TAM / Selected-market opportunity
$53.3BModeled annual account spend
About 8.9M assumed buying accounts × $6K annual organization spend.
The evidence behind the thesis
Research.
With its limits in view.
Primary sources, product boundaries and the questions that should shape an investment conversation.
Reviewed 1 November 2026 · Public research brief
01 / Method
Three kinds of evidence.
External evidence
Official public research, government statistics and vendor documentation. Each source below has a date, a supported finding and a limit. Vendor descriptions establish what a vendor offers, not independent performance or customer satisfaction.
Product implementation
Review of WRKZY’s website, web application, mobile source and maintained acceptance records. Source code can establish that a feature is implemented; it cannot by itself establish production configuration, public distribution, customer adoption or a business result.
Strategic hypotheses
The initial buyer, packaged roles, distribution strategy, defensibility mechanisms and roadmap gates are proposed directions. They require customer and operational evidence. They are not historical performance, contracted forecasts or an approved financing plan.
02 / Product review
What the foundation supports.
The current product links conversations, customer records, commercial activity and work ownership. The AI workforce is the direction built on top of that foundation.
Current customer operations
WhatsApp and email inboxes, CRM records, pipelines, catalog, quotes, campaigns, tasks and configured automations. Workspace access, approval surfaces and history provide controls for the work people do today.
Inspect the product areas ↗Current AI assistance
Summaries, drafts, rewrites and suggestions use bounded context. Reviewed workspace settings keep automatic AI sends and CRM changes disabled. AI-generated automation proposals create inactive drafts for review, rather than silently enabling workflows.
Review the AI capability boundary ↗Mobile implementation
Native iPhone and Android apps support daily customer work. The reviewed October acceptance record documents internal TestFlight and Android build distribution, with public store release and physical production acceptance remaining separate gates. Screens on the website are labeled interface previews using synthetic records.
View the mobile previews ↗Calls and voice
Customer call records and phone handoff are human work surfaces, with availability dependent on the device and configuration. They are not a voice AI agent, an audio recording or a transcription service.
Read the calls scope ↗Future governed execution
Autonomous multi-step goals, proactive monitoring, voice agents and coordinated AI roles remain direction. Existing permissions, approvals and workflow history are building blocks; full agent release gates still require evaluated actions, scoped identity, recovery and complete execution traces.
See the capability gates ↗The capability boundary
Today, WRKZY can summarize, draft, recommend, and run configured workflows. Autonomous multi-step agents are part of the product direction and will launch only with scoped permissions, approvals, and complete audit history.
The review establishes implementation and documented status as of 1 November 2026. It is not an independent security audit, certification, uptime assessment or validation of commercial traction.
Timing / Research and interpretation
Why this operating layer matters now.
More business work is digital
ASUSE reports rising business internet use in covered Indian establishments. This provides more digital context; it does not establish paid adoption or serviceability.
MoSPI · digital participation ↗Model capability is more accessible
Stanford HAI’s historical comparison shows lower inference costs at a fixed capability level. Full workflow delivery still includes integration, review, retries and support.
Stanford HAI · historical inference costs ↗Delegation needs context and controls
Microsoft reports growing agent activity in its ecosystem. WRKZY’s interpretation is that business records, permitted actions and observable outcomes become more valuable as execution expands. This is our thesis, not third-party validation.
Microsoft · ecosystem activity signal ↗Competition / Alternatives and strategic tests
A plan to win must meet real alternatives.
The overview describes WRKZY’s approach. These alternatives already have strengths; comparative deployment quality, retained value and economics need direct evidence.
AlternativeExisting strengthThe test for WRKZY
Customer records, distribution and increasingly capable agents inside an existing suite.
Can WRKZY get a focused team to a useful, connected workflow with less setup and ongoing administration?
Agent builders and workforces
Microsoft Copilot Studio · Relevance AI · Lindy
Flexible tools, integrations, orchestration and enterprise or team deployment options.
Can packaged customer roles reduce the configuration and maintenance customers must do themselves?
A focused support outcome or an accessible role-based AI employee proposition.
Can connected commercial context and follow-through justify choosing WRKZY for a complete customer workflow?
Native access to customer messaging and distribution through Meta’s business ecosystem.
Can WRKZY create durable value across customer records, teams and commercial work beyond the channel?
03 / Global market context
Global context.
A separate launch model.
The company ambition is global. The selected international model below is a company opportunity sensitivity; India remains a separate entry case. These worldwide category estimates and forecasts provide context rather than a bottom-up buyer count.
$128B / Existing market
Historical market estimate · Worldwide CRM · 2024
Broad CRM taxonomy. This is category context, not WRKZY’s TAM or a claim that every CRM budget is serviceable.
Read the primary research abstract or release ↗Up to $234B / Category transition
Disruption forecast · Application spending exposed through 2030
Potential redistribution of existing spending, not incremental market growth or obtainable WRKZY revenue.
Read the primary research abstract or release ↗$985B / Workforce opportunity
Forward-looking spending forecast · Agentic AI software · 2030 forecast
Broad software forecast with adoption uncertainty; not current spending, an AI-employee-only market or WRKZY’s addressable revenue.
Read the primary research abstract or release ↗The scopes overlap and reference different years. Do not sum these figures, treat a forecast as observed revenue, or label them WRKZY TAM/SAM/SOM. Regional population definitions and dates are recorded in the source register.
Regional context / Different populations
Explore the global business base.
A worldwide ambition
Business work crosses borders.
Explore distinct regional populations. These are market contexts, not a customer footprint or committed rollout map.
United States
6.37Msmall employer firms
SBA / Census · 2022 reference year
The national profile identifies employer firms with fewer than 500 employees. This employer-only cohort is more relevant to team workflows than a total dominated by owner-only businesses.
Reference year 2022, published in the 2025 profile. Excludes nonemployers; neither workload fit nor software spending is observed by this firm count.
Read the official source ↗Definitions and reference years differ. We do not sum these populations into a global census. International rollout requires local channel, language, legal, support and willingness-to-pay evidence.
Company opportunity / Selected international markets
From a larger opportunity to a venture outcome.
The India entry case is not the company ceiling. This model explores selected employer-business cohorts and organization-level monetization, then makes the distribution required for meaningful ARR explicit.
US · UK · Canada · Australia — about 9.9M employer businesses before assumed consolidation. Selected cohorts, not a worldwide census; India, the EU and other markets are excluded.
Scale case · modeled opportunity, not observed market spend or company guidance
SAM / Potential expansion fit
$8BConditional serviceable opportunity
About 1.3M modeled accounts after 30% workload fit and 50% deployment fit.
SOM / Five-year capture illustration
$101MIllustrative year-five ARR
About 16.8K retained accounts; 1.3% of this modeled SAM. A capture case, not a forecast.
Assumptions: $6K/organization/year (not current pricing or observed ARPA); 90% buying-account factor; 30% workload fit × 50% deployment fit. Potential SAM is unverified expansion. Cohort definitions and years differ. Capture uses 90% annual logo retention and year-end acquisition; in-year churn could lower results. Not a forecast.
See the five-year capture path and execution requirements
- $1.5MYear 1
- $5.8MYear 2
- $17.3MYear 3
- $45.5MYear 4
- $101MYear 5
$0 baseline
The venture outcome to test
What reaching $100M ARR requires.
16,667 retained paying accounts at $6K aggregate annual account revenue. That is 1.3% of the potential SAM in this case.
The distribution engine
The capture illustration’s final year requires about 4,167 qualified opportunities/month at 20% conversion, and 834 new-account onboardings/month.
This is a hypothetical acquisition path, not demonstrated demand or onboarding capacity. Customer value, retention, acquisition cost and contribution margin after AI delivery and human review need evidence.
| Covered market | Source cohort | Definition, reference and derivation |
|---|---|---|
| United States | 6,374,594 | Small employer firms, fewer than 500 employees 2022 data · SBA 2025 national profile National all-industry small-employer total; excludes nonemployers and 21,041 large firms. Official source ↗ |
| United Kingdom | 1,409,395 | Private-sector SME employers, 1–249 employees Start of 2025 · DBT, 2 October 2025 1,417,730 employers minus 8,335 businesses with 250+ employees (Table C). Official source ↗ |
| Canada | 1,096,141 | Employer enterprises, 1–499 paid employees December 2024 · ISED 2025 report 1,079,188 small plus 16,953 medium employer enterprises; excludes unclassified enterprises. Official source ↗ |
| Australia | 990,837 | Actively trading businesses, 1–199 employees 30 June 2026 · ABS, 18 August 2026 996,203 employing businesses minus 5,366 with 200+ employees. Official source ↗ |
These are distinct national statistical cohorts, not harmonized size bands or globally unique procurement organizations. The UK excludes owner-only businesses; Canada uses enterprise counts; Australia uses actively trading businesses. The US uses the size-specific SBA/Census profile rather than the FAQ’s inconsistent employer classification.
| Assumption | Selected value | What must be validated |
|---|---|---|
| Source records → buying accounts | 90% | Assumed reduction for shared procurement or overlap; not observed deduplication. |
| Aggregate annual organization spend | $6K | Total workspace/usage/role revenue once; not a catalog quote or observed ARPA. |
| Customer-workload fit | 30% | Share with a recurring, valuable workload suitable for the proposed offering. |
| Deployment fit within that workload | 50% | Future channel, language, integration, trust and rollout suitability; not verified present eligibility. |
| Annual logo retention | 90% | Applied to the preceding year-end account balance. |
| Qualified opportunity → paid conversion | 20% | Used to reverse-engineer required pipeline; not observed visitor or signup conversion. |
| New paid account targets, years 1–5 | 250 / 750 / 2,000 / 5,000 / 10,000 | Hypothetical capture inputs; not an approved sales plan or demonstrated demand. |
| Driver | Year 1 | Year 2 | Year 3 | Year 4 | Year 5 |
|---|---|---|---|---|---|
| New paid accounts | 250 | 750 | 2,000 | 5,000 | 10,000 |
| Accounts lost from opening balance | 0 | 25 | 98 | 288 | 759 |
| Year-end retained accounts | 250 | 975 | 2,878 | 7,590 | 16,831 |
| Year-end run-rate ARR | $1.5M | $5.8M | $17.3M | $45.5M | $101M |
| Required qualified opportunities / month | 105 | 313 | 834 | 2,084 | 4,167 |
| Required new-account onboardings / month | 21 | 63 | 167 | 417 | 834 |
Reproduce the model
TAM = selected national employer cohorts × assumed buying-account share × annual organization spend.
Potential SAM = modeled TAM accounts × workload fit × deployment fit × the same annual spend.
Year-end accounts = opening accounts × retention + new accounts capped by remaining SAM. SOM = year-five retained accounts × annual spend.
Simplified year-end acquisition timing: no same-year churn on newly acquired accounts. Constant annual spend; no separate expansion revenue, tax, FX conversion, sales-cycle timing, budget or staffing model. Monthly requirements are rounded upward. Starting from zero is a comparable scenario convention, not a statement about WRKZY’s actual customers. The model shows what an outcome would require; it does not establish that demand, capital or delivery capacity exists.
Market model reviewed 1 November 2026. USD throughout; scenarios are sensitivities, not confidence intervals or financing valuations.
04 / India entry-market illustration
An India entry case, not the company ceiling.
This original India-only model remains useful for examining a narrow customer segment and a small acquisition engine. Its year-five output is not a ceiling on WRKZY’s global opportunity; international and workforce monetization are modeled separately above.
Expand the India scenarios and source calculations
Working case / Illustrative planning assumptions
TAM / India entry market
₹8,226 croreModeled annual opportunity
About 16,45,170 buying accounts with the relevant customer workload.
SAM / Current product fit
₹2,468 croreModeled serviceable opportunity
About 4,93,551 accounts after digital-readiness and supported-workflow filters.
SOM / Year-five capacity case
₹14.1 croreIllustrative year-five ARR
About 2,817 retained accounts, derived from acquisition, onboarding and retention.
All three outputs use INR and annual account revenue. TAM and SAM are annual revenue opportunities; SOM is a year-five run-rate ARR scenario, not cumulative revenue or company guidance. Inputs below are assumptions unless marked as survey data. ₹1 crore equals INR 10 million.
| Input | Value | Evidence status / purpose |
|---|---|---|
| Trade / other-services locations | 2,42,15,500 / 3,35,78,100 | ASUSE 2025 Table 2; counts converted from hundreds. |
| Hired-worker shares | 16.47% / 12.62% | ASUSE Tables 5 and 6; at least one regularly hired worker, not a five-person team. |
| Derived hired-worker locations | ≈ 82,25,849 | Sector counts × their respective hired-worker shares. Approximate because survey percentages are rounded. |
| Locations per buying account | 1.25 | Assumption: consolidates locations that may share one subscription decision. |
| Relevant customer-workload share | 25% | Assumption: recurring enquiries, quotes, follow-ups and team handoffs. |
| Annual revenue per paying account | ₹50,000 | Hypothetical subscription spend; not observed ARPA, a product quote or AI employee pricing. Excludes taxes and provider fees. |
| Digital readiness within the TAM cohort | 50% | Assumption; the survey’s all-establishment internet rate is not used as a conditional rate for this cohort. |
| Supported-workflow fit within ready accounts | 60% | Assumption: current channels, integrations, operating needs and rollout fit. |
| Qualified opportunity → paid account | 15% | Assumed annual cohort conversion; not visitor-to-signup conversion or observed funnel performance. |
| Annual account retention | 85% | Assumption applied to each preceding year-end balance; new accounts enter at year end. |
The five-year bridge to SOM
Annual new accounts are capped by the lower of converted qualified opportunities and onboarding capacity, then by remaining SAM. Each year keeps the retained opening balance and adds new accounts once.
| Driver | Year 1 | Year 2 | Year 3 | Year 4 | Year 5 |
|---|---|---|---|---|---|
| Qualified opportunities / month | 60 | 150 | 300 | 500 | 800 |
| Onboarding capacity / month | 10 | 25 | 50 | 80 | 130 |
| New paid accounts / year | 108 | 270 | 540 | 900 | 1,440 |
| Accounts lost from opening balance | 0 | 16 | 54 | 127 | 243 |
| Year-end retained accounts | 108 | 362 | 848 | 1,620 | 2,817 |
| Year-end run-rate ARR | ₹0.5 crore | ₹1.8 crore | ₹4.2 crore | ₹8.1 crore | ₹14.1 crore |
Year-five retained accounts equal 0.6% of modeled SAM in this case; that share is an output, not an input. Counts are rounded for display; calculations retain precision. Scenarios start from zero, assume constant annual spend and exclude expansion revenue, in-year billing timing and market growth.
Reproducible calculation
TAM = hired-worker locations ÷ locations per account × workload fit × annual account spend
SAM = TAM × conditional digital readiness × supported-workflow fit
Year-end accounts = prior accounts × retention + capped new accounts
Year-five SOM ARR = year-five retained accounts × annual account spend.Source: MoSPI ASUSE 2025 factsheet, Tables 2, 5 and 6 ↗. Scope: covered unincorporated trade and other-services establishments in India. Incorporated businesses, manufacturing and international customers are outside this model. The scenarios are not a validated count of software buyers.
TAM / Covered India customer-workload opportunity
The starting cohort is hired-worker locations in ASUSE’s unincorporated trade and other-services sectors. A locations-per-account assumption reduces double counting; a workload-fit assumption narrows it to recurring customer operations. This is a scoped entry-market TAM, not all Indian business software or a global AI workforce total.
SAM / A conditional current-product filter
Digital readiness is assumed within the workload-qualified cohort. Supported-workflow fit then narrows those ready accounts to the current channels, integrations and rollout requirements. Neither rate is an observed ASUSE cross-tab or measured WRKZY conversion rate; they need customer validation.
SOM / A capacity-based five-year illustration
This is a cohort roll-forward from zero, capped by acquisition and onboarding capacity and applying annual retention. It does not incorporate WRKZY’s actual pipeline, staffing, cash budget or current customer base. It should become an operating forecast only after those inputs are reconciled privately.
Global and role expansion / Separately earned
International customers, incorporated businesses and additional AI roles are outside this India appendix model. The global workforce vision needs country-level buyer definitions, deployment and language support, and role-level value evidence. We do not add adjacent software categories or labour budgets to produce an inflated combined market.
What could change the estimate materially?
The paying-workload share, locations per buying account, annual spending capacity and conditional product readiness drive TAM/SAM. Qualified acquisition, onboarding effort and retention drive SOM. The cases are sensitivity illustrations, not statistical confidence intervals. Validate these assumptions before external fundraising circulation.
05 / Economics & measurement
Measure the whole cost of completion.
Agent adoption and cheaper inference create an opening. They do not establish willingness to pay, workflow reliability or gross margin.
Compare the same unit of work
Competitors charge by seats, credits, conversations, outcomes or combinations of these. A credit can represent different actions across products. Compare a defined workflow and its full delivery cost instead of treating headline rates as interchangeable.
For future role or outcome pricing, define eligibility, completion, partial success, human intervention, retries and disputed results before assigning a price.
Account for the work around the model
Track inference, retrieval, orchestration, channel fees, storage, evaluations, retries and external services. Separately record onboarding, integration maintenance, human review and support effort. Apply a consistent accounting policy when reporting gross or contribution margin.
Report latency and completion quality alongside cost. A cheap attempt that creates rework is not a cheap outcome.
A practical measurement contract
Eligible attempts. Verified completions. Human interventions. Full delivery cost.
Use fixed cohort windows, representative tasks and explicit failure categories. Track incorrect external actions, escalations and recovery as first-class outcomes. Benchmark improvements against the customer’s prior workflow.
06 / Private diligence
The questions behind the next conversation.
These are diligence requirements, not claims that the following results or materials already exist.
Customers & retention
Paid and active businesses; cohort definitions; activation time; retained workflow usage; references and measured customer outcomes.
Revenue & economics
Reconciled revenue and collections; account expansion; gross and contribution margins; cost allocation; acquisition payback; burn and runway.
Team & execution
Founder backgrounds, relevant operating experience, current responsibilities, hiring needs and evidence of learning from deployments.
Technology & trust
Architecture review, production gates, tenant isolation, access and deletion controls, incident practices, testing and independent assurance where available.
Agent readiness
A named bounded responsibility; representative evaluations; allowed actions; approval policy; traceability; recovery; human intervention and completion cost.
Financing & ownership
Corporate structure, cap table, IP ownership and licences, financing history, intended use of capital, milestone budget and proposed terms.
WRKZY is operated by AIRBRILS PRIVATE LIMITED; the trademark is owned by AIRMESH PRIVATE LIMITED. See the official company details. The investment entity, IP rights and financing structure require confirmation during private diligence.
Four gaps that market numbers cannot close
Founder-market fit and current team responsibilities; verified paid demand and retained customer value; a funded milestone budget and runway; and the investment entity’s rights to the product IP. These require founder-approved records and evidence. The website and the model should not imply that missing information was independently verified.
07 / Primary source register
Follow the sources.
All links are to the original research, government release or vendor. Pages can change after our review; dates and limitations keep each finding in context.
International sizing inputs
SBA / Census · US small employer firms ↗
2025 national profile; underlying SUSB reference year 2022; reviewed 1 November 2026
The national profile reports 6,374,594 employer firms with fewer than 500 employees, using Census SUSB data.
Read with this limit: National all-industry firm count, not locations or qualified software buyers. Excludes nonemployers. The February 2026 FAQ uses an inconsistent employer total; this model uses the size-specific national profile instead.
International sizing inputs
ISED · Canada employer enterprises ↗
2025 report, published January 2026; reference December 2024; reviewed 1 November 2026
Tables 1–2 report 1,079,188 small and 16,953 medium employer enterprises: 1,096,141 with 1–499 paid employees.
Read with this limit: Enterprise-level employer counts, excluding unclassified enterprises. Not harmonized with every country's statistical coverage or a count of qualified WRKZY buyers.
International sizing inputs
ABS · Australia actively trading employer businesses ↗
18 August 2026; population at 30 June 2026; reviewed 1 November 2026
Reports 996,203 employing businesses; subtracting 5,366 with 200+ employees gives 990,837 with 1–199 employees.
Read with this limit: Actively trading statistical businesses, not every registered ABN. Employment threshold and coverage differ from US, UK and Canada. Snapshot the reference period when refreshing this latest-release link.
Global market context
Gartner · Worldwide customer experience and relationship management, 2024 ↗
11 June 2025
Gartner estimates the worldwide CRM software market at $128 billion in 2024.
Read with this limit: Historical estimate for a broad category, based on the public research abstract. It is not WRKZY TAM, a serviceable budget or current-year market revenue.
Global market context
Gartner · Agentic AI spending in software markets ↗
17 February 2026
Gartner forecasts agentic AI software spending of $985 billion in 2030.
Read with this limit: A broad forward-looking forecast, not current revenue or an AI-employee-only vendor market. It overlaps other software categories and is not WRKZY’s addressable revenue.
Global market context
Gartner · Enterprise application spending exposed to agentic arbitrage ↗
1 July 2026
Gartner identifies up to $234 billion of enterprise application spending exposed to agentic arbitrage through 2030.
Read with this limit: Potential disruption and redistribution of spending, not incremental market growth. Do not add it to CRM or agentic-software estimates or describe it as WRKZY SOM.
Regional business context
SBA Office of Advocacy · Small business FAQ, 2026 ↗
February 2026; underlying reference years vary
Reports approximately 36.2 million US small businesses, including about 6.4 million employer and 29.8 million nonemployer firms.
Read with this limit: The publication combines reference years and labels 6,395,635 employers as small, although that includes 21,041 large firms. The new sizing model instead uses 6,374,594 small employers from the detailed SBA/Census national profile. The FAQ total is not a qualified buyer count.
Regional business context
Eurostat · Final 2024 business-economy statistics ↗
30 September 2026; reference year 2024
Reports approximately 34.0 million enterprises in the EU business economy in 2024, superseding the preliminary estimate.
Read with this limit: Specified business-economy sectors, not the whole economy. Different scope from US small businesses, UK private businesses and Indian establishments; do not sum these as a harmonized census.
Regional business context
UK government · Business population estimates, 2025 ↗
2 October 2025; population at start of 2025
Estimates 5.7 million private-sector businesses, including about 1.4 million employers, at the start of 2025.
Read with this limit: Includes nonemploying and unregistered businesses. Official statistics in development; broader than VAT/PAYE-only counts and not a count of paying software prospects.
Market sizing inputs
MoSPI · ASUSE 2025 factsheet ↗
Released March 2026; reference period January–December 2025
Tables 2, 5 and 6 provide trade and other-services establishment counts and their hired-worker shares, used to derive the model’s starting location population.
Read with this limit: Survey estimates and rounded shares; locations are not unique buying accounts. Coverage is unincorporated activity, not all Indian businesses. Workload fit and willingness to pay are not measured here.
Why now
MoSPI / PIB · ASUSE 2025 detailed results ↗
6 May 2026
Business internet use rose from 26.68% in ASUSE 2023–24 to 39.37% in 2025 across covered establishments, indicating increasing digital participation.
Read with this limit: This is the full survey population, not the model’s hired-worker cohort. It is context, not a conditional serviceability rate or evidence of paid software adoption.
Market context
World Bank · Small and Medium Enterprises Finance ↗
Undated; reviewed 30 October 2026
SMEs account for approximately 90% of businesses and more than half of employment worldwide.
Read with this limit: Business prevalence is context. It does not establish the number of software buyers or WRKZY’s addressable revenue.
Market context
Government of India · Empowering the Grassroots Economy ↗
14 May 2026
The release reports more than 7.9 crore enterprises registered through Udyam and Udyam Assist by March 2026, including informal micro enterprises.
Read with this limit: A dated registration count is neither a count of active companies nor an estimate of paying prospects. Do not multiply it by a subscription price to claim TAM.
Why now
Microsoft · 2026 Work Trend Index ↗
5 May 2026
Microsoft reports 15× growth in active agents in its ecosystem between March 2025 and March 2026.
Read with this limit: Vendor ecosystem telemetry, not market-wide adoption or ROI. Activity can mean one user-initiated usage day or one autonomous run in a rolling 28-day period.
Why now
Stanford HAI · AI Index 2025 in ten charts ↗
7 April 2025
At a fixed GPT-3.5-equivalent benchmark level, reported inference prices fell from $20 to $0.07 per million tokens between November 2022 and October 2024.
Read with this limit: A historical, fixed-performance comparison. Complete business workflows also incur integration, review, retry, channel and support costs.
Work and reliability
NBER · Generative AI at Work ↗
2023 working paper, revised November 2023
A field study of customer support found improvements in issues resolved per hour with AI assistance, with larger gains among less experienced workers.
Read with this limit: Evidence from a particular workplace and assistive deployment. It is not a WRKZY result or evidence that an autonomous employee can replace a role.
Work and reliability
Anthropic · Building effective agents ↗
19 December 2024
Distinguishes predefined workflows from model-directed agents and describes composable patterns, tool design and evaluation considerations.
Read with this limit: Engineering guidance from a model provider, not independent validation of WRKZY’s agent readiness or performance.
Work and reliability
METR · Task-completion time horizons ↗
Last updated 8 September 2026; reviewed 30 October 2026
Evaluates agent capability using task difficulty and a specified success probability, including 50% and 80% success horizons.
Read with this limit: The page is no longer actively updated. The suite primarily covers technical tasks; its horizons do not establish reliability across business roles.
Competitive landscape
Salesforce · Agentforce pricing ↗
Vendor page; reviewed 30 October 2026
Offers customer-facing and employee-facing agents with consumption and licensing options, including Flex Credits and conversation pricing.
Read with this limit: An incumbent with CRM context and distribution. Action credits, conversations and user licences are different units; headline prices are not comparable total costs.
Competitive landscape
Microsoft · Copilot Studio pricing ↗
Vendor page; reviewed 30 October 2026
A platform for agents, connectors and flows, with Copilot Credits and pay-as-you-go options.
Read with this limit: A powerful build alternative. WRKZY must demonstrate that deploying and operating a packaged business workflow takes less customer effort.
Competitive landscape
Intercom · Fin and platform pricing ↗
Vendor page; reviewed 30 October 2026
Fin uses outcome-based billing and can work with an existing helpdesk. It is a direct specialist benchmark for customer support work.
Read with this limit: Use the vendor’s definition of an outcome and applicable commitments. A billed outcome is not automatically equivalent to another vendor’s resolved conversation.
Competitive landscape
HubSpot · AI credits ↗
Vendor page; reviewed 30 October 2026
Breeze Customer Agent uses credits for conversations resolved without human help. The platform offers usage limits and spending controls.
Read with this limit: Existing CRM context and AI spending controls are competitive requirements, not unique WRKZY inventions.
Competitive landscape
Zoho · Zia Agent Studio reference ↗
Vendor documentation; reviewed 30 October 2026
Supports configuring agents with instructions, knowledge, tools and guardrails inside Zoho’s business software ecosystem.
Read with this limit: Relevant competition for integrated business software. Permissions and a broad feature set alone cannot establish a defensible advantage.
Competitive landscape
Sintra · AI employee offering ↗
Vendor page; reviewed 30 October 2026
Markets twelve role-based AI employees with shared business context, integrations and usage credits.
Read with this limit: Direct category competition. WRKZY should not claim to have invented the AI employee category; promotional offers are unsuitable long-term price anchors.
Competitive landscape
Lindy · AI assistant plans ↗
Vendor page; reviewed 30 October 2026
Offers an AI teammate model with shared context, routines, integrations, usage credits and approval controls.
Read with this limit: Human approval is already present in competing products. Differentiation must include deployment effort, workflow depth and observed results.
Competitive landscape
Relevance AI · AI workforce offering ↗
Vendor page; reviewed 30 October 2026
Explicitly markets AI workforces, agents and tools, with enterprise evaluations, analytics, permissions and audit capabilities.
Read with this limit: Direct workforce orchestration competition. A platform architecture or audit log is not sufficient evidence of uniqueness.
Competitive landscape
Meta · Business Agent announcement ↗
3 June 2026
Introduced Meta Business Agent and the Business Agent Platform for customer interactions, including WhatsApp, Messenger and Instagram.
Read with this limit: Meta is both a channel dependency and a competitor. WRKZY’s investment thesis must extend beyond a WhatsApp reply assistant.
Back to the investment thesis ←
Research reviewed 1 November 2026. External findings are not WRKZY performance claims.