The tolerated pain
A customer emails service@ to book a repair. Another asks for a quote on a maintenance contract. A third reports that the last call-out did not fix the problem. All three land in a shared mailbox that five people can see. Each person assumes someone else is dealing with the ones they did not open. One message is read on a phone, marked as read and forgotten. Another is answered twice, with different dates. The third is never answered at all.
Nothing looks broken. The mailbox is busy, people are replying, and most requests get handled. But the business has no list of open requests, no owner for each one and no way to know what was missed until a customer phones, angry, or simply goes elsewhere.
This article focuses on customer service requests: the repairs, call-outs, warranty claims, bookings and change requests that become billable work or protect existing revenue. Internal requests handled over email are a related but different problem, covered in the pain of processing routine requests through email chains.
Why businesses tolerate it
- Email is already there. A shared mailbox costs little to set up and everyone knows how to use it.
- It worked when the business was small. With two people and twenty emails a day, memory was enough.
- Customers prefer email. The business does not want to force customers into a portal.
- The losses are invisible. A missed request leaves no trace. There is no report of jobs that never happened.
What a shared mailbox is, and is not, designed for
Shared mailboxes are a reasonable tool for receiving mail at a common address. Microsoft's guidance about shared mailboxes in Microsoft 365 lists support and reception addresses among their intended uses, and notes they need no separate licence up to 50 GB. The same page also sets out limitations: a shared mailbox supports a maximum of 25 users, and too many people accessing it at once can cause connection failures or duplicated messages; and the business cannot prevent users from deleting messages in it.
None of this makes a shared mailbox wrong. It does mean that a shared mailbox, on its own, is a place where mail arrives, not a system for tracking whether each customer request has been completed. There is no built-in owner, status or due date for a message, so those have to be managed by habit.
Dedicated service tools work differently. Freshdesk's documentation on converting emails into tickets describes forwarding the existing support address so that each incoming email becomes a ticket with the customer recorded as the requester, assigned to a chosen group, with follow-up replies threaded onto the existing ticket rather than creating new ones. Customers keep emailing the same address; the business gains a record per request. Other service desks and CRMs offer similar features.
Operational symptoms
- customers chasing requests they sent days ago;
- two staff replying to the same customer with different answers;
- colour-coded flags and folders that mean different things to different people;
- nobody able to say how many requests are open, overdue or waiting on the customer;
- service quality that depends on which person happened to open the email, a pattern described in when service quality depends on who picks up the ticket.
How the pain becomes money
- Lost billable work: a missed repair or quote request is revenue the business never earns.
- Customer churn: customers on maintenance or service contracts who feel ignored are less likely to renew.
- Contractual penalties: where service-level agreements apply, late responses may trigger credits or penalties.
- Duplicate labour: two people handling the same request, or a technician dispatched twice.
- Chasing cost: time spent answering "what is happening with my request?" calls, a problem explored in why customers have to ask what is happening.
An illustrative cost model
This example is illustrative; every figure is an assumption. Picture a Bloemfontein refrigeration and air-conditioning service company receiving 400 customer requests a month through service@.
| Cost line | Illustrative assumption | Monthly cost |
|---|---|---|
| Requests missed entirely | 2% of 400 = 8 requests; half would have become jobs; R3,500 gross profit per job | R14,000 |
| Duplicate handling | 15 duplicated requests × 30 minutes × R250 per hour | R1,875 |
| Chasing calls | 60 calls × 10 minutes × R250 per hour | R2,500 |
| Total | R18,375 |
That is about R220,000 a year on these assumptions, before counting any lost contract renewals. The rates are invented to show the method. To estimate your own, sample a month of the mailbox and count requests with no reply, requests answered twice and follow-up emails from customers asking for updates.
Why better inbox habits are not enough
The first response to lost requests is usually a rule: flag what you are handling, move finished emails to a folder, reply-all so others can see. These rules help for a while, but they depend on every person following them every time, including on a phone, on a busy Monday or while covering for a colleague. One person marking a message as read is enough to hide it from everyone else.
There is also a trade-off to respect. Introducing a ticketing tool adds a system to learn and maintain, and a poorly configured one can bury customers in automated replies. Keep the first version simple: one queue, a few categories, an owner and a due date. Add automation only once the basic habit is working.
When it becomes commercially urgent
- Customers regularly chase requests or complain about non-response.
- More than a handful of people work from the same mailbox.
- The business sells service contracts with response commitments.
- Management cannot report how many requests were received, completed or overdue last month.
What better looks like
Every customer request becomes a record with an owner, a status and a due date, whether in a service desk, a CRM case module or a job-management system. Customers can keep emailing the same address. Requests are categorised on arrival and routed to the right team. Staff see their own queue, managers see the overdue list, and customers receive an acknowledgement with a reference number. Repeat contacts attach to the original request instead of starting a new thread. Reporting shows volume, response time and completion, so missed work becomes visible. The broader dynamics are covered in why shared inboxes become operational bottlenecks.
Practical next actions
- Sample one month of the shared mailbox and count unanswered, duplicated and chased requests.
- List request types: bookings, repairs, quotes, complaints, warranty claims.
- Decide where requests should live: an existing CRM, a service desk or a job system.
- Forward the existing address into that system so customers notice no change.
- Define ownership and response targets per request type.
- Review the overdue list daily for the first few weeks until the habit forms.
Mellorca's digital operations improvement work often starts with exactly this kind of request flow, because it sits so close to revenue.
Bottom line
A shared inbox can receive customer requests, but it cannot make sure each one is completed. When requests are tracked by memory, a small share is inevitably lost, and each lost request is service work, revenue or goodwill the business has already earned the right to. Giving every request an owner and a status is a modest change with a direct financial return.
Sources and further reading
- About shared mailboxes in Microsoft 365 - Microsoft 365 admin (last verified 10 October 2026)
- Converting Support Emails into tickets in your support portal (last verified 10 October 2026)
Related Mellorca reading
- When Customers Have to Repeat Themselves Every Time
- Sales, Finance and Operations Each With Their Own Customer: The Cost
- "I Thought You Were Doing It": The Labour Cost of Unclear Responsibilities