Can You Trust Track Changes? Why Lawyers Run Independent Comparisons
Track Changes is the backbone of contract negotiation workflow. One side sends a draft. The other turns on Track Changes, makes edits, and sends it back. The markup shows exactly what was changed. Both sides review the tracked edits, negotiate further, and repeat until agreement.
It works. Most of the time.
The problem is that "most of the time" is not a standard any lawyer should be comfortable with when reviewing contracts. Track Changes has specific, well-known limitations that can result in an incomplete or misleading record of what changed between versions. These limitations are not bugs. They are design characteristics of the feature. Understanding them is the difference between treating Track Changes as a reliable audit trail (it isn't) and treating it as a useful convenience that needs verification (it is).
This post covers what Track Changes actually records, where it falls short, and why experienced lawyers run an independent comparison alongside every Track Changes review.
What Track Changes actually records
Track Changes in Word is a revision tracking feature. When enabled, it records insertions, deletions, formatting changes, and moves made to the document. Each change is attributed to a user and timestamped. The result is a visual markup that shows what was added, removed, or modified.
For contract negotiation, this is genuinely useful. When opposing counsel sends back a draft with Track Changes, you can see their edits in context: strikethrough for deletions, colored text for insertions, formatting change indicators in the margins. You can accept or reject individual changes. You can filter by reviewer. You can navigate from one change to the next.
The critical word in that description is "when enabled." Track Changes records edits made while the feature is turned on. It does not record edits made while the feature is turned off. It does not know about edits that happened before it was enabled. It does not retroactively detect changes. It is a recording tool, not a comparison tool. The distinction matters.
Where Track Changes falls short
These are not edge cases. They are normal occurrences in routine contract negotiation.
Edits made with tracking turned off
This is the most significant limitation. If someone turns off Track Changes (whether accidentally or intentionally), makes edits, and turns it back on, those edits exist in the document but are invisible in the Track Changes markup. The document looks like it has complete tracking. It doesn't.
How does this happen in practice? A reviewer opens the document and Track Changes is on. They scroll through, accepting and rejecting edits. At some point, they click the Track Changes button to turn it off, intending to "clean up" a section. They make edits. They may or may not remember to turn it back on. When they save and return the document, the Track Changes that are visible represent a partial record.
There is no indication in the document that tracking was interrupted. Word does not flag the gap. A reviewer looking at the Track Changes markup has no way to know that edits were made outside the tracking period.
Multi-author editing with inconsistent settings
In many negotiations, multiple people on one side edit the same document. A partner reviews the substantive terms. An associate handles the boilerplate. A paralegal cleans up formatting. If their Track Changes settings are inconsistent (different usernames, different tracking preferences, or one person editing with tracking off), the resulting markup is a partial, potentially misleading view of what changed.
Word attributes changes to the username configured in the application. If two people on the same team use the same generic username ("Author" or the firm name), their individual edits are not distinguishable. If one of them turned off tracking, their edits are invisible regardless of username.
Accepted changes that are then re-edited
A common workflow issue. The reviewer accepts a tracked change (removing the markup), then edits the now-clean text. The new edit is tracked, but the original change that was accepted is no longer visible. The Track Changes record shows the new edit but not the original change that preceded it. For someone reviewing the document later, the history looks like a single edit rather than two sequential changes.
This matters when the "accept then re-edit" pattern introduces subtle shifts. The original tracked change said "30 days." The reviewer accepted it, then changed it to "15 days." The Track Changes record shows "15 days" as a new edit. The reviewer on the other side sees the change from the base text to "15 days" but does not see that an intermediate version said "30 days" and was already accepted. The negotiation history is compressed in a way that can obscure the progression of concessions.
Changes in complex document elements
Track Changes handles paragraph text well. It handles other document elements less consistently. Edits inside text boxes, form fields, certain table operations (cell merging and splitting, column adjustments), and header/footer changes are not always tracked completely. Content controls and structured document elements can also behave unpredictably with Track Changes.
For contracts that rely on tables for commercial terms (pricing schedules, SLA matrices, milestone tables), this is particularly relevant. A change to a value inside a table cell is usually tracked. A change to the table structure itself (added rows, merged cells, column widths) may not be. If the structural change affects which values appear in which cells, the content implications of the structural edit can be missed.
Template conversion resets
When a document is converted between templates (common when each side has their own firm template), the conversion process can interact with Track Changes in unexpected ways. Some template conversions strip Track Changes history. Others reset the tracking state. In the worst case, the converted document appears to have clean, complete Track Changes that actually represent only the post-conversion edits, not the full editing history.
This is not just a bad-faith problem
When lawyers hear "Track Changes can be incomplete," the first reaction is often: "Are you saying opposing counsel is hiding changes?" That is not the primary concern.
Intentional manipulation of Track Changes does happen. But it is rare compared to the accidental causes listed above. The much more common scenario is a competent lawyer or paralegal at a reputable firm who accidentally produced an incomplete tracking record through normal workflow friction: a mis-click that turned off tracking, a template conversion that reset the history, a multi-person editing session where settings were inconsistent.
The point is not that you should distrust the other side. The point is that Track Changes is a convenience tool, not an integrity mechanism. It was designed to make collaborative editing visible, not to provide a tamper-proof audit trail. Treating it as the latter overestimates what it can do.
The appropriate level of trust is: "Track Changes is useful for reviewing what the other side intended to show me. Independent comparison tells me what actually changed."
What independent comparison catches
An independent comparison takes two versions of a document and reports every difference between them, regardless of how those differences were introduced. It does not rely on Track Changes, revision history, or metadata. It reads the content of both files and compares them directly.
This means it catches:
- Untracked edits: changes made while Track Changes was off. The comparison doesn't care whether tracking was enabled. It compares what's in the file.
- Accepted-then-re-edited changes: the comparison shows the net difference between the two versions, regardless of the intermediate editing history.
- Table structure changes: a comparison tool that reads .docx structure can detect row additions, cell merges, and value changes in tables that Track Changes may have missed.
- Template conversion artifacts: any content that changed during a template conversion shows up as a difference between the pre-conversion and post-conversion versions.
- Formatting changes: depending on the tool, formatting differences (font changes, style changes, spacing adjustments) are also detected, even when Track Changes was configured not to track formatting.
The comparison result serves as a verification layer on top of Track Changes. If the Track Changes record is complete and accurate, the comparison should match it exactly: same changes, same locations. If the comparison finds differences that are not reflected in Track Changes, those are the gaps you need to investigate.
The "trust but verify" workflow
The goal is not to replace Track Changes with independent comparison. Track Changes is useful. It shows you what the other side intended to change, attributed to specific people, in a navigable format. The goal is to add a verification step that catches what Track Changes might have missed.
Here's the workflow in practice:
Step 1: Review Track Changes normally. Accept, reject, and comment on the tracked edits as you would in any negotiation round. This is your primary review and it drives your substantive response.
Step 2: Run an independent comparison. Take the clean version you sent in the previous round and compare it against the clean version the other side just sent back (the document with all Track Changes accepted). This comparison shows you every difference between what you sent and what you received, regardless of Track Changes.
Step 3: Cross-check. Do the changes in the independent comparison match the changes shown in Track Changes? If yes, the Track Changes record is complete. If the comparison shows changes that are not in Track Changes, investigate those specific differences.
Step 4: Flag discrepancies. If the independent comparison reveals untracked changes, raise them with the other side. "We noticed a change in Section 5.2 that does not appear in the Track Changes markup. Can you confirm this was intentional?" Most of the time, the answer is a version control mistake. Occasionally, it surfaces something more significant.
This workflow adds a few minutes to each review round. For high-stakes contracts, those minutes are the cheapest insurance available. For routine low-stakes work, you can apply the verification step selectively: at minimum, always verify the final execution copy.
When Track Changes alone is enough
Not every document needs the full verification workflow. Track Changes is reliable enough for many situations:
Internal documents. When both parties are within the same organization and using consistent Track Changes settings, the tracking record is usually complete. The accidental-incompleteness risks are lower because the workflow is controlled.
Simple, short documents. A 3-page NDA with two tracked changes does not warrant an independent comparison. You can read the entire document in less time than it takes to set up a comparison. The risk of hidden changes in a short document with minimal edits is low.
Known counterparties with established workflows. If you've done ten deals with the same firm and their Track Changes have always been complete and accurate, the probability of an issue is lower. Not zero. But lower enough that professional judgment can determine when verification is necessary.
Low-stakes agreements. If a missed change in the document would result in a minor inconvenience rather than material financial exposure, the cost of running a verification comparison may exceed the risk. Match the rigor to the stakes.
The trigger for running an independent comparison is usually one or more of: the document is long, the other side reformatted it, the Track Changes look unusually sparse relative to the scope of the negotiation, or the contract carries real financial or legal exposure. When any of those conditions apply, verify.
The bottom line
Track Changes is a useful collaboration tool. It is not a verification tool. The difference matters every time you receive a document from the other side and need to know, with confidence, what actually changed.
The edits you can see in Track Changes are probably accurate. The question is whether they are complete. The only way to answer that question is to compare the documents independently: your last version against their current version, content to content, without relying on the editing history.
Run the comparison. It takes minutes. If Track Changes was complete, the comparison confirms it and you move on with confidence. If it wasn't, you just caught something that would have been invisible otherwise.
For the verification step, any document comparison tool works: Word Compare for simple cases, or Clausul when the document is long enough that you need formatting noise filtered and changes classified. The point is not which tool you use. The point is that you run the comparison.
Frequently asked questions
Can someone hide changes in a Word document without Track Changes showing them?
Yes. Track Changes only records edits made while the feature is enabled. If someone turns off Track Changes, makes edits, and turns it back on, those edits will not appear in the revision history. The document will look like it has complete Track Changes, but the untracked edits are invisible in the markup. The only way to detect them is to compare the current version against the previous clean version using a document comparison tool. This is not a Word bug. It is how Track Changes is designed: it records edits that happen while tracking is on, and nothing else.
Is it common for Track Changes to be incomplete in legal negotiations?
More common than most lawyers realize, and usually not intentional. Track Changes gets turned off accidentally (a user clicks the button without noticing), gets lost during template conversions (some firm templates reset Track Changes settings), or fails to capture certain types of edits (changes inside text boxes, some table operations, comment-anchored deletions). Intentional manipulation happens but is rare compared to accidental incompleteness. The practical implication is the same: if you rely only on Track Changes to understand what changed, you may miss edits regardless of whether they were hidden deliberately or lost accidentally.
Does Word have a way to show if Track Changes was turned off at any point?
Not reliably. Word does not maintain a log of when Track Changes was enabled or disabled. The revision history shows tracked changes with timestamps and author attribution, but there is no record of gaps in tracking. Some forensic document analysis tools can examine document metadata for clues (inconsistent edit timestamps, revision number jumps, metadata that suggests more editing than the tracked changes reflect), but these are specialized tools, not standard practice. For routine contract review, the practical answer is: you cannot tell from the document alone whether Track Changes was on for the entire editing session.
Should I ask the other side to keep Track Changes on during negotiations?
You can ask, and many firms have this as a standard practice agreement at the start of a negotiation. But it is not enforceable as a technical control. The other side can agree to keep Track Changes on and then accidentally (or intentionally) turn it off. Track Changes is a courtesy convention, not a security mechanism. The only reliable verification is independent comparison: take the clean version they sent you last time and compare it against whatever they send you next time, regardless of whether Track Changes is present. Trust but verify.
What is the difference between reviewing Track Changes and running an independent comparison?
Track Changes shows you what the editing tool recorded during the editing process. An independent comparison shows you every difference between two specific document versions regardless of how those differences were introduced. Track Changes can be incomplete (edits made with tracking off), inconsistent (multiple people editing with different Track Changes settings), or misleading (accepted changes that were then re-edited). An independent comparison has none of these limitations because it does not rely on the editing history. It compares the actual content of two files. Both are useful. Only the comparison is reliable as a verification step.
Is running an independent comparison standard practice in law firms?
At large firms, yes. Most Am Law 100 firms have comparison tools (typically Litera Compare) and use them as a verification step alongside Track Changes review. At smaller firms, the practice is less consistent. Some small firms rely entirely on Track Changes, which works most of the time but creates exposure when it does not. The trend is toward comparison as a standard step, driven partly by risk awareness and partly by the availability of self-serve comparison tools that do not require enterprise procurement. If your firm does not currently run independent comparisons, it is worth starting. The first time it catches something Track Changes missed, the habit justifies itself.