How often does a mortgage handoff lose the file's next action?

A handoff can look complete because the file moved to a new queue. The receiving owner opens it and finds a note that says the borrower was contacted, without a date, a document, or a next step. The file is now owned by someone who cannot act on it. That gap is the subject of this protocol.

Question and scope

The study asks how often a handoff leaves the receiving owner unable to state the file's next action. Scope names the handoff types, the stages, the teams, and the period. A handoff from setup to processing is different from a handoff from processing to closing. Report each type separately.

Define the receiving owner before sampling. The owner is the person or role expected to act next, not the person who last touched the file. If the queue has no named owner, the handoff cannot be scored because there is nobody to receive it. That absence is itself a finding.

Set the observation window around the handoff. The receiver should be asked within one business day, before the file has moved again. A later interview may capture a reconstructed answer rather than the information available at the handoff. Record the time of the question and the time of the handoff.

Define a handoff

A handoff is a transfer of ownership or a transfer of a task between people, queues, or systems. It can be formal, such as a stage change, or informal, such as a message that asks a colleague to take over. The protocol should include both if the team uses both. A message that transfers a task but leaves ownership with the sender is a consultation, and the protocol should say so.

Record the handoff event with a timestamp, the sending owner, the receiving owner, the channel, and the record that proves the transfer. A stage change without a note may still be a handoff if the queue is the formal record. The team should agree on which evidence counts. If the team cannot name the record, the handoff may be happening through a channel that the study cannot see.

Do not count a consultation as a handoff. A question that leaves ownership with the sender is a request for help, not a transfer. The distinction matters because consultations may not need a full next action. The protocol should give examples of each.

Define the next action

The next action is the specific step the receiving owner should take, with the file and the evidence it concerns. It should name a verb and an object, such as request the missing bank statement or confirm the appraisal fee. A note that says review or follow up is not a next action because it does not say what to review or whom to follow up with. The action should also name the evidence that will satisfy it so the receiver knows when the step is done.

The next action should also have a date or a trigger. A date can be a deadline or a planned contact. A trigger can be the receipt of a document or the completion of a vendor step. Without a date or trigger, the action cannot be scheduled or checked.

Allow a range of acceptable answers in the coding. The receiver may state the action in their own words, and two reviewers may agree that the meaning is the same. The test is whether the receiver can describe the action without opening the sender's private notes. If they need the sender, the handoff did not transfer enough context.

Sample the handoffs

Select handoffs from a defined window and include each handoff type. Use a random sample if the volume is high, or all handoffs if the volume is low. Record the file key, the sender, the receiver, the time, and the handoff type. Do not include borrower names in the working table.

Include handoffs that occurred at different times of day and on different days of the week. A Friday afternoon handoff may carry less context than a Tuesday morning handoff. The sample should show whether timing is a factor. If the team cannot sample every handoff, use a rotating schedule.

Record whether the receiver had worked the file before. A receiver who knows the file may need less context than a new owner. The protocol should separate familiar and unfamiliar receivers so the result is not distorted by experience. A handoff to a new owner is the harder test.

Collect the evidence

Ask the receiver to state the next action and its date before opening the sender's message. Capture the answer in their words. Then open the record and compare the answer with the sender's note, the stage, and the documents. The comparison shows whether the information was present and whether it was usable.

Do not prompt the receiver with the answer. A prompt turns the test into a teaching moment and invalidates the sample. If the receiver cannot answer, record the gap and ask what they would need. The question shows the missing field without leading the witness.

Capture the time it took the receiver to find the action. A handoff that requires twenty minutes of searching is not complete even if the answer exists somewhere. The protocol can use a simple threshold, such as two minutes, and record the actual time. The threshold should be set before the sample.

Code the result

Use four codes: clear action, partial action, no action, and wrong action. A clear action names the step and the date or trigger. A partial action names one but not the other. A no-action handoff has no usable next step. A wrong action points to a step that is not appropriate for the file.

Count a lost action when the receiver cannot state the action or states a different action that the evidence does not support. Do not count a late action as lost. Lateness is a separate measure with a different cause. Mixing the two makes the result harder to act on.

Have a second reviewer code a subsample. Compare the codes and discuss the disagreements. A disagreement about whether an action is partial or clear may point to a vague definition. Update the examples and keep the codebook with the study.

Analyze the pattern

Report the share of handoffs with a clear action by type, sender, receiver, and time of day. A high share of lost actions from one queue may point to a template problem. A high share from one sender may point to a workload or training problem. Avoid using the result to rank individuals without context.

Compare the handoff record with the file's next recorded action. If the receiver eventually performed a different action, the handoff may have been technically clear but practically wrong. That pattern is worth a separate code because it shows a judgment gap rather than a documentation gap. A separate code for a wrong but clear action keeps the documentation measure from absorbing a judgment problem.

Look at the handoffs that required the most searching time. Those cases often reveal a missing field or a note that buries the action in a long narrative. A short template with a required action and date can fix many of them. The mortgage operations support lead can test the template on one queue.

Report the result

Open with the share of handoffs that met the clear-action test. Show the sample size and the confidence range if the team can calculate one. Then show the top three causes: no action recorded, action recorded without a date, and action buried in a long note. Each cause should have an example with identifiers removed.

Add a short section on the handoffs that worked. A good example shows the team what a complete note looks like and gives the trainer a model. The example should include the file context, the action, the date, and the receiver's confirmation. Do not use a real borrower name.

End with one change and a review date. The change might be a required field in the handoff form or a short check at the receiving queue. The review date should be soon enough to see whether the change took hold. A report without a test is only a description.

Limits and sources of error

The receiver's answer is a snapshot of understanding, not proof that the action was correct. A receiver can state a clear action that is wrong for the file. Add a reviewer check for a subsample to catch that case. Report the two measures separately.

The observer may influence the receiver by asking the question. A receiver who knows the study is running may prepare before the handoff. Use a short, consistent prompt and do not announce which handoffs will be sampled. The protocol should state how the observer avoided coaching.

The sample may exclude informal handoffs that never reach the queue. Add a short survey question that asks receivers whether they had to ask the sender for context. The answer can estimate the informal gap without requiring a full log. Report the estimate as an estimate.

Review cadence

Repeat the study after a template change, a queue redesign, or a staffing change. Use the same definitions and the same receiver test so the comparison is meaningful. A share that improves while search time stays high may mean the action is present but still hard to find. A share that improves while the search time stays high may mean the note is clearer but still buried.

The loan processing support team can run the review with the senders and receivers in the same room. Ask them to read three handoffs and state the next action. The exercise is fast and shows whether the standard is shared. The goal is a handoff that lets the next person start work without a conversation.