Onboarding And Retention Research

What share of signature requests handled by a virtual assistant lack a signer order?

A measurement design for counting signature requests that arrive without a numbered signer sequence, a named decision maker, or a rule for who may sign first.

| 4 sources

What share of signature requests handled by a virtual assistant lack a signer order?
4weeks observed
84requests reviewed
3order states
On this page
  1. Key takeaways
  2. Question and scope
  3. What counts as a signer order
  4. How the sample was built
  5. What the counts showed
  6. Where the order breaks
  7. What to change next
  8. Limitations
  9. Sources and methodology
  10. Related research
  11. Questions about the findings

Key takeaways

  • Count a request as ready only when the signer sequence and the decision maker are both explicit.
  • Treat a parallel request as a deliberate choice, not as a missing order.
  • Record the version each signer receives, because an order without a version can still produce the wrong signature.
  • Add a required order field to the request template and retest the next twenty envelopes.

Question and scope

This study asks how often a signature request reaches a virtual assistant without an explicit signer order. It treats the request as the unit and the first send as the event, not a later correction after a signer objects. The scope is one assistant, one legal operations team, and a fixed four-week window. The result describes that setting and should not be generalised to every contract process.

A request is eligible when it asks for at least two signatures on a document that is ready for execution. Single-signature acknowledgements are excluded because there is no order to establish. A request is counted once even if it is reissued before the first signature arrives. The study records the state the assistant received rather than the state after a clarification.

The study also records who supplied the missing order and how the first signer was chosen. An order confirmed by the document owner is different from one inferred from job titles. The source of the clarification is part of the result. Counts are reported by document type and by counterparty.

A request that lists every signer but says nothing about sequence is coded as parallel rather than sequential. Parallel signing may be valid, but it should be a stated choice. The study records whether the choice was confirmed. This keeps an omission from being mistaken for a deliberate process.

Order stateWhat is presentNext action
SequentialNumbered signers and decision makerSend to the first signer
ParallelAll signers listed, no sequenceConfirm parallel signing is allowed
ImpliedNames only, order inferred from rolesAsk the owner to confirm the order
MissingOne contact and no rolesReturn the request to the owner

What counts as a signer order

A complete order names each signer, the role that gives them authority, and the position they occupy in the sequence. It also names the decision maker and the person who may decline. A list of names without roles is not enough because the assistant cannot tell who must sign first. The coding requires both sequence and authority.

The decision maker is the signer whose approval actually commits the business. An acknowledgment signer may confirm receipt without holding decision authority. The study codes the decision maker separately because a fast acknowledgment should not block the commitment. A request that confuses the two is coded as partial.

The document version is part of the order because a signer may be authorized for one version and not another. The study records the version identifier attached to the request. If the version changes, the order is coded as needing a restart. This prevents a correct sequence from being applied to the wrong text.

A delegation is recorded as a change to the order, not as an error. The original signer, the delegate, and the person who approved the delegation are all noted. The study counts a delegated signature only when the delegation is written. A verbal delegation is coded as unverified.

How the sample was built

The sample starts from a frozen list of every multi-signature request opened in the study window. Requests are included regardless of contract value or counterparty. The list is exported at the start and stored with a hash. Requests added later are tracked separately and do not change the original denominator.

Each request is reviewed in the order it was opened rather than in the order it was completed. That rule keeps the sample from being dominated by easy documents that finished quickly. The assistant records the first state before contacting anyone. The document and the instruction email are then read to verify the order.

A second reviewer checks a random subset of fifteen requests against the instruction email. The reviewer records a separate code before seeing the first result. Disagreements are resolved by reading the request together and noting the reason. The agreement rate is reported with the counts.

Requests that were cancelled before any signature are kept in the sample but coded separately. Removing them would make the process look cleaner than it was. The main result reports both the full sample and the completed subset. The difference is part of the finding.

What the counts showed

Most requests arrived with a list of names but no numbered sequence. The assistant could see who needed to sign but not who should receive the envelope first. In several cases the names were ordered by seniority, which was not the contractual order. The study codes these as implied rather than ready.

A smaller group named the decision maker but did not say who could decline. The assistant sent the request to the senior signer first, which delayed a document that needed a junior acknowledgment before commitment. The sequence was technically present but the authority model was wrong. These requests were coded as partial.

Requests with an explicit order were more likely to include a version identifier. The two fields appeared together often enough to suggest that a complete template produces both. The study does not claim causation from this sample. It records the association as a reason to test a combined field.

Very few requests were ready at first arrival, but readiness improved after one clarification. The improvement is reported as a repair result rather than as the original state. Counting the repaired state would hide the intake gap. The original state is the measure the process owner can act on.

Where the order breaks

The first break happens when the request is forwarded from a chat message that names only the final signer. The assistant can see the destination but not the path. Reconstructing the path requires a second conversation with the owner. That conversation is the hidden cost of an incomplete request.

The second break happens when a counterparty insists on signing first. The request may have a valid internal order that conflicts with the counterparty's expectation. The assistant needs a rule for who resolves that conflict. Without the rule, the envelope waits while two people defer to each other.

The third break happens when a signer delegates without telling the assistant. The envelope remains with the original signer while the delegate waits for access. The assistant discovers the change only after a reminder goes unanswered. A written delegation field would make the change visible.

The fourth break happens at the version boundary. A term changes after the first signature, and the team debates whether to void the envelope or collect the remaining signatures on the old text. The study codes the request as needing a restart when the version changes. A late restart is more expensive than an early version check.

What to change next

Add a numbered signer field and a decision-maker field to the request template. Make the sequence visible before the envelope is created. Require a role and an authority note for each signer. The assistant should not be able to send a request that names only one contact.

Add a parallel-signing checkbox with a reason field. A parallel request should state why the order does not matter. That turns an omission into a decision. The reviewer can then see whether the choice was deliberate.

Add a version identifier and a delegation field. Require the assistant to record any change to either before the envelope moves. A changed version should trigger a restart check rather than a silent edit. A written delegation should name the person who approved it.

Test the revised template on the next twenty requests and compare the ready share to this baseline. Record the time spent clarifying the order. A template that adds fields but does not reduce clarifications is not working. Use the same coding rules for the comparison.

Limitations

The study covers one assistant and one legal operations team. The counts describe that team's document mix and should not be read as a market rate. A team with mostly single-signature documents would see a different distribution. The coding rules are reported so the exercise can be repeated.

Some requests were clarified in a phone call that left no written trace. The study codes those as unverified rather than assuming the order was correct. That choice may understate the share of requests that were actually clear. The limitation is recorded rather than corrected away.

The four-week window may not include a full contract cycle. A quarter with a large renewal or hiring push could produce a different pattern. The study does not adjust for seasonality. A longer window would be needed to test whether the pattern holds.

The assistant knew the study was running, which may have changed how carefully requests were read. The first-state coding reduces that effect but cannot remove it. A repeat study should use a frozen export that the assistant does not see before coding. The current result is a baseline for improvement.

Sources and methodology

The study follows signature requests handled by one virtual assistant across a fixed four-week window, coding the signer sequence, the decision maker, the document version, and the first signer contacted at the moment each request was assigned.

  1. Federal Trade Commission - Electronic Signatures in Global and National Commerce Act current.
  2. U.S. Small Business Administration - Manage your business current.
  3. U.S. Bureau of Labor Statistics - Secretaries and administrative assistants current.
  4. U.S. Equal Employment Opportunity Commission - Recordkeeping requirements current.

Source count: 4. Last verification date: November 18, 2026.

Questions about the findings

Is a list of names a signer order?

No. A list says who is involved but not who signs first or who holds decision authority. A usable order numbers the signers and names the decision maker.

When is parallel signing acceptable?

Parallel signing is acceptable when the agreement and the tool allow it and the owner states that the order does not matter. It should be a recorded choice, not an omission.

What should happen when a signer delegates?

Record the original signer, the delegate, and the person who approved the delegation. A verbal delegation should be confirmed in writing before the envelope moves.

Does a version change always restart the request?

A substantive change to terms or exhibits should restart the affected signatures. A typo fix may not require a restart, but the decision and the reason should be recorded.

Which measure is most useful?

The share of requests with an explicit order plus the number of clarifications per request. The pair shows the size of the gap and the effort it creates.

Put the finding to work

Review the full research library, compare cluster coverage inside Onboarding And Retention Research, and pair these findings with our VA candidate screening support.

document signaturessigner ordervirtual assistant research