Credit Consumption Analysis

February 24, 2026 · 21 external organizations · Production data

Executive Summary

The Target

B1: $500/mo
Starting Orgs
10 new/month · product-driven
B2: $5–10K
Graduated Orgs
25% grad rate · 6–9 months to graduate
B3: $30–50K
Enterprise Deals
2–4 deals · 6–12 mo sales cycle
$255K
Base Case M12
$500K stretch · depends on enterprise

Where Those Credits Come From

Credits flow through four product areas. Each area is shipped and working. The question is where to focus to drive expansion from $500/month to $5–10K/month per org.

Product AreaHow It Generates CreditsCurrent EvidenceThe Opportunity
SQL Execution Query seconds → query credits. Pipelines run daily. Marinade: 54K credits in 35 days. CMC/Stockholm: high volume during trials. Proven credit driver. Orgs building data pipelines consume 1,800–6,000 credits/day.
Chat AI tokens + triggered queries. Sessions are one-off. 6 orgs active. Laminated, Avalanche, Monad, Aleo exploring. 79% AI, 21% query. Entry point for new orgs. Low credits today ($50–$200/mo) because sessions don’t recur.
Agentic CLI Agent reasoning + iterative queries. 3–60x more credits than UI. Internal power users: 300–500 credits/day. Zero external adoption. Highest credits per user. An agent session triggers 3–10 queries that iterate and self-correct.
Automations
incl. alerts, briefings, scheduled queries
Recurring credits without human action. Runs on schedule. A few orgs have automations — all created by Flipside team. Best NRR lever. Converts one-time exploration into baseline recurring consumption.
The core insight: people come here for the data. The path to $500K/month is getting orgs from $500 to $5–10K by using agents to build their data workflows.

Across 21 external orgs, 86% of credits come from programmatic access — not from clicking around in the UI. Users who access the platform programmatically generate 3–60x more credits. The highest-spend orgs are building SQL pipelines by hand. If they use agents to build those pipelines instead, it drives both AI credits (agent reasoning) and query credits (the pipeline it produces).

The blocker isn’t features — everything is built. The blocker is packaging and direction. There’s a lot you can do on the product, but it’s not clear to users what the one thing is they should do next. The Agentic CLI is our highest credit-per-user surface with zero external users — not because it doesn’t work, but because nobody knows how to set it up. Automations work but nobody knows what they’d create one for.

Product focus for the next 2 months:
1. Package the Agentic CLI for external orgs — make “use agents to build pipelines” the default path
2. Make Automations discoverable — clear path from “the agent built this query” to “this runs every day”
3. Show existing orgs how agents can build and iterate on their workflows faster than writing SQL by hand

What Good Looks Like — April 2026

Two months from now. Here’s what we should be able to point to.

MetricCurrentApril ’26 TargetWhy This Matters for $500K/Mo
Orgs at $500+/mo Active orgs across tiers 5 First 5 prove the $500/mo floor works. These are the top of the funnel.
External Agentic CLI users 0 external 3+ Highest credits/user surface. Must prove it works outside internal team.
Orgs using agents to build queries 0 external 3+ Core thesis: agents building pipelines drives AI + query credits together.
Self-serve Automations Built, team-created only Self-serve Recurring credits = NRR. Orgs must create their own automations.
Orgs using 2+ product areas Orgs siloed in one area 3+ Cross-surface usage (SQL + Chat, SQL + CLI) = sticky, higher ARPU.
The 2-month gate: If we can’t get 5 orgs above $500/month, 3 external Agentic CLI users, and at least 3 orgs using agents to build their queries by April, the funnel model needs to be revisited. These are the leading indicators that the thesis works.

1. What Drives Credit Usage

Thesis: programmatic usage → credit consumption → revenue. Credits are consumed in two categories: AI tokens (chat completions, agent reasoning) and query execution (SQL seconds). Across 21 external orgs, 86% of all credits come from programmatic access (API, CLI, automated pipelines) — not from users clicking around in the UI. If we want to grow revenue, we need to invest in product that drives programmatic usage.

Programmatic vs. Manual: How We Classified

SignalClassificationWhat It Means
session_id = NULLProgrammatic — Direct API Query or AI call made outside any chat session. Pure API/SDK usage.
meta.source = “cli” | “api”Programmatic — CLI/MCP Chat session initiated from CLI, Cursor, or MCP tool. Has a session, but not human-in-the-UI.
meta.source = “web” | “agents-page”Manual — Web UI User in the browser, chatting with agents or running queries manually.
Burst queries (<2s gaps)Programmatic — Pipeline Rapid-fire queries at machine speed. Marinade: 90% burst rate. CMC: 98%.
Validation: Jim Myers described himself as “largely programmatic.” The initial session-presence heuristic classified him as 50/50 — wrong, because CLI-initiated sessions have a session_id. After incorporating meta.source, his classification moved to 97.7% programmatic (233 of 298 sessions are CLI-originated, plus all sessionless query events). This confirmed the heuristic works.

Platform-Wide Findings

86%
External Credits Are Programmatic
API calls, CLI, automated pipelines
98%
Query Credits Are Programmatic
Nearly all SQL execution is API-driven
91%
AI Credits Are Session-Based
Most AI usage happens inside chat sessions

Three Consumption Patterns

PatternOrgs% of CreditsProgrammatic %What They Do
API Pipeline380%~100% Automated SQL pipelines via API. Machine-speed burst queries. Zero AI. High volume, low stickiness.
AI Chat613%~15% Exploring data through agents in the UI. Human-paced. High AI %, low query volume. Undermonetized.
Mixed311%~60% Both AI exploration and query execution. Closest to the target pattern. Highest stickiness potential.
The thesis confirmed: programmatic access drives dramatically higher credit consumption. The highest-volume orgs (Marinade, plus CMC and Stockholm during trials) burn more credits in days than AI chat explorers burn in weeks. But volume alone isn’t health — pipeline orgs have zero AI usage, making them commodity consumers (any SQL warehouse will do). The highest-value pattern is mixed: both AI and query, both programmatic and manual.

The Revenue Multiplier by Access Pattern

Credit consumption follows a clear hierarchy. Each step up the programmatic ladder multiplies revenue per user:

Access PatternAvg Credits/DayRev MultiplierWhy
Manual UI only50–1001x Browsing, occasional chat. Human-speed ceiling caps consumption.
Manual chat + API queries200–3503–5x Chat for exploration, then direct API for extraction. Eric Stone’s pattern.
Agent-driven (CLI/Cursor/MCP)300–5005–10x One agent session triggers 3–10 downstream queries that iterate and self-correct.
Automated pipeline (API)1,800–6,00020–60x Machines run 24/7 at burst speed. No human bottleneck.
This is the investment thesis for the product roadmap. Every user who moves from “manual UI” to “agent-driven” generates 5–10x more credits. Every org that adds automated pipelines generates 20–60x. Features that enable programmatic access — better APIs, CLI tools, MCP integrations, scheduled queries, agent-driven workflows — directly multiply revenue per user. Features that only improve the manual UI experience have a 1x ceiling.

External Org Consumption (21 orgs)

Organization Credits AI % Qry % Cred/Day Est Rev/Mo Pattern
Marinade Finance64,025 3%97% 1,829$1,646 Pipeline
CMC Capital SRL20,817 0%100% 5,204$4,684 Pipeline
Stockholm SSE18,066 0%100% 6,022$5,420 Pipeline
CoW Swap6,922 36%64% 865$779 Mixed
Laminated Labs6,24291%9%223$201AI Chat
Near Foundation5,87931%69%210$189Mixed
Avalanche3,59181%19%257$231AI Chat
Monad2,68471%29%112$101AI Chat
Aleo2,63186%14%329$296AI Chat
Somnia1,30274%26%93$84AI Chat
Karatage1,22834%66%123$111Mixed
Flow82416%84%59$53Light
Attestant62888%12%45$41AI Chat
+ 8 orgs under 500 credits Circle, Solana, Ink, Polygon, Sui, Morpho, Kraken, Movement
Volume and health are not the same thing. The highest-volume orgs (Marinade sustained, CMC and Stockholm during trials) consumed 80% of all credits but have 0–3% AI usage. They’re running raw SQL pipelines — high revenue today, zero moat tomorrow. Meanwhile, orgs with balanced AI:Query ratios (CoW Swap at 36:64, Karatage at 34:66, Near at 31:69) are closer to the target pattern but consume 10–60x less. The gap between volume and health is the core product challenge.

2. The Model — What a Healthy Org Looks Like

The ideal customer uses AI agents to explore data and runs queries to extract it. Query-only users are high-volume but commoditized (any SQL warehouse will do). AI-only users are low-volume and substitutable (ChatGPT + a SQL tool). When both layers are engaged, AI drives queries and queries feed AI — a flywheel that’s hard to leave.

Validated Per-User Economics

Two internal power users with completely different workflows independently converged on similar consumption patterns. This gives us the per-user building block — the question is how many users per org we can drive.

Jim Myers — “Agent-First”Eric Stone — “Explorer”
AI : Query ratio52 : 4844 : 56
Credits / day~500326
Monthly projection~15,000 ($450)~9,800 ($294)
WorkflowCLI agents trigger cascading queriesChat exploration, then direct API queries
Channel78% CLI/Cursor, 49% no-session queries11 chat sessions, 105 direct API queries
InputValueBasis
Credits per user per day300–500Eric = 326, Jim = 500. Conservative: 350.
AI : Query ratio40 : 60Eric = 44:56, Jim = 52:48. Midpoint.
Active days per month22Weekdays
Revenue per user per month$198–$3306,600–11,000 credits × $0.03

What “Healthy” Means at the Org Level

We’re not accepting any customer below $500/month. This is a business product. $500/month = ~16,700 credits/month = 2–3 active users at 300 credits/day. That’s the floor, not the target. The target is $1,155–$2,970/month.
Org StageUsersCred/U/DayCredits/MoRev/MoRev/Year
Entry — small team2–335016,700+ $500$6,000Minimum. 2–3 users, team workflows.
Healthy — data team535038,500 $1,155$13,860Target. Pipelines + agents + analysts.
Strong — power team1045099,000 $2,970$35,640Ideal. Multiple workflows, org-wide.
Enterprise — scaled20500220,000 $6,600$79,200Enterprise. Deep integration across teams.
With a $500/month floor, every customer is worth $6K+ ARR. The goal is graduating orgs from Bucket 1 ($500/month) to Bucket 2 ($5–10K/month) through expansion.

Today’s reality: 21 orgs are active across product areas at various spend levels. The immediate challenge: get orgs above $500/month by driving agent adoption and cross-surface usage.

Why 40:60 AI:Query?

Both power users independently converged on ~40:60. Jim Myers (agent-first, CLI/Cursor) landed at 52:48. Eric Stone (explorer, web chat + API) landed at 44:56. The midpoint is 40:60. This ratio reflects the natural economics: queries are heavier per event (~7 credits vs ~1.5 for AI), and a single agent session triggers 3–10 downstream queries. The query tail is always longer.

The current platform average is 29:71 — skewed by pipeline orgs running raw SQL with zero AI. That’s commodity behavior. Pushing toward 40% AI means users are in the agent layer, which is the differentiated product.

Margins are identical (80%) on AI and query credits. The 40:60 target is about stickiness and moat, not profitability per unit. More AI usage compounds revenue — one agent session at ~$0.75 of AI credits triggers ~$1.20 of query credits.

3. Three Buckets — The Math

At any given time, revenue comes from three buckets. The question is: at what rate do orgs graduate from Bucket 1 to Bucket 2, how long does it take, and how many enterprise deals can we close? That math determines what’s realistic.
Bucket 1: StartingBucket 2: GraduatedBucket 3: Enterprise
Rev/mo$500–$2K$5–10K$30–50K+
WhoSmall team, 2–3 users, running queries.Data team, 8–15 users, agents + automations.Fidelity-scale. Custom feeds, SLA, dedicated support.
How they arriveProduct-driven. Sign up at $500/mo floor.Graduate from Bucket 1 by adding users & workflows.Sales-driven. 6–12 month enterprise sales cycle.
What drives revenueVolume. Many orgs at low ARPU.Expansion. Fewer orgs, high ARPU.Deal size. A few orgs, very high ARPU.

Graduation: Bucket 1 → Bucket 2

An org enters at $500/month. The question is how many graduate to $5–10K and how long it takes.

MilestoneTime from EntryUsersRev/MoWhat Triggers It
EntryMonth 12–3$500 Small team sets up queries + AI sessions.
First expansionMonth 2–33–5$1–2K More analysts join. Pipelines forming. Someone tries the CLI.
Graduated — Bucket 2Month 6–98–10$5–7K Agents building pipelines. Automations running. Multiple daily workflows.
Scaled Bucket 2Month 9–1210–15$8–10K Full org adoption. Embedded in daily workflows. 10+ active users.
Graduation takes 6–9 months. An org that enters in March might reach $5K/month by September–November. This means: only orgs entering in the first 6 months can graduate within the 12-month window. Anything entering after month 6 is still in Bucket 1 at year-end.

Graduation Rate — How Many Make It?

Not every org graduates. Of every 10 orgs that enter at $500/month:

Outcome% of EntrantsOrgs (of 10)What Happens
Churn~35%3–4 Don’t find value. Gone in 2–3 months.
Stay in Bucket 1~40%4 Using the product at $500–$2K. Not expanding. 1–3 users.
Graduate to Bucket 2~25%2–3 Reach $5–10K/month. 8+ users. Agents + automations.
1 in 4 entering orgs graduates to Bucket 2. This is the critical rate. If it’s 1 in 3, the model overperforms. If it’s 1 in 6, it underperforms significantly.

Today’s evidence: Marinade is at $1,646/month (3.3x entry) with zero AI — already mid-funnel on query pipelines alone. If they adopt agents, $5K+ is realistic. But we have zero confirmed graduations yet. The 25% rate is an assumption that needs to be validated in the first 6 months.

Enterprise (Bucket 3) — What’s Realistic?

AssumptionValueWhy
Sales cycle6–12 monthsEnterprise procurement. Legal, security, POC, then contract.
Pipeline startNow (Feb ’26)First close can’t happen before August at earliest.
Deal size$30–50K/monthCustom data feeds, dedicated agents, SLA. Could be higher for Fidelity-scale.
Deals closed in 12 months2–41 by month 6–8, 1–2 by month 9–10, 0–1 by month 12.
Enterprise is lumpy. One Fidelity-scale deal at $80K/month is worth more than 10 Bucket 1 orgs. But you can’t plan around it — it either closes or it doesn’t. Base case: 3 deals, ~$120K/month. Upside: a single large deal doubles that.

The Math — Month 12 Snapshot

Assuming 10 new orgs/month average, 25% graduation rate, 6–9 month graduation time, 3 enterprise deals.

BucketOrgsAvg ARPURev/MoHow They Got Here
Bucket 1: Starting
$500–$2K/mo
~55$1,100$60K Mix of recent entrants ($500) and growing orgs ($1–2K) that haven’t graduated yet.
Bucket 2: Graduated
$5–10K/mo
~10$7,500$75K ~25% of the 60 orgs entering months 1–6, surviving and expanding over 6–9 months.
Bucket 3: Enterprise
$30–50K/mo
3$40K$120K Sales-driven. First deal by month 6–8, 2 more by month 12.
Total at Month 12~68 $255K
ARR $3.1MGP: $2.4M at 80% margin

Sensitivity — What Moves the Number

Three levers. Graduation rate matters most for the funnel; enterprise deals move the total fastest.

ScenarioNew Orgs/MoGrad RateEnterpriseBucket 1Bucket 2Bucket 3Total
Conservative820%2 deals $50K$48K$80K $178K
Base case1025%3 deals $60K$75K$120K $255K
Strong execution1230%4 deals $70K$108K$160K $338K
Enterprise upside1025%3 + 1 Fidelity $60K$75K$200K $335K
Stretch1533%5 deals $85K$150K$200K $435K
Where $500K/month sits: You need the stretch scenario — 15 new orgs/month, 33% graduation rate, and 5 enterprise deals. Or base case funnel + a Fidelity-scale deal ($80–100K/month). Either path requires both the funnel and enterprise to perform.

Base case is ~$255K/month ($3.1M ARR). That’s 10 new orgs/month, 25% graduating to $5–10K, and 3 enterprise deals at $40K. This doesn’t require a hockey stick — it requires steady acquisition and proven expansion.

The biggest unknown is graduation rate. If it’s 20% instead of 25%, you lose $27K/month from Bucket 2. If it’s 33%, you gain $75K. We’ll know by month 6 which rate is real.

12-Month Roadmap — All Three Buckets

B1 OrgsB1 RevB2 OrgsB2 RevB3 RevTotalWhat Has to Be True
Today
Feb ’26
early0 21 active orgs. Zero graduations. Zero enterprise pipeline.
Month 2
Apr ’26
5$4K0$0$0$4K Gate. 5 orgs at $500+. CLI packaged. Enterprise pipeline started.
Month 4
Jun ’26
15$14K0$0$0$14K Early orgs expanding to $1K+. Still too early for graduations. Enterprise conversations active.
Month 6
Aug ’26
25$30K1–2$8K$40K$78K Checkpoint. First graduations happening. First enterprise deal closing. Graduation rate becoming visible.
Month 9
Nov ’26
40$48K5–6$38K$80K$166K Graduation rate validated. 2–3 enterprise deals live. Earliest cohorts at $7–10K.
Month 12
Feb ’27
~55$60K~10$75K$120K$255K Base case. 120 entered, ~65 active in B1, ~10 graduated to B2, 3 enterprise deals.
The month-6 gate: ~$78K/month (25 Bucket 1 orgs + first graduation + first enterprise deal). More importantly: we’ll know the graduation rate by then. If month-1 cohorts are hitting $3–5K by August, the model works. If they’re still at $500, it doesn’t.

Why this model is more honest: It doesn’t require 40 new orgs/month. It requires 10 orgs/month entering at $500, and 1 in 4 of them growing to $5–10K over 6–9 months. Revenue grows because orgs climb the ladder, not because you’re flooding the top. Enterprise deals are additive — each one is a step function, and a single Fidelity-scale deal could add $80–100K in one move.

4. The Playbook — What to Do

Everything is built. The work is packaging, onboarding, and directing users toward the actions that drive credits. The single question for every initiative: does this convert data usage into AI usage?

Next 2 Months (by April ’26)

1. Package the Agentic CLI for external orgs
P0

Agentic CLI
The problem: The CLI works — Jim generates 500 credits/day through Cursor. But zero external users have adopted it because it’s not clear what it is, how to set it up, or what you’d use it for. There’s no onboarding, no setup guide, no “here’s what you get.”
Do: Setup guide + quick-start tutorial. Onboarding flow that gets a user from zero to first agent invocation in 10 minutes. Clear packaging: “Access Flipside agents from Cursor, Claude Code, or any MCP-compatible tool.”
Measure: 3+ external Agentic CLI users by April.

2. Show pipeline orgs how agents build pipelines faster
P0

SQL Execution Chat Agentic CLI
The insight: People come here for the data. Marinade is writing SQL by hand and spending $1,646/month. They’re getting value — but doing the hard work themselves. The pitch: let agents build your pipelines. An agent writes the SQL, iterates on it, deploys it as an automation. Faster for the user, and it drives both AI and query credits.
Do: Hands-on sessions with pipeline orgs. Demo the agent building a pipeline they already have — “you wrote this in SQL, the agent does it in 30 seconds.” Make “build with an agent” the suggested path inside the query interface.
Measure: At least 1 pipeline org with AI% >10% by April.

3. Make Automations discoverable and self-serve
P0

Automations
The problem: Automations are built and working — briefings, alerts, scheduled queries — but users don’t know what they’d create one for or how. We provide examples but there’s no clear path from “I just ran a query” to “this runs every day and sends me a briefing.” Every automation deployed was created by our team.
Do: Clear CTAs in the product after a query or chat session: “Want this as a daily briefing?” “Set up an alert when this crosses X.” Templates for common use cases. Self-serve creation flow that doesn’t require our team.
Measure: 3+ orgs with at least one self-created Automation by April.

4. Qualify 2 new orgs at $500+/month
P0

Why now: 5 orgs at $500+/month is the first proof point that the floor works as a business. These need to be teams (2–3+ users) using multiple product areas, not individuals in one.
Do: Identify from existing pipeline (10 sub-$500 orgs) or net new. Onboard them across SQL + Chat + CLI. Direct them — don’t give them five options, give them one path.
Measure: 5 orgs above $500/month by April.

Months 2–6 (by August ’26)

5. Onboard teams, not individuals
ARPU × n

Why: Most orgs have 1–2 users. The floor is $500/mo (2–3 users). Healthy is $1,155+ (5 users). Every additional user is linear ARPU growth. New users should land in the product area that fits their role: analysts into SQL + Chat, engineers into the Agentic CLI.
Do: Team invite flow, role-based onboarding, shared dashboards/queries, org-level billing visibility. The first experience for user #2 on an org should be as directed as user #1.
Measure: Avg users per paying org > 4 by August.

6. Scale recurring consumption
NRR

Automations
Why: Credits consumed without human action are the purest revenue — and the product already supports it. Once self-serve Automations work (item 3), the goal is adoption: every paying org should have at least one recurring automation. This is what makes NRR > 120%.
Do: Automation templates, suggested automations based on usage, “set it and forget it” flows.
Measure: 30%+ of credits from Automations by August.

The bigger picture: if converting data usage → AI usage works for blockchain orgs, the same pattern applies beyond blockchain. Stockholm SSE accessing economic data through agents is the same motion as a hedge fund or a protocol team doing it. The product areas are domain-agnostic — the wedge is the data.

5. Key Risks — What We Don’t Know

RiskWhy It MattersHow to Validate
Pipeline orgs churn Highest-volume orgs run raw SQL pipelines with zero switching cost — any SQL warehouse works. Without AI adoption, there’s no moat. Get pipeline orgs using agents to build their workflows. AI usage = switching cost. Track AI % per pipeline org monthly.
Per-user economics don’t replicate 300–500 credits/day is validated on 2 internal power users. External users may consume less, use AI differently, or not adopt agents at all. Instrument per-user daily credit consumption for external orgs. Need 10+ external data points in 3 months.
Orgs don’t expand The funnel depends on orgs growing from $500 to $5–10K. If most orgs stay at $500–$1K, the funnel produces ~$50K/month instead of $164K — and the entire model breaks. Track users-per-org and ARPU monthly. If no expansion signal by month 6, reassess the target.
Acquisition engine doesn’t exist yet The funnel needs 120 orgs over 12 months (~10/month avg). More manageable than 245, but still requires a repeatable channel that doesn’t exist yet. Identify top-of-funnel channel by month 4. Test: content, partnerships, PLG viral loops, outbound. Need 5+ qualified orgs/month by month 6.

6. Scoreboard

MetricTodayApr ’26 (2 mo)Aug ’26 (6 mo)
Funnel orgs at $500+/mo early 5 30
Orgs building with agents 0 external 3+ 50%+
External Agentic CLI users 0 3+ 15+
AI % across paying orgs ~0% >10% 30%
Orgs with self-created Automations 0 (team-created) 3+ 10+
Orgs using 2+ product areas siloed 3+ majority
Orgs > $1K/month early 3+ 10+
Highlighted rows are the 2-month gates. If we miss those three — 5 orgs at $500+/mo, 3+ orgs building with agents, 3+ external CLI users — the funnel model needs to be revisited.

Appendix: Methodology & Limitations

Data source: Production OrgUsageEvent table via internal-cli -e prod admin usage-events. 21 external organizations, Flipside internal org (538,644 credits) excluded.

AI:Query classification: Events have category field: ai_usage (chat completions) or query_execution (SQL execution). Credits are calculated by the usage event service based on token counts and query seconds respectively.

Programmatic vs manual heuristic:
session_id = NULL → Direct API (no chat session involved)
ChatSession.meta.source = "cli" | "api" → CLI/MCP-driven
ChatSession.meta.source = "web" | "agents-page" → Web UI

Prototypical org derivation: Based on two validated users: Jim Myers (agent-first, 500 credits/day, 52:48 AI:Query, 298 sessions over observation period, 78% CLI-originated) and Eric Stone (explorer, 326 credits/day, 44:56 AI:Query, 11 sessions + 105 direct API queries, 5.8% burst rate). Conservative blend: 300 credits/user/day, 2.5 users/org, 22 active days/month.

Projection assumptions: New orgs ramp to prototypical over 2–3 months (25%/60%/100% for base case). Existing org growth: 10–20% MoM as users adopt AI + query workflows, capped at 1.5x prototypical per org. Churn: 1–5% monthly depending on scenario.

Limitations:
• Session source metadata only available for internal users (Jim, Eric). External org session sources inferred from session_id presence/absence.
• CMC Capital and Stockholm SSE consumed credits in 3–4 day bursts during trials; they are not treated as sustained paying orgs.
• Eric’s observation window is 5 days — his monthly projection (9,800) is directional, not confirmed.
• The 2.5 users/org assumption is not validated against external org data — some orgs may have 1 user, others 10.