Reviews
Part of Team collaboration tools: the three jobs people usually confuse
Team collaboration plans: how Slack bills seats and limits history
Team collaboration plans turn on the seat unit and the history limit; Slack's active-member billing, 90-day free history and prices show how.
Team collaboration pricing is settled less by the per-person rate than by three things a price list rarely shows: what counts as a billable seat, what a free tier does to your message history, and the points where the bill steps up instead of rising steadily. This page works through them in order. It opens with a five-question unit checklist, uses Slack's published prices and free-tier terms as the clearest documented example of a history limit, maps the three step changes that lift a collaboration bill, describes how fragmented workspaces form, and closes with an eleven-field model you complete with figures you look up yourself.
Most organizations never model the per-person rate at all, because it looks too small to matter. They pay it, the team grows, a department buys its own workspace, contractors get added, and three years later the bill is well past what anyone expected and nobody can say when it changed. The movement was never in the rate. It was in the unit, the free tier and the steps.
Start with the unit, not the rate
The same headline rate means different things at different vendors, so settle the unit before you compare anything. Where an offer describes its unit loosely, ask for the definition in writing: a rate quoted per "user" can mean per account that exists, per account used in the period, or per person who touched the tool at all.
What the unit actually is
- Provisioned or active accounts
- Who counts as a member
- How guests are handled
- Whether connections cost anything
- What the billing period does
- Provisioned or active.Some arrangements bill for every account that exists; others bill only for accounts used in the period. If yours bills provisioned accounts, every leaver nobody deactivated and every account left over from a finished trial is on your bill indefinitely.
- Who counts as a member.A person who only reads, one who only receives automated notifications, and one who works in the tool all day may or may not be priced the same.
- How guests are handled.Free with restrictions, discounted, or full price. This is the largest single variable for any organization that works with contractors, clients or partners, and it is rarely visible on a comparison page.
- Whether connections cost anything.Automated posting from other systems is sometimes limited by count, sometimes metered, and sometimes an entirely separate tier.
- What the billing period does to changes.Whether adding someone mid-period is prorated, whether removing someone produces a credit, and whether the count is taken at one moment or averaged across the month.
Get an answer to each of these in writing, and two offers at the same rate can still differ by a large multiple once the answers are applied to a real organization. The counting itself is cheap: your directory knows how many accounts exist, and the admin console knows how many were used last month. Slack, the example used later on this page, bills the members who are active in the billing period rather than every account that exists.
A free tier is limited by history, not by seats
Free arrangements in this category are usually generous about people and strict about time. The limit is not always a window. It can be a cap on stored volume, or access withdrawn from older material rather than the material itself deleted, which sounds kinder and has the same effect on your day.
Slack's published terms are the clearest documented example available. The free subscription includes 90 days of message history: messages are saved and searchable for 90 days, and an upgrade gives access to the previous 12 months of messages. The same page answers the unit questions above: members who take an action in Slack at any time within a 28-day period are considered active for billing purposes, someone you have already paid for who becomes inactive earns a prorated credit, and new team members are charged a prorated amount for the rest of the current billing period. As served to UK visitors, it lists Pro at £7 per user per month when paying monthly and £5.75 per user per month when paying annually, and Business+ at £14.40 per user per month when paying monthly and £12 per user per month when paying annually; Enterprise+ is sold through Slack's sales team. Prices in pounds show the structure of the offer rather than a quote for a buyer in the US or Canada. On retention, the page says Free Slack deletes data after one year, while paid subscriptions keep it for life by default, with adjustable settings. Slack's paid-versus-free comparison states the boundary from the other side, listing the free arrangement as messaging with a 90-day history and paid subscriptions as unlimited messaging, and it adds that Slack cannot be upgraded for one person: to upgrade Slack, you need to upgrade your entire workspace at once. The same comparison says free teams get access to the previous 12 months of workspace data when they upgrade to Pro, and that there is no limit on the number of people in a team or a channel.
Note what kind of limit that is: a history limit, not a headcount limit, and it is the part organizations plan around last. Not every free plan works this way: Asana's free Personal plan is built for one or two people, so there the limit falls on headcount instead.
How much history you are obliged to keep is a question about your content and your policy rather than about the tier. Retention duties arrive from outside the tool. If your organization is a broker-dealer, SEC Rule 17a-4 and FINRA Rule 4511 impose record preservation obligations that reach electronic business communications. A duty to preserve evidence under the Federal Rules of Civil Procedure can attach to any organization once a dispute is reasonably anticipated. Establish which of these applies to you before you build a working culture on a free plan, because the answer decides whether a history limit is an inconvenience or a disqualifier.
Three questions decide whether the free option is viable:
- What exactly happens at the boundary?Are old messages deleted, hidden, or exportable? These are very different answers, and only one of them is recoverable.
- Does upgrading restore what was lost?Sometimes hidden history returns on payment, sometimes it is gone. Ask before you rely on it.
- What can an administrator export while still on the free arrangement?If the answer is nothing, the free tier is not free; it is deferred.
None of this makes the free option wrong. It makes it a decision with a known cost, taken deliberately, rather than a saving that turns into a bill at the moment you have the least room to argue.
Where the bill steps up
Growth in this category does not raise the bill smoothly. It jumps, usually at three points.
Where the bill jumps
- Free to paidHistory limits hurt or an obligation appears
- Paid to governance tierSSO, provisioning, retention, admin export
- One workspace to managed setDepartments bought separately must be merged
| Step | What triggers it | Why it costs more than expected |
|---|---|---|
| Free to paid | The history limit becomes painful, or an obligation appears | You convert at the moment your archive is most valuable, which is the weakest position to negotiate from |
| Paid to a governance tier | Needing single sign-on, automated provisioning, retention control, or administrative export | The higher rate applies to every account, not only to the accounts that needed the feature |
| One workspace to a managed set | Departments that bought separately have to be merged | The merge is manual, some history does not move, and both bills run in parallel while it happens |
The second row is worth planning around, because the requirement that pushes an organization up a tier is almost never a communication feature. It is a governance feature, and it arrives predictably: with the first regulated customer, the first serious dispute, or the first request for a record. Single sign-on, automated provisioning, retention controls and administrative export sit at paid levels in this category rather than in free plans, and which paid level each one starts at differs by vendor, so read the vendor's own plan comparison rather than a third-party summary. On Slack specifically, paid subscriptions are sold as Pro, Business+ and Enterprise+, and its feature comparison lists single sign-on as OAuth with Google, SAML-based and multi-SAML, so the administrative controls differ by plan.
The same trigger shows up in project management pricing, where approval and reporting needs move the tier before volume does, and in business email pricing, where the meters differ but the question is the same.
The fragmentation problem
The most expensive shape in this category is several teams paying separately for the same product.
Cost of consolidating later
- 1the merge itself
- 2the overlap where both are paid
- 3the history lost in the move
It happens naturally. One team starts on the free arrangement, another buys a paid plan on a card, a third joins somebody else's workspace as a guest. Nobody is doing anything wrong, and the result is multiple bills, no shared history, guests where colleagues should be, and no administrative view of anything.
Consolidating later costs money in three ways: the merge itself, the overlap where both arrangements are paid at once, and the history lost in the move. All three are smaller the earlier you act, and none of them appears on a price list. If you can see the pattern forming, the cheapest time to fix it is now. If it has already formed, add up the total across every card before concluding that one managed arrangement costs more; frequently it does not.
The first sign of fragmentation is not the invoice but the workaround: people pasting screenshots into mail or keeping a parallel file sharing folder because two workspaces cannot see each other. That is the symptom to watch, and it usually shows up well before anyone reads up on software integration alternatives.
Build the model with your own numbers
Fill this in with figures you look up yourself. The fields are what matter, because rates change and units do not.
Fields for your pricing model
- Accounts that exist
- Accounts used last month
- Read-only or receive-only people
- Guests and external participants
- Separate workspaces
- Tier required by governance
- History retention required
- Export capability and its tier
| Line | What to enter |
|---|---|
| Accounts that exist | Including everyone nobody deactivated |
| Accounts actually used last month | The gap between these two lines is money |
| People who only read or receive | And whether a cheaper level exists for them |
| Guests and external participants | Now, and honestly, in a year |
| Separate workspaces in the organization | Including any you only found out about while reading this |
| Tier required | Set by governance needs, not by messaging features |
| History retention required | Driven by an obligation or by how far back people genuinely search |
| Connections and automation | And whether any of them are metered |
| Export capability | Which tier it starts at, since this is your exit |
| Administrative time | Hours per month for channel hygiene, guest reviews and offboarding |
| Cost of consolidating | If you already have more than one workspace |
Run the model at today's size and at the size you would consider a good outcome in three years. Then compare the second figure with the alternative: doing nothing new, or using the collaboration features bundled with a suite you already pay for, such as Microsoft 365 or Google Workspace. That bundled option deserves to be a serious candidate rather than a fallback, and the testing sequence in our guide to team collaboration software exists to compare candidates on the same tasks rather than on feature lists. A per-seat figure from another market is not a substitute for your own quote either, as the note on per-seat prices in different markets explains.
The model is not a forecast. It is a way of putting the two lines people actually negotiate over — accounts that exist and accounts that are used — on one page before a renewal conversation starts.
Common questions
When is the right moment to negotiate?
Before you are dependent, which in practice means at your first renewal rather than your third, and while you still have a working export. Anything you agree while your entire archive sits inside the product is agreed from a weak position.
Should we pay annually?
Annual is usually cheaper and normally fixes your account count. In a team whose size moves a lot — one that leans on contractors, for instance — the monthly premium buys real flexibility. Check how mid-term additions are priced and whether the tier you need is even sold monthly, because that is where an annual commitment quietly stops saving money.
Is the per-person rate worth comparing at all?
Only after you have applied the unit definitions above. Two offers at the same rate can differ by a wide margin once guests, inactive accounts and the tier your obligations require are counted. Converting every offer into your own units, the habit our business email comparison method describes, matters more here than in most categories.
What is the most commonly missed line?
Administrative time. Somebody spends hours every month on channels, guests and access reviews, and it never appears on a price list. Cost it at a real internal rate and it frequently exceeds the subscription, which changes what "cheap" means in this category — along with the account administration overhead you were already carrying.
Does a free tier work if we have a retention duty?
Only if the vendor's free terms let an administrator export or preserve the material you are required to keep. In this category free tiers typically restrict history rather than headcount, which is exactly the wrong axis when your obligation is to keep records. Where a retention or litigation-hold duty applies, the free arrangement is usually the wrong starting point regardless of how few people use it.
