Montreal Nonprofit Market Analysis

Market analysis for Sondely in the Greater Montreal nonprofit sector, covering market structure, funder-driven demand, segment fit, a prioritized target list, product gaps that block adoption, and a discovery outreach plan.

Status: Research and hypotheses. Nothing in the target list has been validated through conversation yet. Every fit score below is a hypothesis to test, not a finding. Public figures are cited; anything uncited is an assumption and is labelled as such.

Research date: July 2026. Sector funding numbers reflect the most recent published cycles (2024 to 2025).

1. Executive Summary

The opportunity. Greater Montreal has a dense, well organized community sector where a small number of funders set the reporting rules for hundreds of organizations at once. Centraide of Greater Montreal alone invested $70.3 million across 375 community agencies in 2024 to 2025, and one of its stated evaluation criteria is an organization's capacity to measure results. Quebec's PSOC program requires an annual accountability filing from every funded health and social services organization. That means the buyer's pain is not optional, not seasonal, and not something they can talk themselves out of. It has a deadline attached.

The gap Sondely fills. Most community organizations in Montreal report activity counts (participants served, meals distributed, workshops run) because that is what their tooling produces. Funders increasingly want outcome change (did the participant's situation actually improve). The distance between "we served 400 people" and "of the 260 people we followed for six months, average stability scores rose 1.4 points" is exactly the distance Sondely's baseline-versus-latest comparison closes. The platform already computes per respondent deltas, per question direction of change, and an aggregate org level summary.

The wedge. Not individual organizations. Umbrella bodies. RIOCM groups roughly 360 Montreal community organizations. TCRI groups roughly 155 organizations serving refugees and immigrants across Quebec. The Collectif des entreprises d'insertion du Québec groups 41 insertion enterprises that take in about 3,000 trainees a year. Each of these has a shared reporting burden, a shared vocabulary, and a newsletter. One conversation with a regroupement reaches more qualified prospects than fifty cold emails.

The blocker, correctly scoped. The beneficiary-facing survey experience is already fully multilingual: an AI translation service (33 languages including French, Spanish, Haitian Creole, Arabic, Somali, Ukrainian), per-survey language selection, human review of translations, and a language switcher on the survey-take page are all built and tested. A respondent can take a survey in French or Spanish today. What is not built is the admin-facing platform chrome (dashboard, survey builder, org management, assignments, nav, buttons, form and validation messages, and the notification emails), which is still hardcoded English. In Montreal's community sector the program staff who administer surveys and file funder reports work in French, so an English-only admin console is a real adoption barrier even though the survey a beneficiary sees can already be French. A French admin UI plus French and Spanish email templates is the remaining entry-ticket work. It is a bounded i18n project, not a from-scratch build, because the harder half (survey content translation) already exists. Discovery conversations can start now. Full paid deployment should wait until the admin chrome and emails are localized. See the Platform Localization Spec.

The recommendation. Run 15 to 20 discovery conversations over the next cycle across the three highest-fit segments, entered through umbrella bodies, using the existing English product as a live demo rather than a deliverable. Use those conversations to (a) confirm or kill the funder-deadline hypothesis, (b) price the pain, and (c) decide whether the French admin UI plus PDF export is the minimum shippable Montreal bundle. Do not attempt paid pilots before the admin console and reminder emails are localized.

2. Market Structure

2.1 Size and shape

There is no clean public count of "nonprofits in Montreal." The registry counts legal entities, most of which are tiny or dormant. The number that matters for Sondely is much smaller and much more addressable: organizations that receive structured public or philanthropic funding and owe someone a report. Centraide's 375 funded agencies and RIOCM's ~360 members overlap heavily and together are a reasonable proxy for the serious addressable base in the metro. Call it 400 to 600 organizations with a real reporting obligation, of which perhaps 150 to 250 run programs where before-and-after measurement on individuals is meaningful.

That is the realistic TAM for the Montreal beachhead. It is small. That is fine, and it is arguably the point: it is small enough to enumerate, small enough to reach through five umbrella bodies, and dense enough that reference selling works.

2.2 Who sets the reporting rules

Three funders effectively define what "proving impact" means in this market. Winning them as an influence channel is worth more than any individual customer.

Centraide of Greater Montreal. Raised $72.5 million in the 2025 campaign; invested $70.3 million in 375 agencies in 2024 to 2025, reaching roughly one in five people in Montreal, Laval and the South Shore, about 800,000 people. It also invested $3.8 million across 31 neighbourhoods through the Projet Impact Collectif. Its published evaluation principles include alignment with community needs, action strategies that promote real social change, capacity to measure results, collective partnership practices, and sound governance. Centraide frames evaluation as understanding what agencies do while ensuring each investment produces concrete, durable results. Sources: Centraide investments 2024-2025, Support for agencies and projects, 2025 campaign result

PSOC (Programme de soutien aux organismes communautaires). Quebec's health and social services funding stream for community organizations. Funded organizations sign three-year financial support agreements and must file an annual reddition de comptes within 120 days of fiscal year end. The standard indicators are membership counts, activities, and participants. The purpose of the annual filing is to determine whether the organization's activities fall within the mission it is funded for. Sources: Gouvernement du Québec, Cadre de gestion du PSOC Montréal

Important nuance, and a risk to the pitch. PSOC's baseline requirement is mission compliance plus activity counts, not outcome measurement. Do not walk into a PSOC-funded organization claiming Sondely is required for their reddition de comptes. It is not. The honest pitch is that PSOC gives them a hard annual deadline and an existing data habit, and that project funders layered on top of PSOC (Centraide, foundations, ministry project envelopes) are the ones asking for outcomes. Overstating the compliance angle will get the pitch dismissed by any director who knows the program, and directors in this sector know the program.

Fondation du Grand Montréal. Community foundation running the Signes vitaux reports and, with the Communauté métropolitaine de Montréal, the Indicateurs vitaux du Grand Montréal platform, built with 82 municipalities and over 20 partners. FGM is a data-forward funder and a natural conversation for benchmarking and aggregate data, closer to a partnership than a sale. Sources: FGM Vital Signs, Indicateurs vitaux

2.3 The evaluation support ecosystem (partners, not competitors)

Quebec has an unusual amount of publicly funded infrastructure dedicated to helping the social economy measure impact. These bodies produce methodology and training but not software.

• TIESS (Territoires innovants en économie sociale et solidaire) is Quebec's reference for knowledge transfer in the social and solidarity economy, and maintains a dedicated space for social impact measurement including theory of change tooling. TIESS, Theory of change tool (PDF)
• Dynamo provides evaluation services in a community development context and has partnered with TIESS on evaluating the Montréal en commun program. TIESS project page
• Others in this space include Communagir, Coop Niska, and PhiLab (UQAM).

Strategic read. These organizations tell nonprofits what to measure and hand them a theory of change on paper. Nobody hands them the instrument that collects it. That is a clean complementarity: TIESS produces the logic model, Sondely produces the questionnaire, the recurrence, the scores and the delta. A co-marketing or referral relationship here is higher leverage than a direct sale and should be tested early.

3. Why Now: Three Forcing Functions

3.1 Funding scarcity raises the burden of proof

In Quebec, 21% of organizations raised less than 50% of their 2023 fundraising goal (Imagine Canada). When money tightens, funders concentrate on organizations that can demonstrate results, and organizations compete on evidence. Scarcity is the single strongest driver of demand for outcome measurement, and it is the current condition of this market.

3.2 Law 25 changes the tooling calculus

Quebec's Law 25 came into full effect in September 2024, applies to nonprofits, requires explicit informed consent before collecting personal information, requires privacy impact assessments before transferring personal information outside Quebec, and carries penalties up to $10 million. Beneficiary survey data on housing status, mental health, food security, or immigration status is sensitive personal information by any reading. Sources: BCLP, McCarthy Tétrault on PIAs, DPO Centre

This is Sondely's most defensible differentiator in this market, and it is largely accidental. The platform already runs on Azure Canada Central (rg-catalyzeup, Cosmos DB and Redis in region). A US-hosted competitor asking a Montreal organization to put beneficiary mental health data in Virginia triggers a cross-border transfer assessment. Sondely does not.

Caveat, and it must be honest. Canadian data residency is a necessary condition, not compliance. Sondely has no consent management UI, no data export for access requests, no deletion workflow, no retention policy, and no privacy documentation. UC-017 in the use cases doc specifies all of it; none of it is built. Claiming "Law 25 compliant" today would be false. The claim that is true today is "your data stays in Canada." Everything past that is roadmap.

3.3 Language is a market condition, not a feature request

French is the working language of the Montreal community sector. PSOC reporting is in French. Beneficiary populations in the highest-fit segments (newcomers, seniors, low-income families) are predominantly francophone or allophone. Some organizations, notably Dans la Rue and several English-community organizations, operate bilingually.

The good news is that the beneficiary half of this is already solved: the survey a respondent takes can be served in French, Spanish, and 30+ other languages today, with per-survey language control and human-reviewed AI translation. The remaining work is the staff-facing side, the admin console and the notification emails, which is a contained localization project rather than a language build from zero. The Platform Localization Spec covers it.

4. Segment Analysis

Segments are scored on four dimensions, each 1 to 5:

• Outcome fit: does the program produce measurable individual change over time?
• Funder pressure: is someone actively demanding outcome evidence?
• Recurrence fit: do beneficiaries stay in contact long enough for a second measurement?
• Reachability: is there an umbrella body or concentration point?

Segment scorecard

Segment 1: Employment integration and employability (Beachhead)

Why it is the best fit. These programs are structurally built around before-and-after. A participant enters with a baseline (unemployed, low confidence, no local credentials), goes through a defined program of fixed duration, and exits with a measurable change. The program has a start date, an end date, and a follow-up obligation. That is Sondely's data model, already implemented: assignments with due dates, recurrence, sessions, attempt numbering, baseline versus latest impact comparison.

Ecosystem and scale.

• Collectif des entreprises d'insertion du Québec (CEIQ): 41 insertion enterprises across 13 Quebec regions, roughly 3,000 trainees a year supported by about 1,000 staff, head office in Rosemont-La Petite-Patrie, Montreal. CEIQ
• Carrefours jeunesse-emploi: network serving people aged 16 to 35 with job search support, résumé workshops, and interview preparation. Multiple locations across Montreal boroughs. 211 Grand Montréal
• EXTRA network: CREMCN describes itself as one of 91 employment advice centres in the EXTRA network, which collectively help more than 80,000 people find employment in Quebec each year. CREMCN

Named targets: Collectif des entreprises d'insertion du Québec (umbrella, highest priority), individual insertion enterprises in Montreal, Carrefours jeunesse-emploi Montreal locations, YES Montreal (Youth Employment Services, anglophone), CREMCN, AIEM (Accueil et intégration Montréal-Est, runs the JEMVE pre-employability program for 16 to 30 year olds).

The pitch: intake assessment on day one, mid-program check, exit assessment, six-month follow-up. Sondely computes the delta per participant, per question, and in aggregate, and generates the narrative summary. Replaces a spreadsheet that nobody has time to analyze.

What must be true (test in discovery): they currently track outcomes in Excel or not at all; the funder actually asks for change data rather than placement counts; participants remain reachable by email six months out.

Segment 2: Newcomer settlement and integration (Beachhead)

Why it fits. Settlement is a multi-year trajectory across exactly the dimensions the Maslow template already covers: housing stability, food security, safety, belonging, language confidence, employment. The Maslow pyramid assessment is arguably a better instrument here than in any corporate wellness context it was originally imagined for. This is the segment where the existing template needs the least reinvention.

Ecosystem. TCRI (Table de concertation des organismes au service des personnes réfugiées et immigrantes), founded 1979, groups roughly 155 community organizations serving refugee, immigrant and undocumented families across Quebec, with a large Montreal concentration. It is the single highest-leverage introduction in this segment. TCRI, 2024-2025 activity report (PDF)

Named targets: TCRI (umbrella, highest priority), CACI (Centre d'appui aux communautés immigrantes), Centre social d'aide aux immigrants, Hirondelle, PROMIS, Collectif Bienvenue, and TCRI member organizations by borough.

Multi-language is both the acute need and the shipped strength here. These organizations serve people who speak neither French nor English on arrival. The survey-content translation system (English source of truth, additive translations, AI-assisted translation with human review) is already live and is a near perfect match for this segment's real need. A settlement organization can deploy the same instrument in French, Arabic, Spanish, Ukrainian and Haitian Creole and get comparable scores across all of them, which no generic survey tool gives them. This segment is where the existing translation engine is the strongest differentiator. The remaining gaps for it are the French/Spanish admin console (staff-facing) and RTL layout support for Arabic, both covered in the Platform Localization Spec.

Sensitivity warning. Immigration status is among the most sensitive data categories imaginable, and current policy conditions in Quebec make organizations in this segment cautious. Sondely's mandatory-authentication, no-anonymous-surveys design (a stated product constraint) is a real obstacle here. Some organizations will not attach beneficiary identity to responses at all. Expect this objection and go into discovery prepared to hear that anonymous or pseudonymous response modes are required. That would be a meaningful product decision, not a small one, because the entire impact comparison feature depends on linking attempts to a person.

Segment 3: Youth development and school perseverance

Why it fits. Youth programs run in cohorts on academic-year cycles with natural measurement points, and self-esteem, belonging and future orientation are exactly what a grouped or pyramid assessment captures. Recurrence is excellent because the same youth return year over year.

Named targets: Dans la Rue (homeless and at-risk youth in Centre-Sud, operates in both English and French, services span shelter, support, education, employability and housing) Dans la Rue, Réseau réussite Montréal and its partners, Boys and Girls Clubs of Montreal, Jeunesse au Soleil / Sun Youth, Maisons de jeunes network, Regroupement des Auberges du coeur.

Complication: consent for minors. Anything involving under-18 respondents adds parental consent requirements under Law 25 and organizational policy. Prefer 16-plus programs for the first pilots.

Segment 4: Mental health and psychosocial support

Why it fits. Outcome measurement is native to this field; many organizations already administer validated instruments. Sondely's grouped scoring, weighted questions, per-question rating labels and multiple result visualizations (pyramid, radar, grouped breakdown) map cleanly onto standard scale-based instruments.

Complication and opportunity. This segment will ask about clinical validity, about which validated instruments are supported, and about clinical data handling. Sondely does not ship validated clinical instruments. It ships a builder that could carry one. That reframes the pitch from "our assessment" to "your instrument, our engine, your longitudinal data." That is a defensible position and worth testing, but it puts the burden of instrument selection on the customer, which some will welcome and some will not.

Named targets: Revivre, AMI-Québec (anglophone, caregivers), Suicide Action Montréal, Réseau Avant de Craquer member organizations, CAP Santé Mentale.

Segment 5: Housing, homelessness and shelters

Why the funder pressure is highest and the fit is still only partial. Homelessness is the most politically visible file in Montreal and carries heavy reporting demands from all levels of government. But the recurrence assumption breaks: people in acute crisis are not reachable for a follow-up survey by email in three months, and asking someone in an emergency shelter to complete a rating-scale assessment is often inappropriate.

Where it does work: transitional and supportive housing, and reintegration programs, where the person is housed and stable for months. Welcome Hall Mission reports that 200 people who sheltered with them transitioned into permanent housing in the past year; that transitional cohort is exactly the measurable population. Welcome Hall Mission

Named targets: RAPSIM (Réseau d'aide aux personnes seules et itinérantes de Montréal, umbrella, member of RIOCM), Welcome Hall Mission / Mission Bon Accueil, Dans la Rue, Chez Doris (women, now including overnight shelter), Old Brewery Mission, Maison du Père, L'Itinéraire.

Positioning note: pitch the housing continuum, not the emergency service. Confusing the two will make the pitch sound naive to anyone in the field.

Segment 6: Seniors and social isolation

Isolation and well-being are measurable and change slowly, which suits recurrence. Funder pressure is real but less acute than in employment or homelessness. The blocker is digital access: a self-serve web survey excludes a large share of the population, so realistically the intervener administers it during a home or phone visit, which changes the UX requirement substantially (a staff-facing entry mode rather than a respondent-facing link).

Named targets: Little Brothers / Petits Frères, Le Réseau des centres d'action bénévole, FADOQ regional chapters, Contactivity Centre.

Segment 7: Food security and emergency aid (deprioritize)

High funder pressure and excellent reachability (Moisson Montréal, Welcome Hall Mission's food operation, which the organization describes as the largest direct-to-public food security operation in the province), but the fundamental measurement is throughput, not individual change over time. A food bank counts baskets. It does not run a six-month longitudinal assessment on the person picking one up, and pushing it to would be both unwelcome and undignified.

Where a narrow opening exists: organizations that pair food aid with accompaniment or budgeting support and follow households over time. That is a thin slice and not worth building a campaign around.

5. Product Gap Analysis

Assessed against the Implementation Status audit (86 done, 1 partial, 15 not done as of April 2026).

5.1 Blockers: cannot sell into this market without these

5.2 Strong blockers for specific segments

5.3 Meaningful but not blocking

5.4 Strengths already shipped that matter here

These are real and should anchor every demo:

• Baseline versus latest impact comparison with per-respondent and per-question deltas, direction classification, aggregate stats, and auto-generated narrative summary. Fully implemented. This is the product.
• Assignments with due dates, recurrence with auto-reset, overdue tracking, extend and waive. Matches how programs actually run.
• Escalating reminder engine with quiet hours, per-survey mute, one-click unsubscribe. Directly addresses the completion-rate problem.
• Grouped and structured assessments with pyramid, radar and grouped breakdown visualizations, weighted scoring, and the Maslow template.
• Public survey links with QR codes, sessions with auto-save and resume.
• REST API with org-level API keys, for the minority of organizations that have a data person.
• Azure Canada Central hosting.

Honest reading of the product's readiness: the measurement engine is genuinely strong and more complete than most of what small nonprofits use today. The gaps are at the edges, in language, in output, and in consent. Those edges are exactly where Montreal lives.

6. Competitive Landscape

Defensible position for Sondely in this market, in priority order:

1. Data stays in Canada. No competitor on the list can say this to a Quebec organization without qualification.
2. Bilingual by design (once shipped), with an architecture built for additive languages beyond French.
3. Priced for an 8-person organization, not a 200-person one.
4. Longitudinal impact is the core feature, not an upsell tier.

Weakness to be honest about: no track record, no reference customers, no sector credibility. In a small, tightly networked, relationship-driven sector, that is a heavier disadvantage than any missing feature. The first two or three named reference organizations are worth more than the next ten features.

7. Target List and Outreach Plan

7.1 Tier structure

Tier 0: Umbrella bodies (do these first). Not sales conversations. Ecosystem conversations. The goal is understanding the shared reporting burden and, if it goes well, an introduction or a newsletter mention.

Tier 1: Best-fit direct targets. Employment integration and settlement organizations in Montreal, roughly 15 to 25 names drawn from CEIQ members, Carrefours jeunesse-emploi locations, TCRI Montreal members, plus YES Montreal, CREMCN, AIEM, CACI, PROMIS, Hirondelle.

Tier 2: Strong fit, second wave. Dans la Rue, Boys and Girls Clubs of Montreal, Jeunesse au Soleil, Réseau réussite Montréal partners, Regroupement des Auberges du coeur.

Tier 3: High pressure, partial fit. Welcome Hall Mission, Old Brewery Mission, Chez Doris, Maison du Père, focused strictly on transitional and supportive housing programs.

Do not contact yet: food banks, arts and culture, environment, sport. Fit is too weak to justify the conversation.

These are publicly known organizations matched to segments by program type. No contact names, emails or relationships are recorded here, and none should be assumed. Named-person contact details must be gathered directly and consensually, and must not be scraped.

7.2 Approach

Frame every first conversation as research, because that is what it is. The survey experience can be French today, but the admin console and reminder emails are not yet localized, so the product is not ready for an unsupported paid rollout in this market. Attempting a hard sell before the staff-facing side is localized risks burning a small, well-connected sector. The honest and effective opening is: we are building an impact measurement tool for community organizations, we are Montreal-based, the surveys already run in French and Spanish, and we want to understand how you report to your funders before we finish localizing the rest. We would like 30 minutes.

That framing is true, it is flattering rather than extractive, and it produces better information than a demo would.

Sequence:

1. Umbrella bodies first (Tier 0). Understand the shared burden, and earn the right to be introduced.
2. 15 to 20 discovery conversations across Tier 1, weighted to employment integration and settlement.
3. Demo the existing English product live in those conversations as evidence of seriousness, never as something they can use tomorrow.
4. Identify two or three design partners willing to shape the French admin release in exchange for early free access.
5. Ship the French/Spanish admin console and localized emails, plus PDF export.
6. Convert design partners to paid pilots. Ask each for a reference and a quote.
7. Use those references to go back to Centraide, TIESS and the regroupements, this time with proof.

7.3 Discovery script

Ten questions. The goal is to leave with facts, not interest.

1. Who do you report to, and how many separate reports do you file a year?
2. What exactly does each of them ask for? Activity counts, or change in the people you serve?
3. Walk me through how you produced your last funder report. What tool, how many hours, who did it?
4. Do you measure anything about a participant at the start and again later? If yes, how? If no, what stops you?
5. What percentage of participants are still reachable six months after the program ends?
6. What language do participants take anything you give them in? How many languages do you need?
7. Has a funder ever asked you for something you could not produce? What was it?
8. Where does participant data live right now, and who has access to it?
9. Has Law 25 changed anything about how you handle participant data?
10. If a tool did all of this for you, who in the organization would have to approve buying it, and out of which budget?

Kill criteria, decided in advance. If fewer than half of Tier 1 respondents say a funder has asked them for outcome change data, or if fewer than half report meaningful six-month reachability, the Montreal beachhead thesis is wrong and the segment priorities in section 4 need to be rebuilt before any French work is justified by this market alone.

7.4 Pricing hypothesis

No monetization is implemented (subscription tiers and billing are on the not-started list), so this is a starting point for the pricing question in discovery, not a plan.

Reality check. Community organizations in Quebec run on thin, restricted, non-discretionary budgets. Software is frequently unfundable out of program envelopes. Two hypotheses worth testing explicitly in discovery: (a) the cost is embedded as an evaluation line item in project grant applications, which funders do routinely fund; (b) the regroupement or the funder pays for member access. The second is the more interesting business and should be tested with Centraide and CEIQ directly.

8. Risks

9. Immediate Next Steps

1. Build the concrete Tier 0 and Tier 1 contact list (organization, program, public contact route). Roughly 25 to 30 names.
2. Book the first five conversations, starting with CEIQ, TCRI and RIOCM.
3. Run the 10-question discovery script. Record answers verbatim into a shared sheet, one row per organization.
4. In parallel, close the two highest-leverage demo gaps: side-by-side attempt comparison UI, and a Program Impact Baseline template. Both are small relative to their demo value.
5. Ship the gating release: French/Spanish admin console and localized notification emails (survey content translation is already done), plus PDF export. See Platform Localization.
6. Decide the pseudonymous-response question before Segment 2 outreach goes past discovery.
7. Review this document after 10 discovery calls. Update or kill the segment priorities in section 4 based on what was actually heard.

Sources

• Centraide of Greater Montreal, 2024-2025 investments
• Centraide of Greater Montreal, support for agencies and projects
• Centraide du Grand Montréal, 2025 campaign result
• Gouvernement du Québec, soutien aux organismes communautaires (PSOC)
• Cadre de gestion du PSOC Montréal (PDF)
• RIOCM, membres
• TCRI, qui sommes-nous
• TCRI, rapport d'activité 2024-2025 (PDF)
• Collectif des entreprises d'insertion du Québec
• 211 Grand Montréal, directory of community organizations
• 211 Grand Montréal, job search resources
• CREMCN
• Dans la Rue
• Welcome Hall Mission
• Fondation du Grand Montréal, Vital Signs
• Indicateurs vitaux du Grand Montréal
• TIESS, who does what
• TIESS, theory of change tool (PDF)
• TIESS, évaluation Montréal en commun
• Institut de la statistique du Québec, survey of non-profit organizations, cooperatives and mutual companies
• Imagine Canada, about the sector
• Imagine Canada, trends Q4 2024
• Imagine Canada, Statistics Canada nonprofit sector data
• BCLP, Quebec Law No. 25
• McCarthy Tétrault, Law 25 privacy impact assessments
• DPO Centre, Quebec Law 25 guide
• Sopact, impact measurement software comparison
• Sopact, survey software for nonprofits
• UpMetrics pricing overview