions
- the cost of preparing, dispatching, and reading comments about interim drafts, and
- quality control (supervision).
Again, Time to Deliver Final Result is a milestone which the client can and does measure (how much time did this transaction take? how painful was it? ).
Costs which fall into the "use of document processing software" bucket represent a portion of the firm's investment in document production tools such as:
- Word processing software (e.g., Microsoft Word)
- Document management software (e.g., Interwoven (iManage), Hummingbird, OpenText), including workflow
- Document assembly/automation software (e.g., SpeedLegal, HotDocs)
- Deal room/negotiation/redlining software
- Voice recognition software
Of these tools, the document assembly/automation software costs should be accounted for (in whole or part) as part of the Precedent Cost.
Training Costs include salary costs of teaching a lawyer how to draft this type of document. I have identified Training Costs as a separate line item, since training is affected somewhat by the strategy discussed in this paper.
Next I look at how these costs may be reduced.
The objective here is to reduce costs without adversely affecting quality or responsiveness. Of the various costs identified above, the ones we’ll look at here are Labor to Deliver First Draft, and Labor to Deliver Final Result.
A key opportunity for cost reduction is to reduce the salary costs in preparing the First Draft. This cost is simply: (salary cost per hour) x (hours spent).
Lets examine the “(salary cost per hour)” first. There are two ways in which this could be lowered:
- Reduce the salary of the person(s) performing the work
At the time of writing, this approach is generally not appropriate, so I don't consider it any further.
- Migrate the work to a cheaper resource
This will be practical where the saving is not outweighed by the increase in the quality control / supervision component. To ensure this doesn't happen, it is important to build know-how into the precedents while keeping them easy to use.
Next, let's consider the “(hours spent)”.
To reduce the number of hours spent preparing the First Draft, you have to reduce the gap or distance between the precedent and the First Draft, or cover that distance more quickly.
- To "reduce the distance" between the precedent and the First Draft, you have to build appropriate options into the precedent.
To do this effectively, document assembly software is a great help. Document assembly software represents the options in a way which a computer can process. This reduces the risk of mistakes which a human may make if they have to manually work through the options and their interdependencies. It also allows the lawyer to work his/her way through the options more quickly. Finally, with document assembly software, you can represent options which are complicated or rarely used, which would be less feasible using a manual approach.
- To "cover the distance" more quickly, you need to look at how a lawyer (and word processing operators) tailors the precedent into a First Draft. These can be broken down into
· amendments to existing clauses,
· re-arrangement/re-ordering of existing clauses, or
· the insertion of new clauses
There is probably not much you can do to speed up amendments and re-arrangements, since (in today's world) these must be done by someone using a word processor.
You can speed up the insertion of some new clauses, by using a clause library. This works best if appropriate options are built into clauses, so a clause can be customised on the fly. For example, a confidentiality clause would be mutual or unilateral depending on which option the lawyer selected.
The major takeaway here is that the cost of getting to a First Draft can be substantially reduced by three initiatives (each of which is facilitated by document automation):
1. making the precedent richer (ie build in appropriate options),
2. accessible to more junior staff, and
3. supplementing the precedents with a clause library.
In addition to reducing cost, the ideas presented above have several key attractions:
- the First Draft is likely to be closer to the final document than it would have been without a good precedent, since you are starting from a better position
- the First Draft is likely to be delivered to the client more quickly (which they will notice)
- the First Draft will not contain the names or details of other clients, as sometimes occurs when one works from a recent example document rather than a properly sanitised precedent.
Before moving on, I touch briefly on some of the other smaller components of this cost:
- Receiving instructions
Giving the lawyer a list of the things about which instructions are needed for that particular precedent should help to reduce the cost of receiving instructions.
- Time to find precedent(s) (search time)
Precedent consolidation - made possible by document assembly software - reduces search time. For example, if you are looking for a confidentiality precedent, instead of trying to choose between a number of different confidentiality precedents e.g., mutual/unilateral, IP acknowledgement etc, you simply choose the single confidentiality precedent, and answer the questions to customise it.
- Quality control (supervision)
I have noted above that supervision costs are reduced by improving the precedent document.
The keys to reducing Labor/Time to Deliver Final Result are:
- A closer First Draft
- A more efficient negotiation procedure
As noted above, the ideas for cutting costs associated with getting to the First Draft have the side effect of giving you a First Draft which is closer to the final document.
A more efficient and effective negotiation procedure is a very important way of reducing the cost of labor to get to the Final Result (and reducing the pain perceived by the client). This is a major topic in its own right. It suffices for now to note that there are two major aspects to improving the negotiation procedure:
- software support for managing and tracking changes to various clauses, and the reasons for those changes; improvements to how those changes are circulated to the client, and the other side, and how feedback is received;
- soft negotiation skills, which help you to reach a better or equivalent result in less time.
The analysis in the preceding section suggests that one of the major opportunities for cost reduction in document preparation lies in reducing Labor to Deliver First Draft. That is, through appropriate use of document automation technology, the labor involved in producing the First Draft can be reduced while simultaneously improving the quality/completeness of that document, and delivering it to the client more quickly. Importantly, we will see shortly that this can also translate into increased profits.
The following diagram indicates how the costs of preparing a given document ought to change over time.
- total cost decreases significantly over time
- the proportion of the cost involved in preparing the First Draft is greatly reduced on an ongoing basis (how much exactly will vary from document type to document type – for example, it may be greater for a construction agreement than for a confidentiality agreement)
- there is an initial increase in precedent investment as the firm invests in improving the precedent; this is then followed by maintenance.
- the cost involved in preparing the final result has also decreased, since the First Draft is closer to what is required. The costs which remain there reflect value-add, but also the fact that we haven't attempted to improve the negotiation procedure (this being outside the scope of this paper).
The secret to achieving this significant and sustained reduction in cost is to use technology so that your expenditure can be an investment in a continuously improving re-usable precedent product, rather than a throw-away cost in the repeated re-invention of a document which is not easily leveraged.
If we are successful, we will be spending less time on each matter, while delivering higher quality documents to the client more quickly.
Because our costs are going down, the firm is more profitable provided the client is billed the same total amount.
In fact, the firm could choose to reduce the bill somewhat to share some of the cost-savings with its client, showing that the firm appreciates that budget constraints / cost control is the most frequently cited management issue cited by Chief Legal Officers. This approach allows the client to reduce its legal spend over time, without in any way compromising quality of service, and may also help the firm to gain market share at the expense of other firms.
However, as the reader is no doubt aware, if the firm:
- continues to bill its clients by time expended,
- does not increase its hourly rates,
- does not bill for the value of the precedent,
then, because the time expenditure has decreased, it has effectively passed the entirety of its cost saving on to its client. This is a recipe for lower revenues with no improvement in profitability.
To avoid this result, and to achieve the desired increase in profitability, the firm must bill the client for the value of the precedent.
In this paper, “Precedent Usage Fee” means the amount the firm bills the client for the use of a precedent.
Worked examples of calculating the Precedent Usage Fee are included below.
The following diagram shows a healthy trend in the components of the bill for preparing a given document.
Figure 2: Revenue Components of Bill over Time
1. the total size of the bill is (or can be) reduced somewhat
2. the AS-IS Time to First Draft is roughly the same amount as the TO-BE Precedent Usage Fee plus the TO-BE Time to First Draft
3. the Precedent Usage Fee is larger than the TO-BE labor component for preparing the First Draft.
In summary, the key idea is revenue replacement - you replace a portion of the revenue generated by billing time for some piece of work (which might yield a profit of around 40%), with revenue by billing a Precedent Usage Fee (which ought to yield double the profit) instead.
Quoting for Work - Implications
Rules governing professional conduct typically require a firm to provide details of the method/basis of costing their legal services.
So, although a rational client ought to care only about the total, and the quality of service provided for that total, the rules applicable in your jurisdiction are likely to require you to say up front that the bill will be based in part on a fee for precedents used, unless you are charging a fixed fee.
You might consider a fixed fee for the delivery of the First Draft only. See the FAQ: “How can I justify the Precedent Usage Fee”?
“Investing in a precedent” includes planning what will be in the precedent (features/scope), taking the time to draft the basic clauses, and building options and know-how into it, then review and sign-off. It also involves maintaining the precedent, including responding to legislative change, and soliciting feedback, evaluating it, and incorporating it where appropriate. From an accounting point of view, the investment also includes a share of overheads, such as necessary IT infrastructure.
The key question then is what to invest in developing a particular precedent?
- If (as often happens) a firm under-invests, it will experience higher costs and will be less profitable and/or less competitive.
- If you over-invest, your ROI will be smaller (or even negative).
One example of over investing is building options into the precedent which aren’t used frequently enough to justify the time spent in constructing and maintaining them.
Optimal investment in each precedent is critical to law firm performance.
Before you can work out what to invest in a particular precedent, you have to be clear on your business objectives.
For example, these might be:
Business Objective 1: Realise a higher profit on transactions of type XXX than before (in absolute dollars, and as a percentage of revenue)
Business Objective 2: For transactions of type XXX, deliver the following measurable improvements to the client:
· 40% reduction in Time to Deliver First Draft
· 20% reduction in Time to Deliver Final Result
· 5% reduction in total bill
The sample business objectives refer to particular types of transactions, since the actual percentages are likely to be different for each different precedent (ie type of agreement).
Let’s look first at Business Objective 2, the three improvements measurable by the client.
Calculating the Precedent Usage Fee
If the Precedent Usage Fee is to make up for the reduction in profit from time billed, it can be calculated from the following figures:
o Average “as-is” Time to Deliver First Draft (for this type of agreement)
o Average Time to Deliver Final Result
o charge out rate,
o profit on charge out rate
See the example below.
The first sanity check is to ask yourself whether the client will bear the amount you need to charge for the precedent having regard to the type of transaction.
Now let’s turn to Business Objective 1.
Knowing how many transactions the precedent will be used in over a given period of time (say three years), and the requirement that profitability increase as a result of this initiative, set a maximum on the amount you can invest.
After working the numbers, a second sanity check is to ask yourself whether the time reduction targets are realistic based on the automation planned, and whether that level of automation can be achieved with an investment of not more than the maximum amount you have calculated you can invest in improving the precedent.
All the better if the necessary improvements can be implemented for less than this maximum amount.
Consider documenting a transaction for which the market will bear a cost of $10,000.
Assume that starting with an existing simple precedent, we can currently do the work for $12,500.
Breaking the bill down:
average charge out rate = $300
Time to Deliver First Draft = 15 hours = $4,500
Time to Deliver Final Result = 25 hours =$7,500
If the profit on hourly rate = 40%, we make $4800 profit.
Note that this assumes we can find clients who are prepared to pay $12,500 in a market which generally bears $10,000. Even though we make $4800 profit, we still can’t lower our charge to $10,000 simply by reducing the total unless we cut our hourly rates or write off time, since it is actually taking us 40 hours to do the work.
Calculating the Precedent Usage Fee
As stated, our business objective is to perform the work and bill $10,000 or less.
This can be achieved if through precedent improvement, we can:
reduce Time to Deliver First Draft to 5.25 hours (35% of previous) = $1,575
reduce Time to Deliver Final Result to 17.5 hours (70% of previous) = $5,250
charge Precedent Usage Fee: = $2,775
With these fees, we perform the work and bill $9600 (ie $400 under the competition).
To achieve these quicker delivery times, suppose it was necessary to invest $20,000 (mainly in labor) in automating the precedent in year 1 (and smaller amounts in years 2 and 3), for a total of $27,000.
Assuming the precedent is charged out 50 times per annum, over a 3 year assumed lifetime, the Precedent Cost works out at $180 per use.
On this basis, we now make $5,325 profit, an improvement of $525 per matter.
Across 150 documents over 3 years, this is an additional $78,750 profit, and we are now delivering at a competitive price.
In the above example, it was necessary to invest $20,000 in automating the precedent.
What does this actually mean?
The major component of precedent investment is labor - planning what will be in the precedent (features/scope), taking the time to draft the basic clauses, building options and know-how into it, review and sign-off, and ongoing maintenance. Investment in IT infrastructure is comparatively small, when amortized across a typical precedent portfolio.
So how do we cost time?
In practice, fee earners will not be doing precedent work if there are transactions they could be working on. So it makes perfect sense to use their salary cost as opposed to their charge out rate for the purposes of converting this dollar figure into a number of hours.
There are three ways the model could treat the time lawyers now have for other work:
§ Optimistic Approach
Assume that it will be used for other work and recognise the revenue billed as a benefit of precedent automation.
§ Conservative Approach
Assume that it will be used productively (hopefully) for other work, but don’t count on this as being other than revenue neutral
§ Pessimistic Approach
Assume that the lawyer will do nothing other than draw a salary in the time saved
The model presented in this paper takes the Conservative Approach. Any profit on the time saved is an extra benefit not accounted for in this model. However, if some proportion of that time could not be used productively for something else (i.e. if there was no other work to do), then the cost of that non-productive time would need to be taken into account.
The direct benefits of the approach advocated here are:
o increased profit and profitability (as a percentage of revenue), and/or competitiveness
o happier clients, through:
o possibility of reducing the size of the client bill (while retaining increased profits)
o faster delivery of higher quality First Draft
o faster delivery of Final Result
The indirect benefits include:
o freeing up lawyers to work on other things (without reducing revenue)
o lawyers avoid mundane repetitive error-prone treadmill work, and concentrate on what really needs their intellect if it is to be done
To use this approach, we need to know the following figures:
- average time to deliver this type of document
- the average charge out rate
- the profit on those blended charge out rates
- how many transactions the precedent will be billed in over a given period of time
For the purposes of the calculations, the number of times a given precedent is likely to be billed can be estimated by an informed partner, who may choose to analyse existing transaction documents in order to come up with an answer.
The estimate ought to come from the partner sponsoring the development of a particular precedent, since the estimate is part of the partner's business case. I call this the “Volume Commitment”.
In this paper, “Volume Commitment” means the number of times a given precedent is likely to be billed over its lifetime, as stated by the relevant partner.
Once the precedent exists, the precedent system ought to be able to track precedent usage. It should be linked to the billing system.
I suggest that you trial the approach advocated in this paper with a handful of suitable documents from one or more practice groups.
Whatever document you choose, it is critical that there be a partner who is championing this approach for that document. In fact, it is best to start with a partner who is already impatient to improve the process in their practice group (“Sponsoring Partner”). You don’t need to start with a firm wide initiative, although that would be feasible provided the Knowledge Manager has the relevant resources and partnership support.
This approach works with both relatively simple high-volume documents, and more complex documents (see What about Complex Transactions?).
Nevertheless, you might want to base your first Document Product on a simple high-volume document in order to familiarise yourself with the process and to start to get a feel for what inputs to the model are realistic.
As a first step, you should work with the Sponsoring Partner to create a Business Plan for the chosen document.
This document-specific business plan should:
· articulate in general terms the rationale behind automating this document, including which clients it is intended to please;
· include AS-IS metrics (actual or estimated):
· average Time to Deliver First Draft / Final Result
· size of average bill
· current profitability
· historical usage volumes
· interview the lawyers and describe in prose the factors which contribute to those average times, so you can identify the parts of the document which would benefit most from automation
It is critical that the lawyers who have day-to-day responsibility for drafting and negotiating this sort of document are involved in both the planning of the improvement effort (ie the preparation of the business plan), and its execution. Unless this happens, the improvement effort will miss the mark.
· summarise the main variations (features) the automated document will cater for, the expected lifetime of the document, and its maintenance requirements
· include target TO-BE metrics which are achievable based on the AS-IS metrics, the interviews, and the planned features of the automated document. For example:
· 50% reduction in Time to Deliver First Draft
· 20% reduction in Time to Deliver Final Result
· 5% reduction in total bill
A phased approach is recommended, so that the planned improvements are delivered in part, measured for 6-12 months, and then delivered in whole
· calculate the proposed Precedent Usage Fee, and sanity check it
· include a high level project plan for the automation work, including dates and resources
· estimate the investment of time and effort (and by whom) which is necessary to achieve the quicker Time to Deliver targets
The Sponsoring Partner will need to ensure that these resources can and do commit to doing the work in the time frames specified. In some cases the Sponsoring Partner may need to smooth the way for this to happen by re-prioritizing the other responsibilities of those resources.
· include the Volume Commitment
· confirm that profitability improves based on the required investment and Volume Commitment
· calculate expected Return on Investment
A SmartPrecedent which will prompt you with the questions necessary to automatically generate a first draft of such a business plan for a particular precedent may be found on the SpeedLegal website.
In order to implement the plan for a document, you need:
· a financial model (spreadsheet) which allows you to calculate what to invest in the precedent, and the Precedent Usage Fee
· a suitable document assembly/automation system, such as SpeedLegal or HotDocs
It is critical that the system you choose be capable of setting up the automation which is necessary to achieve the targeted quicker Time to Deliver metrics, and easy enough to use that the labour and skills to do so are available. The project will fail otherwise.
· preferably, billing codes which allow you to separately track Time to Deliver First Draft, Time to Deliver Final Result
· a billing system able to bill the Precedent Usage Fee
· resources to develop the precedent as per the plan, and billing codes to track that investment (at cost).
Ultimately the precedent department/function (however constituted - including knowledge managers, librarians, professional support lawyers, and practice automation/IT analysts) should develop into a center of excellence for document product:
- work closely with transaction lawyers and marketing to plan investment in particular precedents, including targets for the three things clients can measure (ie Time to First Draft, Time to Final Result, amount they are billed)
- develop rules of thumb for the investment likely to be required in order to achieve the targets (for example, a typical improvement project might require 3 times the number of hours currently taken to prepare a First Draft for a client)
- project manage the development of the precedent, on time and within budget
- track and report actual measurements against targets
- improve usage forecasts
- document best practices for document production.
This paper has identified three key indicators of efficiency which a firm can't help but expose to its client.
I have shown how an appropriate investment in precedents improves efficiency by each of these measures.
Using the approach suggested, we can measure:
- our investment - how much a precedent costs to produce and maintain,
- how much we charge for it (the Precedent Usage Fee), and
- how many times it will be used over its lifespan.
Based on numbers for the precedent portfolio as a whole, we can calculate:
- the firm's total investment in its precedents
- what that investment will generate, expressed as ROI, or absolute contribution to firm profitability in each year
either as forecast or actual.
Tracking these numbers will be of great interest to both the partnership (as it is part of the value of the firm), and the precedent department.
This paper has presented a systematic approach for increasing the efficiency of the document production process, focusing in particular on measuring things clients care about, like quality and timeliness.
By thinking clearly about documents as a product to be delivered (albeit a tailored one), and using document automation technology as a manufacturing process to replace a significant portion of the labor component with a precedent usage fee, the client receives higher quality of service at lower cost, and the law firm enjoys greater profits (and profitability as a percentage of revenue) whilst at the same time positioning itself to react to changing market forces.
It is worth saying up front that the client benefits not just from faster service (and a possible cost reduction), but also from higher quality and the reduced risk which comes from the re-use of well-tested material.
Clients will be aware of this, and for these reasons, in favour of precedent-based work. However, they will still pressure you to lower the fee.
If you don’t charge a fixed price, at least for the First Draft, a prospective client is likely to ask you to justify the Precedent Usage Fee, with a view to paying less.
Most firms would be reluctant to justify the Precedent Usage Fee by giving the client a copy of the precedent.
However, you might consider:
· allowing the client to look at it at your premises
· explaining to the client that all of their instructions are addressed by the precedent, except for things specifically itemised (which will appear in the First Draft, charged at hourly rates)
· keeping statistics as to the amount of effort invested in the precedent (eg “this precedent is version 72, which is the result of 400 hours of effort by 15 lawyers over 4 years”)
Alternatively, quote a fixed price for the matter (or just delivery of the First Draft), so that the client is oblivious to the breakdown between time and the Precedent Usage Fee.
The point of this paper is to identify ways to make the firm more efficient (and therefore more profitable) by finding ways to reduce the costs of document preparation. Figure 1 in Implications of Cost Reduction shows how these costs might go down over time.
I've suggested that some of these cost savings could be passed on to the client.
The result is that a client buying a particular agreement in 2006 should pay less for it than if they bought it in 2003.
This result is entirely consistent with the natural dynamic of document pricing, which is that a document becomes less valuable over time (even as it improves).
The chart below shows this for some common documents:
A number of factors contribute to this trend:
· when a document is first introduced (due to some innovation or regulation), expertise is relatively scarce
· there is an initial green fields demand for the document which, once satisfied, is followed by a (smaller) replacement or incremental demand
· when a document is first introduced, example documents are hard to come by
· how to do a particular agreement spreads rapidly through the legal community since the document goes to the other side's lawyers for "review" (a network effect), and it has proven difficult to prevent "re-use"
· industry sectors tend to standardise their key documents
· For optimal profitability, it is best to position the firm to supply new documents from the beginning of their lifecycle, when demand and price are highest
· It is important to automate as early as possible, since that is when the the profits can be harvested. That’s because later, competitive forces are likely to lead to price reductions.
Is it possible to stop the documents from depreciating over time, or at least slow down the depreciation?
· It is not enough to add features to the document
Unlike consumer products (and software), adding features won't arrest the decline, since you don't sell a document by enumerating its features (for example, “we’ve got a better termination clause”). The closest you can get to this is to say that "the precedent addresses all your instructions [with the noted exceptions]".
Even though adding features won't arrest the decline, you have to do it anyway, since that is how you reduce the cost of Labor to First Draft. As previously noted, to manage these additional features, document assembly is very useful.
· Prevent other firms from copying your work
It is assumed that procedures are in place which prevent precedent documents from "walking out" with disgruntled or misguided employees.
It is more of a challenge to prevent other firms from borrowing from documents they have received during negotiations. If you wish to discourage this, you need prominent copyright notices, and you need to ensure that your precedent system records authorship of clauses and documents. You might also consider the digital rights management technologies introduced in Microsoft Office 2003.
The reality is that we can expect to see faster depreciation in the future, as the internet contributes to rapid commoditisation. This paper will help you to position your firm to prepare for this.
Example of Volume Sensitivity?
Consider a scenario where the firm must significantly reduce its bill if it is to compete in some area.
Suppose the market price for a Superannuation Fund is $270.
Suppose further that the firm currently charges an average of $540 for this sort of transaction (double the market price).
The relevant AS-IS data:
Time to Deliver First Draft = 2 hours, at $150 per hour
Time to Deliver Final Result = 1 hour, at $240 per hour
volume: 1 per week, over 3 years
profit on hourly rate = 40%
profit is $216 (although we are at risk of losing the work)
In this example, our objective is to make a profit if we bill a total of $270.
This can be done with:
First Draft: 0 (we spend no time, through client self-service)
final draft: $120
Precedent Usage Fee: $150
How much profit do we make? That depends on how much we invest in automating the document, and whether we can improve the volumes (ie amortize the cost across more documents).
The following table shows some examples:
Investment (year 1-3)
Volume (total over three years)
150 (current volume)
750 (since we’re price competitive now, and can leverage our brand to win more business)
You can see how important it is to get the investment and projected volumes correct. The margin for error in the amount invested becomes much less important if you can increase the transaction volumes (compare the figures for volume of 150 versus 750).
With a total investment of $33,750, and volume of 750 (5 per week over three years), we make $153 per document. That is less per document than the $216 we used to make ($32,400 over 150 transactions), but the total profit on the increased volume is better ($114,750 on 750 transactions).
We can do even better if we refine our offer a little further.
We might consider two offerings:
1. Something similar to the one already described ie including some labor for the final draft.
The example above included $120 charge for labor for the final draft. Since Feest is not providing any manual drafting for $270, this is actually a value-add.
2. A completely automated offering, where the client serves themselves via the firm extranet, and there is no review by the firm.
This is equivalent to Feest’s product, but assuming we still charge $270, we make more because we don’t incur any labor cost associated with producing the final draft.
Put another way, we charge $270 for automated output, and extra if the client wants to talk to a real lawyer.
The same technologies which help the firm’s lawyers to produce documents more efficiently can, in principle, be used to allow clients access to self-service document creation.
The model presented in this paper can also be used to help ensure that such services are profitable for the law firm.
Some points to consider:
- provided the service is priced properly, self-service can be more profitable than traditional legal services, since the labor cost is avoided
- because the labor cost is avoided, the service scales well. Look to offer documents for which you can grow the demand
- however, if the objective is to avoid labor (in order to assure immediate turn-around), then most (if not all) of the options in the precedent need to be completely automated. This suggests a higher degree of automation in certain areas (and therefore investment) than is necessary for documents which the firm’s lawyers complete themselves.
- it is therefore sometimes more effective for the firm to “finish off” the document manually; to this extent, customer self-service amounts to a way of receiving instructions.
- be aware of your target audience. Is the client a lawyer or lay-person? If they are a lawyer, are they familiar with this sort of document? These considerations will affect how you word the questions (and indeed, what options you offer), and the degree to which a document designed for an internal audience can be re-purposed for a client audience.
- it is prudent to ensure that your document automation system supports customer self-service, if only because one or more key clients will demand it sooner rather than later. This implies that the document automation system has a web-based interface with support for whatever browsers and word processors your customers are using, usage measurement, and appropriate workflow features.
As the margin pressure example suggests, with greater volumes, you can afford to invest more, since you get your money back more easily.
This is a real advantage of a large (or specialist) law firm over a smaller one. A larger or specialist law firm can afford to invest more to create a better precedent, not because it has more resources (although it does), but because the precedent will get used more. And a better precedent means of course the bill includes more of the profitable Precedent Usage Fee, and less of the relatively less profitable billable time.
Of course, there is a limit to what you can sensibly invest in a document. For example, a confidentiality agreement is inherently less complicated than a distribution agreement, which is less complicated than an outsourcing agreement or certain finance documentation.
But, where document complexity can be high, volume is critical to making investment attractive. And so, for complex documents in particular, it is important to drive growth in volume.
Publishers, such as Wolters Kluwer, who offer law firms and the public the ability to generate particular documents via the web have a volume advantage. If their precedents are used by many firms, then the high volumes will mean they can invest aggressively in having the best precedents, which in turn will lead to more use as the reputation of their documents gets better and better.
It is sometimes said that it is not possible to measure precedent usage accurately, since lawyers can access the precedent once (which is measured), but then store it locally for re-use in subsequent transactions (which can’t be measured).
There are a few things to note about this:
- this behaviour doesn’t in itself invalidate the model presented here, since the model relies on a certain level of measured usage. “Pirate” usage is unfortunate, but not fatal - provided the Precedent Usage Fee is billed in the volumes committed to by the Sponsoring Partner.
- however, it is clearly unfortunate, since the unmeasured use is a use of the precedent for which the client is not billed, and is therefore lost revenue and firm profit.
- building variable clauses and other automation features into the precedent should reduce the tendency for lawyers to keep a copy in their “bottom drawer”, since it is likely to be more work to edit the bottom drawer copy to meet the needs of the transaction, than it is to answer the automation questions to generate a tailored copy.
- suppose the Precedent Usage Fee is $1000, and the Precedent Cost is $100. If the lawyer was credited some amount up to $900 against their budget for a billable use of the precedent, then they would use it. Indeed, it would be prudent to make the billing of the Precedent Usage Fee and the budget credit something which must be approved by the responsible partner.
For reasons which should now be clear, smaller firms in particular should make a “buy versus build” decision when the need for a new precedent is identified. This paper provides a simple way to determine whether the price a third party is charging for a “smart” precedent is such that buying it as needed is more sensible than building one internally.
In complex transactions, precedent documents and precedent clauses may only get you part of the way there. You may need to write the rest from scratch.
How far the precedent documents get you really depends on what you’ve put into them. This is perfectly acceptable. It reflects the fact that some text which is necessary for a particular transaction is used rarely enough that it doesn’t pay to incorporate it as a complete part of the precedent. Put another way, it is more economical to finish it off as called for in the course of a transaction.
This is also consistent with a “start small” approach, where you partially automate a document, see how it goes, and if demand for a particular variation/option justifies it, invest in automating that later.
It is worth emphasising that even with short documents, partial automation may be all you should do.
In summary, the model espoused here works well for complex transactions – for complex documents and for transactions involving multiple documents. The percentage reduction you realise in Time to Deliver First Draft will probably be lower (or achieved more gradually) but the model will guide you as to what is worth pursuing.
For simplicity, this model has been presented in the context of a precedent document, rather than a precedent clause.
However, similar principles apply to the development of precedent clauses.
The additional wrinkle is that a precedent clause can be used in many precedent documents, and also directly in a one-off client document.
If the client is to be billed for a precedent clause, then at what rate? And how do you work out what to invest in the first place?
The answer to these questions is that as the precedent department starts to get a feel for the cost and value of precedent documents, it will become confident to work out (estimate) these numbers for precedent clauses.
This paper has distinguished between preparing a First Draft, and preparing a final draft, and noted that a key component of the bill for preparing the First Draft ought to be the Precedent Usage Fee.
There is a risk that, once high quality documents are available online, clients will come to law firms with their own first draft, saying “We’ve already created a first draft ourselves, so we don’t need your precedent, and therefore don’t need to pay the Precedent Usage Fee”.
Of course every lawyer has experienced a client who presents them with an unworkable document and asks them to “just check it quickly”, or “fix the worst problems”.
That was okay in a world where the document was bad and one billed purely for time. In the scenario envisaged here, the document will be good, and so the work you are asked to perform is likely to be less.
The most effective response to this threat is clearly to charge a competitive Precedent Usage Fee. If you can do this, then there is no reason why you wouldn’t offer web-based self-service, so that the client who presents you with a “first draft” is presenting you with your own document, which they’ve paid you for.
 Schmidt, “Law in the Partner Jungle”, BRW Magazine (http://www.brw.com.au/), September 25-October 1, 2003
 Schmidt, “Lawyers in Limbo”, BRW Magazine (http://www.brw.com.au/), December 4-10, 2003
 Case Study: Firm A has a stated strategy of only going after high value work. Why? Because other work is less profitable, so that if a partner does it, the average profitability per partner is lower. Why not employ a solicitor, para-legal, or computer to do it? The risk in only doing the high value work is that it is too easy to ignore your costs, which means that over time, you will become inefficient relative to other firms. The brutal reality is that it is not difficult for an efficient (read boutique) firm to attract high calibre partners (based on high per-partner profits) to start doing high-value work (often for clients they have brought with them). It may be a greater challenge for a large firm focused on high-value work to become efficient enough to profitably perform lower value work, or indeed, to recognise that it needs to perform its existing work more efficiently. Often this realisation hits home when competition bites.
 In “Knowledge Management: Measuring Success”, published in June 2003 in "The Business of Partnership", and available at http://www.legalweek.com/ViewItem.asp?id=15487, Matthew Parsons explains how the personnel dedicated to implementing Jones Day’s strategic KM plan were made redundant, a pattern which he says is being repeated in other US firms.
 The actual cost of the firm's investment in the precedent are the salary costs directed at developing the precedent which weren’t billed to and recovered from some client. That is, the Precedent Cost accounts for time spent only at the firm's expense.
 Note that there is obviously a difference between the actual number of hours spent preparing the First Draft, and the time which elapses before the client receives it (this is turnaround time). Reducing Labor/Time to Deliver First Draft will help to reduce the turnaround time. If you wish, you can measure “Attentiveness” as the total number of billable hours spent on a matter, divided by turnaround time (measured in Business Hours).
 Third party studies are starting to measure this metric as well. For example, the Aberdeen Group’s “The Contract Management Benchmark Report” (June 2003), contains the finding that it takes an average of 20 to 30 days for a company to create, negotiate, and finalise a contract, after the initial sourcing cycle. That report does not however break out figures showing the effect of law firm assistance on cycle time.
 Building know-how into precedents is an important supplement to other forms of training. One of its key advantages is that it is "just-in-time", that is, delivered exactly when it is required.
 The inadvertent inclusion of the names of other clients in a document invites a client to lose confidence in the firm:
1. “you charged me a lot for a slap dash search & replace, and didn’t even read the resulting document carefully”
2. “you have poor attention to detail, and poor controls in place to prevent my sensitive information leaking out”
 It is often very helpful to review the first draft marked up against the precedent (once the appropriate options in it have been selected). This also helps in identifying possible improvements to the precedent.
Training, document processing software, and miscellaneous costs are not shown, since these may remain fairly constant.
 Association of Corporate Counsel & Altman Weil Inc, “2003 Chief Legal Officer Survey”, page 10, available at http://www.acca.com/Surveys/closurvey/2003.pdf
 This desirable state of affairs will be reached most quickly with common documents with relatively less variation. It may take longer to achieve with other documents, if ever. While we are looking generally to increase the precedent contribution over time, there comes a point for any document when the effort of incorporating yet another option into the precedent outweighs its usefulness. See the What about Complex Transactions? FAQ for further discussion.
 For the typical approach in many US states, see Rule 1.5 (b) of the ABA’s Model Rules of Professional Conduct; for England and Wales, see paragraph 4(f) of Solicitors’ Costs Information and Client Care Code 1999; in New Zealand, the requirement in Rule 3.01 of the Rules of Professional Conduct for Barristers and Solicitors is that the fee be fair and reasonable. In Victoria Australia, see section 86 of the Legal Practice Act 1996; in New South Wales, section 175 and 177 of the Legal Profession Practice Act 1987.
 Sometimes you won’t know up front which precedents you will need to use. However, if the firm has a fixed Precedent Usage Fee for each precedent, it will be possible to say that any precedent used will be charged for at the current price at the time. Whether it is necessary or desirable to say that is left open here.
 I have defined Precedent Cost to include the labor to incorporate improvements to the precedent devised during this matter, and included this. For simplicity in this example, no net present value discount rate is used.
 It is not absolutely necessary to distinguish between Time to Deliver First Draft and Time to Deliver Final Result, since the Precedent Usage Fee can be calculated and charged without doing this. However, it is strongly recommended, since it gives you invaluable visibility into customer outcomes. For this reason, it is best if the billing system provides billing codes which differentiate between time spent on the First Draft, and time spent on the final result.
 If you take a phased approach, this fee will be different in each phase.
 In this paper, I have assumed the fee for some particular precedent will always be the same, irrespective of which options/alternatives are selected. Clearly it would be possible to charge different amounts depending on the options selected, but in most cases the additional complexity this would introduce is not warranted.
 A simplified version of this approach is to ignore the distinction between first and final drafts, instead measuring total drafting time. This means that you don’t need to introduce a new billing code for preparation of first draft, but does make it significantly harder to measure improvement.
 In fact, (server-side) automation offers good protection here, since lawyers don’t retrieve the entire precedent (with all its options and complexity in a takeaway form), but rather, just a particular view of it (determined by the options they have selected).
 The price at http://www.feest.com.au/, as at the time of writing.
 For three reasons:
1. it will take longer to build up the precedents to include all the options which are required, and while some are missing, those parts need to be drafted in the traditional way
2. some parts may be so unique to this transaction, that they never make it into precedent clauses
3. various other things may be billed under this billing code which can’t be avoided
This paper Copyright © SpeedLegal, 2003-2004.