Home / Research / Prebill workflow

What share of prebill holds clear without partner action?

A protocol for measuring how many prebill holds are released by the billing team, how many need a partner decision, and how long each path takes.

Legal Billing Assistant Editorial Team · Published

Editorial illustration representing What share of prebill holds clear without partner action?
Table of contents

Method note: this brief proposes a source-led review approach. It does not claim that private client or firm data has been collected.

Question and scope

This protocol asks what share of prebill holds clear without a partner or other authorized attorney decision. The purpose is to understand how much of the hold queue is administrative and how much genuinely requires professional judgment. The unit is a hold event, not an invoice, because one invoice can carry several independent holds. The result describes the observed queue and does not evaluate any individual partner's responsiveness.

Define the observation window, the client population, and the prebill cycle. Include holds opened during the window and holds that were already open at the start, with a flag for each. Exclude draft records that were never submitted for review. A clear scope prevents the denominator from drifting as the study proceeds. Record the reason the window was chosen so a later analyst can judge whether the period was typical.

Defining a hold and a release

A hold is a prebill record that is complete enough for review but cannot be released until a named condition is met. A release is the recorded decision that the condition was satisfied. Do not count a status change that merely reflects a new note or a reassignment. The release must change the record's eligibility to move forward.

Classify the release authority before counting. Administrative release includes corrections the billing team is authorized to make under a written rule. Partner release includes any decision reserved to an attorney or another named approver. If a hold requires both, count the highest authority required. Record the classification rule so two reviewers do not split the same event differently.

Population and sampling

The population is every hold event in the window for the matters in scope. If the volume is too large to review completely, sample by hold reason and by week rather than by convenience. A sample drawn only from the busiest week will overrepresent the conditions of that week. State the sampling method and the confidence limits that apply.

For each sampled hold, collect the open date, the release date, the reason, the authority, the evidence attached, and the final action. Include holds that remain open at the end of the window and mark them as censored rather than dropping them. Dropping open holds biases the result toward fast releases. The analysis should account for the observation time available.

Evidence standard

A release is supported when the record contains the evidence named in the hold rule. For an administrative correction, that may be a corrected field and a note referencing the rule. For a partner decision, it may be a dated approval or a documented instruction. A verbal release without a record should be coded as undocumented and reported separately.

Retain the source of the evidence and the date it was attached. Do not infer a release from the absence of a hold later in the workflow. The record may have been reopened, merged, or removed for another reason. The evidence standard should be strict enough that an independent reviewer can confirm the release from the file.

Coding the release path

Code each hold by the first authority that actually resolved it, not the authority named in the initial request. A hold opened as partner review may clear administratively once the rule is clarified. That outcome is important because it shows where the initial routing was unnecessarily high. Keep both the requested and the resolving authority.

Add a code for holds that were withdrawn because the underlying charge was removed. A withdrawal is not the same as an approval, and mixing them inflates the administrative release rate. Add a separate code for holds that remain open and one for holds that were escalated beyond the original owner. The coding scheme should cover the outcomes the queue actually produces.

Analysis and reporting

Report the share cleared by administrative authority, the share cleared by partner authority, and the share still open at the end of the window. Include the median and the oldest age for each group. A single percentage without the age distribution hides the difference between a fast administrative fix and a slow partner backlog. Include the count of holds still open at the end of the window. Show the denominator beside every percentage so the reader can judge the base.

Break the results down by hold reason and by client. A client whose guideline requires frequent partner approval will look different from a client whose holds are mostly missing matter fields. Compare the requested authority with the resolving authority to identify routing that can be corrected. The report should show where a rule change or a clearer checklist would reduce the queue.

Limits and sources of error

The result depends on the accuracy of the hold reason and the completeness of the release evidence. A queue that allows free-text reasons will be harder to code consistently than one with a controlled list. A release recorded only in email may be missing from the system. Report the share of records with incomplete evidence and the effect on the estimate.

Partner availability and client deadlines can change the mix during the window. A holiday week or a major filing deadline may produce a different pattern than a normal week. Do not generalize from a short window without noting the conditions. If the window is not representative, extend it or report the result as a period-specific observation. Include the number of holds that were reopened during the window because a reopen changes the release path, and report that share beside the cleared share.

Retention and privacy

Keep the coded dataset in a controlled location with access limited to the study team. Remove client names and matter numbers from the working file when the analysis does not need them. Retain the linkage key separately if a later audit requires re-identification. Follow the firm's retention schedule for billing records and study notes. State the access rule in the same note so a later analyst can verify who may reopen the file.

Do not include quoted client instructions in a report that circulates beyond the billing team. Summarize the rule instead. If the result is shared with a client, aggregate across matters and remove any detail that could reveal another client's position. The measurement should improve the process without creating a new confidentiality exposure. Keep the redacted version with the report and the full version in the controlled repository.

Applying the result

A high administrative release share may indicate that too many holds are routed to partners by default. Review the hold rules and give the billing team clear authority for the corrections it already makes. A low administrative share may indicate that the queue is capturing genuine judgment questions, which is not necessarily a problem. The response should follow the reason mix. Record the retention period and the owner of the dataset in the study file, along with the access rules that apply to it.

Use the age data to set escalation points for the slowest reason categories. If partner holds are the bottleneck, a standing review slot may be more effective than individual reminders. If administrative holds are slow, the bottleneck may be access or training. Match the intervention to the measured path rather than to the total volume. Use role labels rather than names when the result is shared with the practice group, and remove the names before the file leaves the team.

Review cadence

Repeat the measurement quarterly and after any change to the hold taxonomy. Keep the protocol and the coding guide stable so the trend is comparable. If a new reason category is added, report it separately for the first period rather than blending it into an existing group. The comparison should be transparent about the change. Keep the coding guide with the result so a new analyst can repeat the classification, and assign the review to a named owner.

Share the result with the billing manager and the partner responsible for the queue. Ask whether the administrative release rules still match the firm's risk tolerance. Update the rules if the answer changes, then measure again. The cadence should follow the decisions the measure is meant to inform. Record the date of the review and the name of the owner who accepted the next action.

Reporting the result to the queue owner

A share without context can be read as a target, so report it beside the hold volume, the reason mix, and the observation window. The queue owner needs to know whether a change in the share reflects a new routing rule or a different mix of matters. Present the result as a description of the period, not as a score for the team. That framing keeps the conversation on the process.

Ask the queue owner which holds could be moved to administrative release without weakening the control. Test one change on a small sample before applying it across the queue. Record the change date beside the next result so the comparison is honest. The measure should support a decision about routing, not just a monthly report.

Authoritative sources

These public references inform the operating questions in this article. They do not replace legal, ethics, privacy, financial, or client-specific advice.

  1. U.S. Government Accountability Office Green Book
  2. Legal Electronic Data Exchange Standard
  3. American Bar Association Model Rules of Professional Conduct
  4. Internal Revenue Service recordkeeping guidance
  5. NIST Privacy Framework
  6. NIST Cybersecurity Framework

Put the guidance to work

Choose one recurring billing workflow, identify its authoritative source records, and name the owner who reviews exceptions. We can help turn that boundary into a focused support role.

Discuss your workflow