Best unified inbox for enterprises: help desk or routing?

Ask for the best unified inbox for enterprises and you get ranked lists that put help desks and work-routing tools in the same table. They’re different products, and they don’t compete with each other.

Working out which one you need rules out every vendor in the other category.

The two products

A shared support inbox. Several people work one queue out of an address like support@ or billing@. What buyers shop for: assignment and internal comments on a thread, saved replies, SLA timers, escalation rules, and reporting on first response time, resolution time and volume. Front, Zendesk, Intercom, Help Scout and Gorgias all sell a version of it, and Missive sells the collaborative half with team chat sitting next to the thread. The buyer is usually a support or customer success leader, measured on response time.

Cross-channel work routing. A request arrives wherever the customer or the colleague happened to be, and somebody has to turn it into work with an owner. The mailbox is one of the places it lands. The buyer needs software that reads every channel, recognizes the request inside a message, assigns it, and runs the routine parts before anyone opens the thread. Vendors approach it from different directions. Automation platforms such as Zapier, Make and n8n route messages by rules you write. Chaser turns Slack messages into tracked tasks. Skydive’s AI coworkers take on work across Slack and email. Town is an AI assistant that works across a person’s email, calendar and Slack, with plans for teams. Viktor is an AI coworker that takes requests in Slack and Microsoft Teams. The buyer is usually an operations leader or a COO, measured on how much gets handled without anyone triaging it by hand.

A related question whose answer is a help desk product

People also ask which centralized inbox platform lets multiple teams manage conversations with automated FAQs and performance tracking. Unpack the question and you’ll find it has three parts, each of which maps to a different piece of a help desk.

• Multiple teams on one set of conversations means queues with routing rules, per-team views, and permissions that stop the billing team reading the recruiting mailbox.
• Automated FAQs means deflection against a knowledge base, either in a chat widget before a ticket exists or as a suggested reply inside one.
• Performance tracking means first response time, resolution time, volume by queue and per-agent numbers, on a dashboard somebody reviews every month.

That is a help desk specification, written by someone with a measured problem. SuperOffice tested 1,000 companies for its customer service benchmark and found an average response time of 12 hours and 10 minutes, with 62% of companies never replying to a customer service email at all.

If those three requirements are yours, buy a help desk product and stop reading comparison posts that mix the categories together.

What enterprise adds to either one

The enterprise tier of almost any product is set apart by a list of features and terms procurement asks for: single sign-on, directory provisioning so accounts appear and disappear with the employee, role-based access, an audit trail, retention controls, a data processing agreement with the standard contractual clauses, a published sub-processor list, a questionnaire answered row by row in a spreadsheet, a published uptime commitment, and a bill that arrives as an invoice on net terms. None of these change what the product does for the person using it, but without them few enterprises will be willing to buy it. A product can be excellent at working a support queue and impossible to get through a security review, or the reverse.

IT, security and procurement vet the product and the company that builds it, usually after somebody else has picked a favorite. For software that reads company mail, that vetting starts at the first meeting instead, because the security team is evaluating access to every conversation the company has.

That bar is rising. McKinsey’s State of AI 2026 reports 40% of large organizations, meaning those above $1 billion in revenue, now scaling AI agents in one or more functions, up from 27% a year earlier. Agents acting on company data has become commonplace, so the governance questions that used to close a deal now open it.

Where we fit

this+that is in the second category, work routing. We read Gmail, Outlook, Slack, Microsoft Teams, and Google Chat as one stream, turn what arrives into tracked work, and run the follow-through. We aren’t a help desk.

• Every connected channel in one app. Each keeps its own stream, and a request that arrives in team chat gets handled the same way as one that arrives by email.
• DoBox reads each conversation and turns what it asks for into a task on a board, with the thread attached and a person’s name on it.
• Workflows do the routine part. A trigger can be an email sender, domain or label, a chat message, a DoBox event, a new contact or Brain page, a GitHub event, a schedule, or a person pressing run. The workflow can branch on a condition, loop over a list, call a connected tool over MCP, ask the AI for a judgment call mid-run, draft or send the reply, and hold anything that needs a person at an approval gate before it goes out.
• The Brain grounds the answers, so a reply to a question your team has answered before comes from what the team decided rather than from an AI’s guess. The customer support page shows how that works for a support team.
• Contacts build themselves from the channels you already connected, so whoever picks up an account next starts from a reconciled record.

On the enterprise requirements, we cover part of the list. We have SOC 2 Type I as of July 31, 2026, with Type II underway, and CASA Tier 2. Our DPA sets out GDPR Article 28 terms with the standard contractual clauses and the UK addendum, we publish a sub-processor list with at least 10 days’ notice and an objection right before anyone new is added, and your messages are never used to train AI models, ours or our providers’. The Enterprise plan includes SSO (SAML) and SCIM provisioning. We commit to 99.5% uptime each calendar month, published rather than negotiated deal by deal. The security page has the rest, and the report itself goes out under NDA.

What this+that costs

Team is $35 per user monthly, or $29 per user billed annually, and includes 3,000 credits per user each month. Enterprise starts at $125 per user per month and includes 7,500 credits per user, pooled across the team, plus SSO, SCIM, a quarterly business review and enhanced support. You can top up either plan at any time. Monthly credits don’t roll over, but purchased credits never expire. How credits work covers what each action costs.

Self-serve signups get 30 days with the whole product and no credit card. Enterprise has no trial. A team that wants to try before signing an enterprise agreement signs up self-serve and moves over afterward.

Six questions that settle which product you need

1. Who’s buying, and what are they measured on? A support leader measured on response time is buying a help desk product. An operations leader measured on how much gets handled without anyone triaging it by hand is buying work routing.
2. Does the work arrive at one address or at everybody’s? A queue in front of support@ is a help desk problem. Requests landing in individual mailboxes and multiple chat channels are a routing problem.
3. Once the reply goes out, is the conversation finished? If it is, buy a help desk product. If the reply creates work in various other systems, buy something that can act in them.
4. Which of your channels does the product actually read? Ask for the list rather than reading the integrations page. A request that arrives in a chat channel nobody connected never reaches the automation, whatever else the product covers.
5. When does the security review start? For anything that reads mail, assume week one. Ask for the SOC 2 report, the DPA and the sub-processor list before the demo.
6. What happens when a similar request comes in? The same workflow should run again, without anyone setting it up twice.

Frequently asked questions

Is a unified inbox the same as a shared inbox? No, and the two get bought by different people. A shared inbox is one mailbox several colleagues work out of together, which is how a help desk starts. A unified inbox is every channel a person or team receives on, collapsed into one stream. A company can need both, from two vendors.

Can one vendor cover both products? Large platforms come closest, but what’s usually sacrificed is depth on the routing side. Check which channels the product reads and what it can do after the reply is sent, because a suite that handles a support queue well often has nothing to say about work that continues in a finance tool or a CRM.

Does this+that replace Front, Zendesk or Intercom? For a team running a real support queue, no. Those products do queue management, SLAs and support reporting properly. Teams that choose us instead usually aren’t running a support desk: their work arrives across everyone’s channels and needs routing and follow-through rather than ticketing. Front and Missive have the longer comparisons.

Does this+that support SSO and SCIM? Both come with the Enterprise plan. They’re configured per organization during onboarding rather than switched on from a settings page.

We’re 400 people. Is this+that the wrong size for us? Possibly, and the test is which product you need rather than the headcount. If most of the work in the company still arrives as messages to individuals and gets routed by hand, the product does the same job at 400 that it does at 40. If what you need is a support floor with staffing reports, headcount isn’t the reason we’re the wrong answer.

If what you need is work routing across channels, get started with this+that, or write to sales@thisandthat.chat to start the security review.