Montreal Coworking Spaces: 16-Site Desk and Office Audit

Executive Summary

This audit identifies 16 operator-documented physical locations in a deliberately limited sample of Montreal coworking spaces. It does not establish a citywide location total, a vacant-seat count or an occupancy rate. Inclusion means that a current operator source identifies a physical address, advertises a desk or office product and provides a route to enquire or arrange a visit. The address ledger below is the reproducible basis for the sample count. Checks were completed on October 15, 2026. No historical September inventory is reconstructed.

The sample covers downtown and Old Montréal, the Plateau and de Gaspé corridor, the canal, Angus, Verdun, Queen Mary and Saint-Léonard. Its product coding finds 12 locations with shared or day coworking, 14 with assigned desks and 15 with private offices. These overlapping counts measure explicit product mentions, not seats available to rent. For example, ECTO documents floating and fixed open-space desks, while SPACIAL explicitly describes offices for teams of 2 to 25. Neither disclosure supplies an available-office count. (Source: ecto.coop) (Source: ecto.coop) ^[1]

Prices require their units and conditions. Montréal CoWork advertises private offices from $705 per workstation monthly, plus taxes, while WeWork's 1010 Sainte-Catherine page advertises a two-person office from C$770 monthly, subject to a promotional, twelve-month commitment. Those are different purchasing units. At 2727, the pricing summary records a $350 monthly hot desk, assigned desks from $400, and private offices from $700 in its summary, in Canadian dollars before taxes. The same page’s detailed rate table, private-office section and FAQ instead state from $600/month, subject to live inventory. Confirm the selected room’s price before booking. These are advertised prices, not reserved inventory. ^{[2] [3]} ( 2727coworking.com

Directory totals answer another question. CoworkingCafe displays 52 spaces in its Montréal search results, whereas Coworking Québec describes a province-wide membership list. Neither is this report's denominator. Municipal boundaries also matter: Québec distinguishes Montréal city from its agglomeration and linked municipalities. ^{[4] [5] [6]} For a freelancer, the practical sequence is commute, access schedule, trial day and contract. For a small team, add an explicitly sized enclosed office and a written move-in quote. A virtual address does not substitute for workstation access: 2727 explicitly separates its business-address plans from daily desk access. ( 2727coworking.com

Introduction and Background

Searching for coworking desks in Montréal produces several different objects: a physical workplace, an operator's network page, a booking category and a directory result. A useful location audit must keep these objects separate. WeWork's own Montréal page, for example, reports three locations, while individual building pages describe several workspace products. Counting each product as a location would change the result without adding another address. ^[7]

The same distinction applies to geography. The City of Montréal lists 19 boroughs. The agglomeration includes the city and linked municipalities, while the Communauté métropolitaine de Montréal (CMM), a metropolitan governing body, has a broader municipal membership. A space described commercially as “Montréal” may therefore sit outside the city boundary used here. ^{[8] [6] [9]}

This report serves freelancers and teams choosing where to tour. It does not rank operators, claim that its sample exhausts local supply or turn a website's contact form into proof of bookability. CoworkingCafe itself distinguishes direct booking from enquiry routes that send a request to an operator. That distinction remains relevant even when a location has a detailed listing and a published price. ^[10]

The decision is ultimately a match between a product and a work pattern. A visitor needing occasional daytime seating has a different requirement from someone leaving equipment overnight. A team needing an enclosed room has a different requirement from colleagues willing to sit separately in shared space. Swivl illustrates another distinction: a membership offering bookable private-focus-office hours is a different product from a continuously assigned team office. ^[11]

The contribution here is a dated, address-level evidence trail, a conservative product classification and a shortlist process. Unknown fields remain unknown. Readers can use the ledger to identify relevant operators, then replace website-level evidence with a written confirmation for their dates, headcount and access requirements.

Audit Method, Geography and Evidence Standards

What entered the sample

The audit used varied English and French searches, the association directory and the commercial directory for discovery, then checked operator sources for the included addresses. Coworking Québec is useful for finding names and addresses, but its stated membership scope is all of Québec. It is not a complete municipal register. ^[5]

A row entered only when operator evidence supported:

• Physical address: a specific workplace, rather than a city landing page alone.
• Workspace product: shared workstations, assigned desks or enclosed offices.
• Customer route: an operator enquiry, tour or booking route.
• Geographic fit: an address within the report's Montréal city scope.
• Source trail: a fetched operator page associated with the row.
• Check date: the common October audit update, rather than an inferred historical opening date.

“Operator-documented” is deliberately narrower than independently inspected. No site visit, telephone confirmation, completed reservation or payment was conducted. Website presence supports an advertised offer, with moderate confidence in the product classification; it provides insufficient evidence for current vacancy.

Boundary and deduplication rules

The counting unit is one distinct street address, with suites recorded where supplied. Products at the same address do not create extra locations. Different addresses of the same operator remain separate. No ownership conclusion is drawn from a trading name.

City, island, agglomeration and metropolitan area are not interchangeable. Québec's Commission de toponymie separately identifies Montréal, Dorval and Mont-Royal among the agglomeration's constituent cities. The CMM's municipal directory uses a still broader territory. ^{[12] [9]}

The exclusion and exception log is therefore:

• Longueuil: Halte24-7's South Shore address is outside this city sample. ^[13]
• Royalmount: Hedhofis supplies a Mount Royal address; it is excluded under the municipal rule. ^[14]
• Dorval and forthcoming Laval: Metspace distinguishes these from Saint-Léonard; they do not add city rows here. ^[15]
• Virtual services: address-only plans do not create extra workspaces or desk supply under the audit definition.
• Cafés and meeting-only offers: excluded as standalone categories; an operator with a separate desk or office product can qualify.
• Unretrieved operators: Temps Libre and Fabrik8 were left outside the sample after unsuccessful retrieval attempts, including the permitted fallback. This is an evidence limit, not a statement about their operating status.

Evidence levels and unknowns

The audit separates advertised product, published price, enquiry route, operator-labelled availability and confirmed availability. These are different fields. A general “Available” card is retained only as operator language; it does not establish a dated seat commitment. Dominion's rate cards illustrate why the report can record a price while leaving bookability unresolved. ^[16]

Undated pages are labelled by access date. A webpage modification date is not treated as a rate guarantee. Prices are transcribed with conditions, and network prices are separated from location-specific ones. No directory total enters any sample calculation.

Address-Level Ledger of Montreal Coworking Locations

Table 1 is the core comparison and source ledger. H means shared, floating or day coworking; A means an assigned or dedicated desk; P means a private office. Y means an explicit offering was established. ? means the captured evidence did not establish that product at that address, not that it is absent.

The scope flag records evidence, not ownership: M means multiple physical locations were established in the research; S means only one site for that operator is included here and network or independent ownership status is unassessed. Every row was last checked on October 15, 2026. The final column links to an operator page from which to begin a tour, booking or enquiry.

The table is a sample of locations with documented offers, not a statement that all desks or offices can be purchased now. Metspace's shared and assigned products remain uncertain at this specific address because the captured package rates are operator-wide and availability varies by location. Its location evidence establishes an office site. ^[31]

Numeric team capacity is also mostly unknown. The report retains SPACIAL's advertised office range, Swivl's explicitly sized Queen Mary office and 2727's stated office range where useful; it does not estimate capacity from photographs, meeting rooms or a building's floor area. 2727's tour describes offices for 1 to 10 people. ( 2727coworking.com

Metro and booking fields

Nearest-metro status is unverified for every row. Operator-named nearby stations were established at Queen Mary and 2727. Swivl names Snowdon, on the Orange and Blue lines; 2727's tour names Charlevoix. Neither statement independently measures the nearest entrance or an accessible walking route. ^[32] ( 2727coworking.com

For every address, use the Société de transport de Montréal (STM) network and station-neighbourhood maps to check the route before touring. Its map resources support route planning, not a promise that the office door is accessible from a particular exit. (Source: www.stm.info) Booking status for every row remains unconfirmed; a customer route is a way to ask, not a held seat.

Analysis of Key Segments and Tour Zones

Table 2 groups the ledger into practical tour zones. These are editorial clusters, not administrative borough boundaries. The denominator is always the included address rows, and the product columns overlap. A zero means no explicit mention coded in this sample, never a claim that the zone lacks that product.

The downtown cluster offers several documented office-tour candidates within the sample, including the WeWork building pages, Crew and Hedhofis Old Montréal. The Plateau cluster provides several documented shared and assigned-desk candidates. These are shortlist observations from the ledger, not measured differences in neighbourhood supply density. ^{[33] [34] [35]}

Hot desks and assigned desks

A hot desk is unassigned shared seating under this audit's taxonomy. An assigned desk is a workstation reserved to a member. They should not be pooled when comparing convenience, equipment storage or access. ECTO explicitly distinguishes a different desk each day from a member's own open-space desk. (Source: ecto.coop) (Source: ecto.coop)

Access can matter more than the nominal monthly rate. Halte24-7's Montréal offer separates its standard daytime access from an additional charge for round-the-clock access. Swivl describes most Queen Mary coworking access as running from 6 a.m. to midnight, seven days a week, which should not be rewritten as unrestricted overnight access. ^{[36] [37]}

Where a required product looks thin in Table 2, the proper response is to expand discovery and ask operators. It is not evidence of unmet demand. A location can offer a product that the captured page did not establish, and a smaller sample can miss operators entirely.

Private team offices

A private office means an enclosed workspace product in the source, not necessarily an office that fits a particular team. A generic private-office label cannot establish whether colleagues can sit together, whether extra monitors fit, or whether a room is available on the desired date.

SPACIAL's 2 to 25-person advertised range creates a relevant enquiry route for the target team sizes. Swivl provides a more specific Queen Mary example with up to six workstations. These are product-level descriptions, not confirmations of current stock. ^{[1] [11]}

Spaces and Regus also qualify their round-the-clock access statements by customer type: private-office and dedicated-cowork customers. The same wording should not automatically be extended to a day-pass visitor. ^{[38] [39]} Readers should obtain the access conditions for the exact product quoted.

Building a Shortlist Before Booking Tours

Freelancer requiring regular desk access (Hypothetical Example)

Consider a freelancer who wants a repeatable weekly routine, sometimes works late and prefers an assigned workstation. The first filter is the assigned-desk column, followed by access and commitment. The shortlist should contain locations the person can realistically reach on working days, rather than the largest number of brands.

A Plateau comparison could start with ECTO, Montréal CoWork and Halte24-7. The published ECTO monthly rate assumes a six-month commitment, with a 20% supplement for shorter commitments. Montréal CoWork advertises dedicated desks from $450 per workstation monthly, plus taxes. Halte24-7 advertises dedicated desks at $399 monthly, plus taxes, with an additional access charge to investigate. These are conditional offers, not a single standardized market price. (Source: ecto.coop) ^{[35] [40]}

The freelancer should send the same request to each operator:

• Product: an assigned workstation, rather than an unassigned seat.
• Schedule: the earliest arrival and latest departure needed.
• Term: the desired start date and shortest acceptable commitment.
• Equipment: permission to leave a monitor and accessories.
• Calls: phone-booth access and any booking restrictions.
• Meeting use: included hours, guest rules and extra charges.
• Trial: whether a paid day can test the relevant working hours.
• Total price: currency, taxes, deposit and optional-access charges.

A trial visit should test actual tasks: taking a call, using the preferred desk setup and making the commute at the intended time. Operators' station references can help start route planning, but the reader should check the trip using the STM resources and the full street address. (Source: www.stm.info)

Six-person team requiring one enclosed office (Hypothetical Example)

For a team needing to sit together, six separate desk memberships are not automatically equivalent to an enclosed office. The initial shortlist can include SPACIAL because its published office range includes the requested size, and Swivl because its Queen Mary page identifies an office with up to six workstations. ^{[1] [11]}

The team should ask every candidate:

• Room identity: the specific office number or floor-plan reference.
• Simultaneous seating: whether everyone can work there at once.
• Furniture: desk dimensions and storage included.
• Move-in: the earliest confirmed date, not a general starting price.
• Expansion: whether a larger office can be contracted later.
• Privacy: enclosure, call arrangements and guest access.
• Shared amenities: included meeting time and booking limits.
• Commercial terms: commitment, notice, deposit and total recurring charge.

A per-workstation office rate is useful as a budgeting input, but multiplying it by headcount does not establish that a suitable room exists. Likewise, a small-office promotional price at a WeWork building does not price a larger office. ^[3]

A written response should classify each candidate as confirmed suitable, suitable subject to room/date confirmation, or not established. The useful shortlist is the set that satisfies the actual requirement. No operator in this report is assigned the first classification solely from its website.

Data Analysis and Evidence

Counts and price-disclosure coverage

The sample calculation is count of distinct included addresses = 16. Product counts are sums of the Y flags: H = 12, A = 14, P = 15. A location can contribute to all product columns, so their sum is not a location total. The documentary unknowns are 4 H fields, 2 A fields and 1 P field, calculated as the address denominator minus each product count.

The audit retains a location-specific recurring numerical price for 12 of 16 addresses, or 75%. This is an audit price-record measure, not a citywide transparency estimate. It includes a recorded dollar price even where the currency code remains unconfirmed. The retained rows are 2727, Montréal CoWork, Halte24-7, Crew, ECTO, the three WeWork buildings, Hedhofis Angus, Hedhofis Old Montréal, Swivl and Dominion. Their sources are linked in Table 1 and in the price comparison below.

Among rows without a retained numerical price, SPACIAL directs readers to contact it for pricing, Metspace's captured rates are network-level, and the selected Regus and Spaces prices were not retained because currency or booking locale was unclear. No missing value is converted to zero. ^{[41] [31]}

Published price examples and their units

Table 3 records selected offers rather than a price ranking. C$ and CAD mean Canadian dollars where explicitly established. An unqualified $ preserves the source's symbol, with currency code to confirm before purchase. Every figure was observed in the October update; no historical rate guarantee is implied.

The table demonstrates why a “cheapest Montréal coworking” average would be misleading here. It mixes an unassigned membership, a reserved workstation, an individual office, a per-workstation office charge and a larger team office. Commitment periods also differ. Currency confirmation and an exact product match must precede a numerical comparison.

Reproducible budget arithmetic

For illustration, applying Montréal CoWork's advertised private-office floor to a six-person requirement produces 6 × $705 = $4,230 monthly before taxes. This is a calculation, not a quote for an available room. ^[2] Swivl's separately advertised up-to-six-workstation office starts at $3,000 monthly on a twelve-month term; an enquiry must establish whether that product fits the team and dates. ^[11]

Revenu Québec states 5% Goods and Services Tax (GST) and 9.975% Québec Sales Tax (QST), calculated on the sale price, and specifies a combined 14.975% rate. Where both taxes apply, a before-tax charge is multiplied by 1.14975 before invoice rounding. ^{[47] [48]}

For a confirmed CAD $350 fully taxable monthly charge, the illustrative total is $350 × 1.14975 = CAD $402.41, rounded to cents. The starting charge comes from 2727's price record; the multiplier comes from Revenu Québec. This excludes deposits, optional services and any different invoice treatment. ( 2727coworking.com ^[48]

No mean or median rent, vacancy rate, market share or demand estimate is calculated. The sample is purposive, products are heterogeneous and location floor prices do not describe the distribution of actually rentable offices.

Implications and Future Directions

Use the data without overstating it

For freelancers, the audit makes a neighbourhood decision more concrete by separating floating and assigned seating. For teams, it narrows enquiry routes while keeping room capacity and current availability as explicit next steps. The practical deliverable is a smaller set of questions to resolve before spending time on tours.

A virtual-office comparison should begin with the work requirement. If the need is mail handling and a business address, desk access may be unnecessary. If the need is a place to work, an address plan alone is insufficient. At 2727, the pricing page explicitly states that business-address plans do not include daily desk access. That product distinction should be checked at every operator. ( 2727coworking.com

For local business and commercial real-estate publishers, the reusable finding is the defined sample and its address evidence. The commercial directory total should remain attributed to its publisher and search geography. CoworkingCafe's Montréal results include a Pointe-Claire listing, which helps explain why its displayed total should not be treated as a strict-city inventory. ^[4]

The report also separates site inventory from vacancy. Hedhofis's homepage attributes 42 private offices and 75 dedicated desks to Angus. Those are operator-advertised facility counts; they do not indicate how many can be rented now, and they are not converted into seats or added across the sample. ^[49]

Update log and reproducibility

Initial audit update: October 15, 2026. The sample contains the address rows in Table 1; all product and price measurements use that snapshot. No September audit, historical change rate or opening/closure trend is asserted.

A subsequent update should:

• Preserve addresses: use the street address as the stable counting unit.
• Retain source dates: record access date separately from source-stated dates.
• Recheck products: change a question mark only with new evidence.
• Capture quotes: preserve product, unit, term and tax wording.
• Confirm currency: avoid relying on the dollar symbol alone.
• Request availability: identify room, headcount and move-in date.
• Log exclusions: distinguish boundary cases from retrieval gaps.
• Recalculate: use the revised row denominator for every percentage.

The tables can be copied into a spreadsheet. An update should retain explicit unknowns rather than silently filling them from directories.

Conclusion

Montréal coworking choices become easier to compare when the unit of analysis is a physical address and the unit of purchase is an exact workspace product. This report supplies a bounded operator-documented sample, separates shared desks from assigned workstations and private offices, and leaves current bookability unresolved where the evidence cannot establish it.

The location ledger is most useful as a starting point for tours. A freelancer can filter by seating type, commute and access schedule, then compare commitment periods and the full recurring charge. A team can filter for enclosed offices, require a room sized for simultaneous attendance and obtain a written response for its move-in date. Published starting prices help form questions; they do not secure a room.

The counts describe this audit's evidence, not the whole market. Directory listings, network products, facility inventory and vacant seats have different meanings. Geographic labels also need a stable boundary before any comparison becomes reproducible.

The next decision should therefore be specific: select a workable area, identify the required product and request a dated confirmation from the operator. Retaining unknowns until that confirmation arrives gives readers a more defensible shortlist than a confident-looking total assembled from incompatible listings.