Small-business outsourcing research

How much shared-inbox work sits unassigned after one business day?

A protocol for measuring unassigned shared-inbox work, the age of the backlog, and whether a named owner picks up each conversation.

How much shared-inbox work sits unassigned after one business day?

Research question and scope

This protocol asks how much shared-inbox work sits unassigned after one business day. The unit is one conversation thread, identified by the first inbound message and the shared inbox. The study measures whether a named owner was assigned, when the assignment happened, and whether the customer received a reply. It does not evaluate whether the reply was correct or whether the customer was satisfied.

Set the scope to one small business, one shared inbox, and one assignment process. Record the business hours, the triage rule, and the people who can take work at the start. If the process changes during the window, keep the earlier definition for the full sample or analyze the change as a separate period. A new auto-assignment rule can look like better ownership when it is only a new routing.

Population and unit of analysis

Define the population as every conversation thread that received a customer message during the window, including threads that were answered and threads that were closed without a reply. The unit is the thread, not the message, because a thread can contain several messages. A thread that is reopened after a customer reply remains one unit with an additional interval. A new thread from the same customer about a different issue is a new unit.

Record the channel, the topic, the customer type, and whether the message arrived during business hours. These strata matter because a billing question and a general inquiry carry different deadlines. Include threads that were assigned and then reassigned, because the reassignment is part of the ownership history. Test the unit definition on two ordinary threads before collection begins.

A thread can be assigned to a person who is on leave, which looks like ownership but produces no reply. Record the owner's availability at the assignment time and flag the thread when the owner is away for more than a day. Do not treat an out-of-office reply as coverage or as a customer response. The assignment rate should distinguish an active owner from a name in a field.

Definitions that make the measure reproducible

Define an assignment as a named person recorded as the owner in the inbox, not a label that describes the topic. Define the assignment time as the moment the owner was recorded, not the time the conversation was first read or tagged. Define unassigned work as a thread with no named owner after one full business day. Define a reply as an outbound customer-facing message, not an internal note or an automatic acknowledgement.

Do not count a team queue as a named owner unless the team has one accountable queue lead. Do not count a read receipt as a reply. A thread that is waiting on a customer is assigned when a named person owns the follow-up. Each definition should include one positive example and one boundary example.

Evidence standard and data dictionary

For each thread, capture the customer, the channel, the first message time, the assignment time, the owner, and the first reply time. Add the topic, the customer type, and whether the thread was reassigned before the first reply. Record the current status and the date it was last updated. Keep the thread export with the inbox record.

Also capture whether the thread was tagged and whether the tag matched the topic. A tag without an owner can make a thread look handled when nobody is responsible. If the owner is on leave, record the cover person and the date of the transfer. A second reviewer should be able to reconstruct the ownership timeline.

A shared inbox can apply an automatic label that looks like ownership without naming a person. Record the rule that applied the label and the person who is expected to act. A label that routes work to a team is not an assignment unless the team has one accountable lead. The data dictionary should separate the two.

Sampling and collection

Draw the sample from the inbox export rather than from a list of complaints. Record the export date, the filters, the row count, and the selection rule. Include threads from every channel and topic so the result is not dominated by one source. If the population is ordered by arrival time, a fixed interval spreads the sample across the window.

Collect the thread as it existed at the end of the first business day, not as it looks after a later cleanup. If the assignment happened later, record the later time and analyze the original gap separately. Do not infer an owner from the person who eventually replied. Note the collection date separately from the arrival date.

Analysis and reporting

Report the number of threads, the share assigned within one business day, the share assigned later, and the share still unassigned at the end of the window. Show the counts by channel, topic, and arrival period so a busy Monday does not dominate. Present the median time to assignment and the share that received a reply. Use the same denominator for every rate.

Inspect the unassigned threads and describe the pattern without naming the customer or the team member. A cluster on one channel may point to a missing triage rule, while a cluster on one topic may point to a skill gap. Compare the result with the arrival time, because a late Friday message has fewer business hours before the deadline. Report the reassigned threads separately.

A useful additional view is the share of threads that were assigned but never received a customer-facing reply. An owner without a reply is still a broken handoff, even when the queue looks clean. Show the gap between assignment and first reply and describe the pattern. That view helps the owner decide whether the problem is triage or capacity.

A useful cut is the share of threads that received a first reply before an owner was recorded. That pattern shows work happening without ownership, which makes the queue hard to manage even when customers are answered. Show the sequence and describe the consequence. That view helps the owner decide whether the assignment step is realistic.

Limitations and sources of error

The record cannot show whether the customer was satisfied or whether the issue was resolved. It can show only whether a named owner existed and when a reply was sent. An assigned thread is not proof of good service, and an unassigned thread is not proof that nobody looked. The protocol also cannot capture a conversation handled by phone or in person and never logged.

Inbox tools can auto-assign, reassign, and merge threads in ways that change the history. A thread that looks unassigned may have been handled through a private channel. Record those cases and their treatment instead of dropping them silently. If the same person triages and answers the work, note the independence limitation and analyze that subgroup separately.

Review cadence and next action

Give the result to the owner and the inbox lead and ask them to choose one observable change. The change might be a triage owner for each channel, a one-business-day assignment rule, or a daily review of unassigned work. Record the owner, the effective date, and the measure that will show whether the change worked. Avoid changing the triage rule and the staffing at the same time.

Repeat the protocol with the same definitions after the change and compare the same fields. If assignment improved but replies did not, the owner may be assigned without capacity to act. If replies improved but assignment time rose, the team may be answering without recording ownership. Close the review when the owner accepts the finding and the next sample is scheduled.

Give the inbox lead the same protocol and a short list of the channels with the oldest unassigned work. A rule that works for one channel may fail when the volume shifts to another or a new channel is added. Record the review date and the person who owns each channel. The next sample should use the same definitions.

Sources

Frequently asked questions

Does a favorable sample prove the routine caused the result?

No. The design is observational and supports a bounded operating decision, not a universal causal claim.

What should an owner review first?

Start with high-consequence defects, missing denominators, exceptions, and cases that required the owner to reconstruct the evidence.

Related research