Retrospective context. The original 18 January 2018 date is retained. Microsoft reported 250 million monthly active Teams users on 2 November 2021, not 200 million daily users. Monthly activity is not a provisioned-seat count, and differing Slack metrics do not establish like-for-like market share or purchasing intent. The interesting question is not who won. It is what the shape of the win tells buyers about how enterprise software markets actually resolve — because the same dynamic is running right now in AI assistants, and the collaboration market is the cleanest recent case study available.
How the market actually resolved
Slack defined the category. Channel-based team messaging, searchable history, an integration platform that developers genuinely liked, and an onboarding experience that spread bottom-up through organisations without IT involvement. By the time Microsoft entered in late 2016, Slack was the reference product and the verb. Microsoft did not out-build it. For its first two years Teams was slower, weaker on search, thinner on integrations, and confusing to anyone who had learned channel-based chat elsewhere. What Microsoft did instead was remove the reasons to buy anything else: bundle the product into a subscription already purchased, inherit the identity and compliance footprint, add meetings and then voice so the comparison stopped being about chat, and let default enablement do the distribution work that Slack had to earn one team at a time. The result was not a product defeat. It was a distribution defeat, and the distinction matters because it predicts where the pattern repeats. A specialist product loses to a bundled one when four conditions hold: the bundled version is good enough for the median user, the incumbent owns identity and compliance for the same buyer, the switching cost of the specialist is borne by the customer rather than the vendor, and the specialist's differentiation is legible only to a minority of sophisticated users. All four held here. Slack's response — a competition complaint in Europe about the bundling of Teams with the productivity suite, which contributed to Microsoft later unbundling Teams from some commercial offerings — was an acknowledgement that the contest had moved from the product to the licence. Salesforce's acquisition of Slack in 2021 closed the independent chapter. Slack did not disappear and still holds a strong position in software, media and professional services organisations with developer-heavy cultures. It stopped being the default.
What buyers should take from it
The wrong lesson is "pick the bundle." The right lesson is about how to make a platform decision when one option arrives free inside something you already pay for. Price the switching cost, not the licence. The subscription difference is the smallest number in the comparison. Migration of history, rebuilt integrations, retrained habits and the productivity trough of a change all dwarf it, and they grow every month you defer the decision. Separate the layers of the decision. Chat, meetings, voice and document collaboration are four distinct choices that vendors deliberately present as one. Plenty of organisations ran the bundled platform for meetings and a specialist for team conversation for years, and that was coherent rather than indecisive. Make the concentration trade explicitly. Consolidating identity, email, documents, chat, meetings, voice and compliance with one vendor is genuinely cheaper and more secure to operate, and it hands that vendor full visibility of your switching cost at renewal. That is a defensible trade made once, deliberately, with the exposure written into the risk register — and an indefensible one made by default because a product appeared in a tenant. Distinguish reported activity from enablement or tenant counts, and define the measurement window. A monthly activity figure alone does not establish how a particular team works. The internal metrics worth tracking are how many teams have shifted their real discussion, and whether anyone can find a decision made six months ago.
Practical Guidance for Teams vs Slack Strategy Assessment
- Evaluate switching cost before licence cost. History migration, integration rebuild and retraining dominate the economics; the subscription line is noise by comparison.
- Unbundle the decision into chat, meetings, voice and documents. Vendors present four choices as one, and the bundled answer is genuinely strong in some layers and weak in others.
- Decide a primary platform and write it down. A product enabled by default is not a decision, and undeclared dual running is the most expensive outcome available.
- Keep a specialist only with a stated scope and an owner. Exceptions with defined boundaries are strategy; exceptions without them are sprawl.
- Record the concentration trade in the risk register. Single-vendor consolidation is defensible; unexamined single-vendor dependency at renewal is not.
- Ignore vendor adoption statistics and measure your own. Teams that moved their working conversation, and findability of past decisions, are the only numbers that matter internally.
- Test what leaves with you. Message history, files, integrations and — now — search indexes and embeddings have very different portability, and exit cost is set by the least portable.
- Revisit the decision at each renewal, honestly. Market position changed twice in five years here; a platform choice treated as permanent stops being a choice.
The Regional Angle
The Gulf market resolved faster and harder than the global picture, and for reasons that had little to do with product comparison. Microsoft's enterprise position across government, semi-government, banking and the large family groups is unusually strong, and the enterprise agreement is frequently the single largest software relationship in the organisation — negotiated at group level, renewed as a package, and administered by an integrator who already knows the stack. A specialist platform entering that environment has to win a budget line, a security review, an integrator skill set and an identity integration that the bundled option skips entirely. Regional IT functions are also leaner and more partner-dependent than Western equivalents, which raises the value of the option your existing partner can support on Monday. Historical regulation of consumer voice and video calling in the UAE added a factor with no global equivalent. Enterprise collaboration platforms were progressively permitted for licensed business use while consumer calling apps remained restricted, which meant the choice of platform carried a functional consequence that had nothing to do with features. Organisations that had been working around this found a sanctioned, bundled enterprise voice and meeting capability unusually attractive, and that shaped the market before any comparison of chat experience took place. The contest that actually matters regionally is neither of these products, though. It is WhatsApp. Client conversations, supplier coordination, internal approvals and a good deal of payment instruction still run through personal messaging, and a collaboration rollout here is an attempt to displace that rather than to migrate between sanctioned platforms. Feature comparisons are irrelevant to that contest. The argument that lands with finance and legal is governance — an approval given in WhatsApp leaves the company when the employee does, and with employment-linked residency, departures can be abrupt and complete. The argument that lands with users is a mobile experience good enough for the work, in the language they actually use. That last point is chronically underestimated. A large share of the regional workforce is frontline rather than desk-based — retail, logistics, hospitality, construction, facilities — working across Arabic, English, Hindi, Urdu, Malayalam, Tagalog, Bengali and Nepali, and often without a corporate email address. Licensing and deployment models built around knowledge workers do not reach that population, which means most regional collaboration deployments cover the office and leave the operation exactly where it was. Any market-share statistic describing this region is describing head offices. One compliance note that has grown sharper. Tenant location, archive residency and where search indexes physically live are live questions under Saudi PDPL, the UAE federal framework and the separate DIFC and ADGM regimes. Regional cloud availability for collaboration workloads arrived later than for core infrastructure, which left organisations with in-country obligations running a residency gap in precisely the system holding their day-to-day business communication.
The objection worth taking seriously
The strongest criticism of the "bundling won" narrative is that it is self-serving for the losing side and condescending to buyers. The counter-case is reasonable. Teams improved enormously after its first two years and, for most organisations, is now genuinely the better product for their situation — not on chat elegance, but on the combination that actually matters to an IT function: one identity, one compliance boundary, one administration surface, one support relationship, meetings and voice included, and a phone call to a vendor already accountable for half the estate. Buyers who chose it were not duped by distribution. They correctly valued integration over features, which is the trade enterprise IT has made for forty years and is usually right about. It is also true that Slack's own position was less robust than the category-defining narrative implied. Its pricing model became difficult at scale, its enterprise governance story arrived late, and its strongest advocacy came from engineering teams whose preferences do not always generalise to a finance department or a warehouse. A product beloved by its users is not the same as a product that solves an organisation's problem. Where the criticism of bundling retains force is in the second-order effect, and buyers should weigh it rather than adopt a side. When distribution decides outcomes, investment in the category declines, the bundled product only needs to stay adequate, and the buyer ends up negotiating renewals from a position of total dependency. Competition authorities took the concern seriously enough to act. The practical discipline that follows is small and rarely applied: know what exit would cost, keep at least one workload where a credible alternative is live, and look honestly at whether the bundled product has improved since the last renewal or merely persisted.
Common Questions
Did Teams actually overtake Slack on users?
The verified Microsoft figure here is monthly active users in November 2021. This article does not establish a comparable Slack measure or market-share conclusion. Compare equivalent definitions and periods before drawing one.
Is it ever right to pay for a specialist platform alongside a bundled one?
Yes, with a written boundary and an owner — typically where a developer or operations culture depends on integrations that genuinely do not exist elsewhere. Undeclared dual running, where both are live and nobody chose, is the outcome to avoid.
What is the most common mistake in switching platforms?
Treating it as a technology migration. Without decisions about channel structure, what belongs in chat versus documentation, and where the record of a decision lives, the new platform reproduces the fragmentation of the old one within a year.
Is the same pattern repeating with AI assistants?
It is, almost exactly, and the four conditions all hold again: the bundled assistant is good enough for the median user, the incumbent owns identity and compliance, switching cost falls on the customer, and specialist advantages are legible mainly to sophisticated users. Two things differ and both favour concentration further. First, assistants answer only from what they can reach, so the vendor that already holds your mail, documents and chat has a structural quality advantage that no independent product can match without being granted the same access. Second, the artefacts are less portable than chat history ever was — embeddings, retrieval indexes and tuned configurations generally do not leave with you, which raises exit cost in a way message archives never did. The practical implication for buyers is to ask the portability question at the start rather than the renewal: what leaves with us, in what format, and how long would it take to rebuild retrieval quality somewhere else.
Teams vs Slack Strategy Assessment — price the switching cost, unbundle the four decisions, and write the concentration trade down before renewal.