Is document collaboration in law firms just too “old-school”?

Jack Shepherd
13 min readSep 17, 2020


There are two primary models of document collaboration in law firms. The first is the old-school method of sending out a Word document, receiving comments back, and typing changes into the live document. The second is Google Docs-like co-authoring, where contributors all have access to the live document and make changes simultaneously.

I have lost count of the number of times I have heard people criticise lawyers for being out of date on document collaboration. In this article I look at why lawyers work as they do, and whether their processes are really as out of date as some think they are.

Speed read: I believe the old-school “typing in changes” model of collaboration still has a place in legal teams for some exercises, e.g. when working to collate comments from large numbers of specialist advisors. More modern models such as Google Docs also have a place, eg. among teams looking at a document from the same perspective. The key is to know which to use, and when. The tools now available to lawyers mean that the biggest blocker to improving document collaboration is, as always, culture.


Old-school document collaboration, aka “typing changes in”

The old-school method involves multiple contributors receiving a copy of a Word document. Each person makes their comments whichever way they want (handwritten, typed, track changes, over a call etc.). A single person reviews the changes, and if appropriate, types them into the document in order to progress it. That person has a name: the person holding the pen (I also refer to them as a pen-holder in this article). They are more than just typists; they serve a control function over what goes into the document.

Old-school document collaboration


Many legal projects involve facts that change incredibly quickly. Some will involve issues from numerous angles, such as tax, employment, accountancy, competition and foreign laws. All of these facts and legal angles come together in complex document collaboration processes.

The pen-holder is closest to the facts. The people providing input on the different angles are called specialists. They often relying on the pen-holder to keep them updated on the facts. Compared to pen-holders, specialists are usually spread more thinly over many different projects.

The person closest to the facts acts as the control mechanism

The old-school method allows one person with general oversight (i.e. the pen-holder) to weave the different legal angles together in a way that avoids conflicts and is appropriate to the facts. Having a pen-holder allows a legal team to respond to questions about a document quickly and easily, without passing queries between a cast of thousands. The alternative is often seen as chaotic, e.g. where a tax associate makes changes that conflict not only with the facts but also with essential edits made by other specialists.

But specialists are not always involved in these exercises. For example, some documents only involve three members of a litigation team. Here, everybody is usually up to speed with all the facts and the risk of making changes that conflict with each other is minimal. Indeed, making conflicting changes might facilitate discussions to resolve key points sooner. In these situations, the old-school method of having one person responsible for a document rather than the team seems so…old school.

The old school method seems to offer less where the team as a whole is responsible for a document

Key question: “do I need one person acting as a control function?”.

Version control

When specialist input is required on a document, lawyers tend to see this as obtaining a set of “approvals” or “sign-offs”, rather than true collaboration. This is usually for risk and compliance reasons, or just because this is how teams are used to working.

For example, an international law firm that doesn’t operate in Norway will want to avoid taking responsibility for Norwegian law; that risk should stay with their Norwegian lawyers. Another example would be a standard protocol requiring the formal approval of a tax advisor before a legal team sends a document for signature.

When the old-school method is used, you are usually emailing a Word document to people for their approval. That document won’t change until you send another one. It means it’s very easy to isolate approvals to a given version of the document.

Here, tax and competition lawyers are signing off on the same first draft

The fact that approvals can be pinned to a particular version makes people happy from a risk perspective. For example, they know that when the tax team gives their approval on the document sent out yesterday, it’s fine from a tax perspective. But if the document had changed since it was sent to the tax team, they approved an older version. That gives the pen-holder less comfort than if the tax team had seen all the changes.

Making changes after approval reduces the comfort given by sign-offs and approvals

Where you are not working with specialists, the need to work in this way varies — particularly when members of the same team are working together on a document (e.g. restructuring partner and restructuring associate).

In these situations, I have worked in teams where I have been expected to obtain partner approval before a document is sent out. I have also worked in teams where I have sat around a computer with a partner, making changes together. If you are working in the latter way, the granular version control offered by the old-school method seems less valuable.

In my view, working to get “approvals” from others is not really collaboration in the proper sense. This is not meant in a negative sense, because sometimes working in this way is necessary. But too often, I see legal teams defaulting to this way of working even where there is no obvious risk or protocol forcing them to do so. Such teams may wish to consider working differently in order to make sure the document is progressed efficiently, and everybody is heard.

Key question: “is there some risk or compliance reason why I need to obtain approvals and sign-offs from people, or is this just how I am used to working?”.


The big weakness of the old-school method is that it relies on one person to type the changes of others into a document. This is a slow and error-prone process. It is not the best way of extracting value from highly-trained lawyers.

Having said that, the reliance on “typing in changes” is what gives the old-school method its control function. For example, when a pen-holder blindly “accepts all” changes from a specialist without reviewing them, they are abdicating their control responsibility. That means errors and inconsistencies slip in. Reviewing what others have changed in a document may also serve an important learning and development function.

The ideal outcome here balances the control mechanism with an efficient way to ensure approved changes are reflected in the document. Indeed, this is how software development teams usually work (see, e.g. GitHub). My main wish is that the control function would be less laborious than manually typing in somebody else’s comments (again, see, e.g. GitHub).

Key question: “how onerous will it be to type others’ changes into a document, and is it worth doing this in order to gain more control?”.


Co-authoring — “like Google Docs”

Co-authoring is the opposite of the old-school method, because everybody is looking at the live version of a document rather than separate copies. As a result, all changes are visible to everybody else as soon as they are made. This way of working is available in a number of document platforms, including Office 365 (i.e. the latest version of Word).



As noted above, there are situations where it is easier for the team to take responsibility for a document rather than an individual pen-holder. These are most often the situations where everybody has the same grasp of the facts, and are looking at the document from the same angle. Co-authoring is helpful in these situations because it reduces the burden of sending endless drafts around the team, and somebody having to type them in.

There are also advantages to using co-authoring for documents that are are product of lots of other documents joined together (composite documents). For example, a due diligence report might be composed of hundreds of “questionnaires” compiled by numerous different people. These are often joined up into one document. If you use the old-school model, everybody has to work on a separate document, leaving a mass document assembly exercise afterwards. Again, it’s error-prone and slow.

Old-school: composite documents have to be completed separately and then assembled

In contrast, adopting a co-authoring model here all but cuts out the assembly step because everybody is making changes directly in the composite document. This is another situation where everybody is looking at things from broadly the same perspective, each with a similar level of factual understanding.

Co-authoring: completely cuts out assembly step with composite documents

But some lawyers are uncomfortable with co-authoring.

Firstly, the specialists. When making edits live in the document, they may be scared that others can see their incomplete work product, “warts and all”. This may happen, for example, if somebody takes a break before finishing their edits to get a cup of tea. Or it may happen if a tax trainee needs a tax partner to review their work before it is considered to be complete.

Contributors on documents are scared of showing off their incomplete work

If people are scared about these issues, they have two choices: (i) copy the document, work on that and type the changes into the “live” document when done, or (ii) establish a working culture that makes people feel like they are not being judged on incomplete work products. Or stop being a perfectionist.

Secondly, the pen-holder may feel inherent discomfort with co-authoring. They may wish to exert control over the document by forcing contributors to use track changes to input their comments. That allows the pen-holder to approve or reject changes as appropriate.

But this is not always ideal. For example, if many specialists are likely to be amending the same sections, comments can become intertwined. It can be very challenging for the pen-holder to “see the wood from the trees” and understand why changes are being made. In addition, track changes-like interfaces can be notorious for introducing rogue spaces and typos — see, e.g. the example below.

Perfectionism and the need for control have a chilling effect on the adoption of co-authoring in many situations. However, these become less important when collaboration occurs between those with similar perspectives and a common understanding of the facts.

Key question: “is the team happy that their work in progress will be visible to all, and do I need more control over the changes people are making?”.

Version control

Because edits are being made simultaneously, version control in co-authoring is based on an entirely different model. Whereas the old-school method bases versions on completed sets of edits per contributor, co-authoring bases versions on the time at which each individual edit is made.

Teams wishing to obtain sign-offs and approvals on a particular version will find co-authoring challenging because the document is constantly changing. That makes it difficult to pin down which changes a foreign law firm has “approved”, as opposed to those they haven’t seen yet.

With the old-school model, pen-holders may be used to isolating all the changes made by specialist teams into one copy of a document. That helps them explain to others why the document has changed. Co-authoring does not easily cater for this, particular when contributors make comments that depend on changes others have already made, e.g.:

Two rogue “s” letters appear because one contributor was editing on top of another’s edits

All of these concerns could be addressed if you use the old-school method to re-circulate the document once all changes are made. Possibly, you would have to use a document comparison to “flatten” all the changes to make sure you can get a clear sign-off. Even better, if you do away with the need to operate in a way that requires approvals and sign-offs, you might be able to make do with the co-authoring versioning model (and help teams become more communicative in the process). Whether or not this is possible really depends on the team’s culture and risk appetite.

Key question: “how strictly do I need to obtain sign-offs or approvals on a particular version, or is this just how I am used to working?”


The obvious practical issue to call out here is that co-authoring requires all contributors to have access to the same document management system. This can be a problem with large businesses, as numerous privacy, security and legal hurdles will have to be overcome to enable this. For this reason, where co-authoring is used, it tends to be internal to an organisation.

The more significant practical issue relates to team cultures. The old-school method is the incumbent, and we know that old habits die hard.

For example, co-authoring challenges a lawyer’s need to present anything less than a “perfect” end-product. Many lawyers (especially senior ones) are reluctant to look at anything unless it is considered to be a “final” version of a document, despite them having the ability and time to do so. When this happens, the team is effectively driven into the old-school method, as the entire process becomes sequential, with only “perfect” drafts being exchanged.

These behaviours can have consequences. They result in significant time and effort being expended on issues that could have been dealt with at a much earlier stage. These occasions are sadly common occurrences in some law firms. They have a significant effect on team morale, as people stay late and feel like they have wasted time.

In this scenario, at least 5.5h are wasted because a senior lawyer wasn’t prepared to look at an early draft. The diagram also shows the knock-on effect on the associate’s morale

Other cultural barriers around co-authoring include lawyers not having a sufficient incentive to change the way they work, feeling uncomfortable using new technology, and preferring to make handwritten comments on printed documents. I have seen these kinds of issues successfully navigated when teams discuss their working practices at the outset of a project. Senior lawyers are often unaware of the granular processes used by others to develop documents. Equally, in the heat of a legal project, these topics are unlikely to be the most pressing things.

A pre-project meeting discussing and agreeing these kinds of process issues can be extremely valuable — not only for document collaboration, but for other aspects such as matter management and fee reporting.

Key question: “are there any technical barriers to using co-authoring, and if not, is the team as a whole bought into the advantages of using co-authoring?”

So, are lawyers too old-school?

It depends on the context

We shouldn’t be too quick to judge lawyers for not using tools such as Google Docs. As a general rule, there is a greater need for granular version control when there is a large number of contributors working on a document, and their contributions need careful review and approval. Similarly, where formal sign-offs or approvals are requested, there is a need to lock down changes in a document. In these situations, the old-school method is still appropriate.

But there is less need for rigid version control where smaller teams are working together tightly, and are looking at the document from the same perspective. Switching to use co-authoring will likely be efficient in these circumstances. It will also improve information flow between teams, meaning mistakes are discovered sooner and can be resolved more quickly.

Common situations to use co-authoring

Here are some common situations where I have seen co-authoring work well — I have put these in order of ease of adoption (easiest first).

  • Composite documents — where teams would otherwise have to split the document up into numerous other documents, work separately, and join them together afterwards
  • Tracker documents — where lawyers use a long Word table to track each person’s progress on their relevant work streams
  • Long, separable documents — where contributors are working on completely separate sections (e.g. tax warranty provisions, areas that specifically relate to a foreign law)
  • Same team collaboration— where individuals from the same team recognise that the old-school method provides little value in these circumstances, and have overcome cultural barriers to all edit the same document “live”. From my conversations, I have observed that litigation lawyers see the value in this more easily than transactional lawyers.

It is up to each legal team to decide how to collaborate on their documents. With all the tools available today, my hope is that lawyers at least start talking to each other about how they want to work together. Too often, nobody talks about these “boring” processes, and they end up following a sub-optimal method just because it’s what they are used to. But open conversations about these subjects trigger cultural changes, and help unlock many of the barriers to better document collaboration.



Jack Shepherd

Ex biglaw insolvency lawyer and innovation. Now legal practice lead at iManage. Interested in human side of legal tech and actually getting things used.