# Who should we sell to next?

Independent portfolio case study. The software company and product are hypothetical. No customer interviews, sales campaigns, or product tests have been run.

## Verified market data
Source: U.S. Census Bureau, 2023 County Business Patterns, national file:
https://www2.census.gov/programs-surveys/cbp/datasets/2023/cbp23us.zip
Retrieved October 14, 2026. Fixed historical vintage, not a claim of current market size.

Filter naics to 541810, 541611, 541512. Keep lfo = "-" (all legal forms), so legal-form subcategories are not double counted. Add n10_19 and n20_49 for locations with 10 through 49 employees. These are establishment counts, not unique companies or buying accounts. A multi-location company can appear multiple times. Segment labels are proxies: advertising agencies omit other marketing specialties; management consulting omits other consulting categories; systems design is a subset of IT services. The employee range applies to locations, not parent-company size. CBP excludes businesses without paid employees. No firm-level qualification, software spend, growth, or willingness to buy is inferred.

## Public product evidence
Asana: https://asana.com/agencies
Agency workflow and competitor context: client work, approvals, resource planning, reporting.
Teamwork: https://www.teamwork.com/teams/project-management-software-consultant/
Consulting workflow and competitor context: client projects, resource scheduling, time, budgets, billing.
Atlassian: https://www.atlassian.com/software/jira/service-management
IT workflow and competitor context: support requests, incidents, change management.
Reviewed October 14, 2026. Vendor pages establish advertised capabilities, not prevalence of pain, competitor market share, buying intent, or comparative superiority. Requirements and scores below are analyst hypotheses informed by these examples, not empirical survey results.

## Hypothetical product and fit rubric
Assume an existing product supports task ownership/timelines, client approvals, workload planning, and status dashboards. No billing, time tracking, ticket queues, incident management, or service-level rules.
Agency needs: timelines, approvals, workload, client visibility, status reporting. Covered 5/5.
Consulting needs: timelines, workload, client visibility, billable time, invoicing. Covered 3/5.
IT needs: project tasks, client visibility, ticket queues, incident response, service-level rules. Covered 2/5.
Equal weight per need is a simplification. Score is coverage of the proposed five needs, not proven product-market fit.

## Decision model
Default importance: market proxy 20%, product coverage 60%, monthly account value 20%.
Market score = location count / maximum location count * 100.
Coverage score = covered needs / 5 * 100.
Value score = assumed monthly account value / maximum assumed monthly account value * 100.
Overall score = sum(each score * its importance) / sum(importance).
Monthly account values: agencies $300, consulting $600, IT $900. Illustrative account-level subscription revenue, not surveyed prices or market benchmarks. Values exclude acquisition and service costs, churn, discounts and expansion. All weights, coverage assumptions, and prices can be changed in the page. All-zero weights return no recommendation. All-zero prices give zero value scores. Ties within 0.05 points are reported as ties. Scores compare only these three choices; they are not success probabilities.

Default scores: agencies 72.89, consulting 69.33, IT 62.03. Agency fit falling from 5 to 4 reduces its score by 12 points, so consulting leads. This is a fragile test priority, not a validated expansion decision.

## Proposed test
Start with U.S. advertising agencies, 10 through 49 employees at a location, multiple concurrent clients, and approval delays. Verify actual firm size and independent buying authority first. Proposed buyer: head of operations; trigger: recurring missed approvals. These are hypotheses.
Interview 10 agency operators and 5 consulting operators, then run a 30-day agency pilot. Target 100 qualified, unique buying accounts. Illustrative funnel: 10% start a demo, 40% of demo accounts start a pilot, 50% of pilot accounts become paying accounts. Expected values = 10 demos, 4 pilots, 2 customers; at $300 per account per month, $600 new monthly recurring revenue and $7,200 annualized run rate, not cash collected in year one. Funnel assumptions are not results.

Before the test, define activated = invited a client, added a real deliverable, completed an approval within 7 days. Directional continuation gate: at least 2 paying accounts and at least half of pilot accounts activated; assess after the full paid-conversion window. Require demonstrably shorter approval cycles against each participant's baseline, plus onboarding time and acquisition cost review before scaling. Small samples cannot establish segment superiority. Keep a comparable consulting test as the next step if agency switching resistance or low willingness to pay invalidates the hypothesis.