# What Portland's small-business dashboard tells us

October 20, 2026. Data provided by **Dan Ruan and Mitch Daugherty, Prosper Portland's Office of Small Business**. Analysis by Portland Civic Lab.

Source: [Portland Office of Small Business dashboard](https://www.portlandciviclab.org/data/small-business/osb-dashboard-2026-10-20.pdf), pages 1–3. [Download the transcribed values](https://www.portlandciviclab.org/data/small-business/osb-dashboard-2026-10-20.csv).

## Assistance is visible; its lasting benefits are not yet measured here

The dashboard describes businesses in contact with the Office of Small Business, the requests recorded, staff activity and survey feedback. It makes the service more visible. It does not establish how often a problem was resolved, how long resolution took, what changed for an owner or worker, or what that improvement cost. Missing information in this dashboard does not mean the agency or its partners hold no such records.

The evidence does not by itself justify increasing or cutting the office's budget. It also cannot establish the effectiveness of every small-business program in Portland.

## Different counts describe different parts of the work

| Measure | Reported count | What it means |
|---|---:|---|
| Engagement records | 1,170 | An engagement is not necessarily a distinct business or problem. |
| Engagements with an action log | 1,003 | A linked action does not establish completion or speed. |
| Unique businesses with action logs | 940 | Provider-reported reach, including outside-city and unclassified records. |
| Action logs | 2,183 | Multiple actions may concern the same business. |
| Survey responses | 50 | The number invited and question-level answer counts are not supplied. |
| Appointments | 315 | Appointments are not necessarily distinct owners or businesses. |
| Group events | 178 | Events attended or hosted, rather than participants or successful cases. |

If the engagement cards share the same filters, 1,003 divided by 1,170 is 85.7% with a linked action log. The difference of 167 records is not a count of owners ignored: timing, eligibility, duplicate contacts and logging practices could matter. The dashboard does not establish their frequency.

The 21 monthly engagement counts total 1,170 and span February 2025 through incomplete October 2026. The series begins before OSB's reported May 2025 launch; the relationship between those earlier records and the launched service is unexplained. Exact filter settings are absent. The earlier year-one report uses a different period and potentially different definitions, so subtracting its 759 businesses from 940 does not measure newly served businesses.

## Requests identify service needs among clients

Funding and business assistance lead the recorded requests, with 610 and 449 entries respectively. Those are category counts, not financing awards or distinct businesses receiving money. Finding a program, completing an application, receiving usable financing and achieving stable cash flow are different results.

Needs total 1,716 and services total 3,104, exceeding their corresponding engagement and action-log headlines. Categories may overlap or use different counting rules; they cannot be interpreted as mutually exclusive shares. Some PDF labels are cut off. The webpage uses complete editorial summaries, and the extract preserves the original wording separately. The largest service count is printed as 1.181k and transcribed as 1,181.

The 40 safety/livability/vandalism entries do not establish how important safety is to all Portland businesses. Owners may seek help elsewhere or never contact OSB. Similarly, the food-service and retail shares describe clients in the dashboard's industry chart, not the city's business mix.

## Geography and feedback have limits

The business district categories total 947 against a headline of 940 unique businesses. Event categories total 180 against 178 events. The source does not establish whether overlap, classification or different filters explain these discrepancies. Both the reported totals and category sums remain visible. District counts lack the eligible-business denominators needed to judge equitable access.

The survey's printed positive categories add to 78% feeling supported and 82% likely to recommend OSB. Another 10% are very unlikely to recommend it. These answers describe the people who responded. Without invitations, selection rules and question-level response counts, 50 divided by 940 is not a valid response rate, and the percentages cannot represent everyone served. Satisfaction is useful evidence about experience; it is not a measured economic benefit.

The language and appointment-demographic charts also lack clear denominators and repeat-visit rules. Appointment demographics cannot automatically describe all 940 businesses or demonstrate equal access. Unprinted values remain unknown.

## What would show that the help works?

For an owner trying to open a shop, success may mean an opening with fewer delays and affordable costs. For a solo worker, it may mean more income after expenses and fewer hours of administration. For an employer, it may include more stable cash flow and better jobs. Growth is not every owner's goal.

Evidence of effectiveness would connect a business's original problem to completed assistance and subsequent results, including unsuccessful and unfinished cases. It would distinguish an approved loan from money received, and a referral from a service actually completed. It would also account for businesses that never seek help.

A business improving after assistance does not establish how much the assistance caused. Economic conditions and differences between participants and nonparticipants also matter. Comparisons, sustained outcomes and actual costs are needed to judge which services offer the greatest public benefit. That is the unanswered question for Portlanders behind the dashboard's activity counts.
