The shared mailbox worked for two years and then stopped, and the failure was not dramatic. Nothing was lost. It just became impossible to say, at any given moment, which of the forty open threads somebody was already handling. Two replies went out to the same person on the same day, one of them contradicting the other, and that was the meeting where somebody said the word software.
Team inbox software solves a specific problem: many people, one address, and no way to tell who has taken what. Most of the tools in the category solve it competently. The choice between them turns on things that do not appear in a feature grid, starting with how the price is counted and ending with whether a shared inbox is even the right shape for the traffic in question. What follows works through both.
What the category actually fixes
A shared mailbox opened in an ordinary mail client gives several people the same view and no shared state. Read and unread are per person. There is no field that says who is handling a message. A colleague's draft is invisible until it has been sent, which is why duplicate replies are the signature failure of the setup.
Team inbox software adds three things on top of the mailbox, and all three are about state rather than about mail.
Assignment puts a name on a conversation, so the list can be filtered to what one person owes and what nobody has claimed. This is the feature that earns the purchase. Everything else is refinement.
Status separates a conversation that is waiting on the team from one that is waiting on the other party and one that is finished. Without it, an inbox with nothing in it means either that the work is done or that somebody archived a thread to make the view tidy.
Internal notes and drafts let the discussion about a reply happen against the reply itself, rather than in a chat window where it loses its connection to the thread. This is what stops the forwarded mail chain with four layers of quoting.
Any tool in this category will have all three. The differences start after that.
What the price is counted on
The category prices in at least three different ways, and the difference matters more than the headline number, because the three scale on different things.
| Tool | Entry price today | Counted on | Seat ceiling on the entry plan |
|---|---|---|---|
| Missive | $14 per user per month | Users | Up to 5 users |
| Front | $25 per seat per month, billed annually | Seats | Up to 10 seats |
| Help Scout | $0 for the free plan, then $25 per user per month | Users, plus AI answers at $0.75 per resolution | Free plan up to 5 users |
| Zoho TeamInbox | Priced per user, with a 14 day trial | Users, with shared channels capped by plan | Starter allows 3 shared incoming channels |
Three observations from that table are worth more than the prices themselves.
First, the entry plan almost always has a seat ceiling, and the ceiling is low. Missive's Starter plan covers up to 5 users, and Front's Starter plan covers up to 10 seats. A team of twelve is not choosing between entry plans. It is choosing between second tiers, where Missive's Productive plan is $24 per user per month for up to 50 users and Front's Professional plan is $65 per seat per month for up to 50 seats. Comparing entry prices for a team that cannot fit in an entry plan produces the wrong answer.
Second, some capability is priced separately from the seat. Help Scout's AI Answers add-on is charged at $0.75 per resolution, defined as a conversation resolved without human assistance. Front lists Copilot as a $20 per seat per month add-on. A quote built from plan prices alone will be low if the plan chosen depends on those features.
Third, the thing being counted is almost always people. Very little in this category is priced on volume of messages. That is worth knowing because it means a busy month costs the same as a quiet one, which is the right shape for support traffic and the wrong shape for nothing much.
Check the ceiling against next year's headcount
The seat ceiling is the most common reason a tool chosen in good faith turns into a migration eighteen months later. A five seat ceiling is a hiring plan, not a limit. Where growth is likely, the number to compare is the price of the tier the team will be on, not the tier it starts on.
Where a team inbox is the wrong shape
The category assumes the work arrives as mail. Often it does not. A large share of what lands in a shared mailbox is a notification that something was submitted somewhere else, and the mail is a pointer rather than the thing itself.
This is the pattern with forms. Applications, service requests, event registrations and enquiries arrive as structured submissions. A form tool collects them, and then a notification lands in the mailbox saying a submission arrived. The team inbox is being asked to hold state for records it does not contain, which produces a specific and familiar mess: the conversation about a submission lives in one system and the submission lives in another, and nothing links them except a person remembering.
The symptoms are recognisable. Someone has to open the spreadsheet to see what was actually submitted before answering the mail. The status of a submission is a colour in a cell, while the status of the thread is a label in the inbox, and the two disagree. A reply goes out and the record is not updated, because updating it is a separate deliberate act that has nothing forcing it.
Where the intake is a form rather than a mail, the shape that fits is a form tool that holds the submission and the reply together, with an owner and a status on the submission itself. The assignment then belongs to the record, not to a mail thread that happens to mention it. Reading through how that plays out for applications, service requests and event intake is the quickest way to tell which shape a given workflow wants, and the use cases page is laid out by intake type for that purpose.
Nothing about this makes team inbox software a poor category. It makes it a category for teams whose work genuinely arrives as email from people who write in prose. Where the work arrives as fields, the fit is worse, and no amount of inbox tooling fixes that.
The four questions that separate otherwise similar tools
Feature grids in this category are long and mostly identical. Four questions cut through them.
Which channels count as shared, and how many are included. Some plans cap the number of shared inboxes or channels rather than the number of people. Zoho TeamInbox's Starter plan allows 3 shared incoming channels and its Professional plan allows 5, which is a limit on the shape of the team rather than on its size. A company with one address per department hits a channel cap long before a seat cap.
What happens when two people open the same conversation. The answer separates tools that genuinely prevent collision from tools that report it after the fact. Both are useful. Only one of them prevents the duplicate reply that started the search.
How rules behave when they fail. Missive lists up to 1000 rules on its Productive plan, which is far more than most teams write. The number is not the question. The question is what happens when a rule's condition stops matching, because rules fail by doing nothing, and the conversations they were meant to route go where unrouted conversations go.
Whether the audit trail survives a person leaving. Assignment history, notes and the record of who replied need to outlast the account that created them. This is rarely on a comparison page and is the first thing anyone asks after a handover goes badly.
How to run the trial so it tells you something
Most products in the category offer a trial of around two weeks, and Zoho TeamInbox lists a 14 day trial period on its plans. Two weeks is enough to answer the real questions and easily short enough to waste.
A trial run on invented traffic proves nothing. The only useful test is to point one genuine channel at the tool, with the people who will actually use it, and leave the old mailbox running alongside for the same period. Reading a demo account tells nobody how the team behaves at four in the afternoon when six things are open.
Three things are worth measuring deliberately during those two weeks. How long a conversation sits unassigned, because that number is the whole reason for the purchase and it is easy to check afterwards. How often somebody has to leave the tool to find out what a message refers to, since every one of those trips is a sign that the record lives elsewhere. And how many replies went out without the status being updated, which is the habit that decides whether the list can be trusted in six months.
Then check the one thing a trial hides. Trials usually run on the top tier. The plan that gets bought is rarely the top tier, so it is worth confirming which of the features that made the trial pleasant belong to the tier actually being paid for, and which are add-ons charged per seat or per resolution on top.
The cost that never appears in the quote
Every tool in this category will import mail history and connect to a mailbox in an afternoon. The expensive part is elsewhere.
Rules and assignment conventions have to be rebuilt, because they are not portable between products. Whatever informal system the team used, and there is always one, has to be written down before it can be recreated, and writing it down usually reveals that two people understood it differently. Anything wired to the old mailbox, from autoresponders to forwarding from a website form, has to be re-pointed, and the item that gets missed is always the one nobody remembers configuring.
Budget the convention work, not the import. A tool adopted without an agreed rule for who claims what reproduces the original problem inside better software, and the second migration is harder to justify than the first.
What to change first
Before comparing products, write down where the work actually arrives: as prose from a person, or as fields from a form. That answer decides the category, and the category decides which prices are even worth comparing. Where the intake is a form, the shorter path is to put the submission and the reply on one screen with a named owner, which is what the demo of a form tool like Halict shows in a few minutes.
Q1. What is the difference between a shared mailbox and team inbox software?
A shared mailbox gives several people the same messages and no shared state, so read status is personal and nothing records who is handling what. Team inbox software adds assignment, status and internal notes on top of the same mail. The practical difference shows up as duplicate replies, which the mailbox cannot prevent and the software can.
Q2. How much does team inbox software cost per person?
Entry plans today run from $14 per user per month for Missive to $25 per seat per month for Front on annual billing, and Help Scout lists a free plan for up to 5 users before its $25 per user per month tier. The figure to compare is the tier that fits the team's actual headcount, since entry plans carry seat ceilings of 5 to 10. Add-ons priced separately, such as $0.75 per AI resolution, can change the total materially.
Q3. Is a free team inbox enough for a small team?
A free plan is workable while the team fits under its user cap and needs no automation, and Help Scout's free plan covers up to 5 users. The limits that bite first are usually the number of shared channels and the absence of rules, not the number of people. Where a free plan is being used for anything that has a deadline attached, check what happens to the audit trail when someone leaves.
Q4. Should form submissions go into a team inbox?
They can, but the submission itself stays wherever the form put it, so the inbox ends up holding state for records it does not contain. That is the setup where a status lives in a spreadsheet cell and a label in the inbox at the same time and the two stop agreeing. Where the intake is a form, keeping the submission and the reply in one place removes the split rather than managing it.
Q5. How many shared inboxes will a plan allow?
This is capped per plan in several products and is easy to miss because the marketing counts seats. Zoho TeamInbox allows 3 shared incoming channels on Starter and 5 on Professional, for example. A company with a separate address per department should check the channel count before the seat count.