
Guideline 18 calls for a central, searchable home for public data and for the policy that governs it, so residents, journalists and developers know exactly where to look and what is available.
Of all the provisions found in municipal open data policies, the requirement for a central portal is among the most widely adopted. It is also the provision most often mistaken for the whole policy.
What the guideline means
Data scattered across department web pages is hard to find and easy to forget. Guideline 18 asks a government to create one central location for publishing datasets and explaining its open data policy. That hub can be a general-purpose catalog or a family of topical portals (spending, ethics, transit), as long as related portals point to one another.
A good portal does more than list files. It documents each dataset (see Guideline 13), shows what exists through an inventory, lets search engines index it, and ideally publishes usage analytics so both users and publishers can see which datasets matter most.
Model policy language patterns
Typical portal provisions, paraphrased:
- The city shall maintain a single web portal through which all public datasets are made available, managed by a named department.
- Departments shall publish datasets to the portal, or to an approved departmental site linked from it.
- The portal shall include the open data policy, the dataset inventory, technical standards and a feedback mechanism.
- The portal shall go live within a stated number of days of adoption (see Guideline 27).
Where it appears in adopted policies
Portal language is close to universal among cities that have passed an open data policy. A few illustrations from this index:
- New York City's 2012 Local Law 11 assigned the city's technology department to manage a single web portal for public data sets.
- Boston's open data policy, set by a 2014 executive order, directed departments to publish appropriate datasets on the city's open data portal.
- Minneapolis, MN (2014) required a city open data portal to be available within 120 days of enactment.
- San Francisco, CA (2013) charged its chief data officer with providing and managing a single portal.
Chicago, Philadelphia, Louisville, Austin, Washington DC and dozens of others in the full index include similar provisions.
Common pitfalls
- Portal as finish line. Launching a catalog is visible and satisfying; keeping it current is not. Many portals filled quickly and then went stale.
- Blocking search engines. If datasets cannot be indexed, they are effectively hidden.
- Thin documentation. A table without definitions, units or a contact person generates confusion rather than reuse.
- Fragmentation. Topic portals that do not link to each other recreate the scattered-website problem the guideline was meant to solve.
A practical checklist for portal provisions
- Designate one office responsible for the portal and its uptime.
- Require that the portal list every published dataset and link to the full inventory, including datasets not yet released.
- Allow search engines and other third parties to index the catalog.
- Publish the open data policy, technical standards and annual reports on the portal itself.
- Offer both bulk downloads and programmatic access for each dataset.
- Show usage statistics and provide a visible feedback channel.
- Plan for migrations: stable dataset identifiers and redirects keep links working when platforms change.
What portals teach about the value of operational data
A public portal is a lesson in packaging. Buyers of data, whether civic developers or AI/ML teams, judge a dataset first by whether they can find it, understand it and trust that it will be maintained. The same is true of private operational data: a clean catalog, a data dictionary and a clear refresh schedule raise what a dataset is worth.
If your business holds operational records that could inform models, you do not need to build a portal. You need an honest inventory and a de-identified extract. The guide from open data policy to data value walks through the parallel, and sell my data explains how a de-identified dataset is listed and priced.
We never take HIPAA/PHI data or personal customer data. Everything is de-identified before it leaves your control.
Frequently asked questions
Is launching an open data portal the same as having an open data policy?
No. A portal is infrastructure; a policy sets the rules for what goes on it, how fast, in which formats, under what license and with what oversight. Many cities launched portals before adopting a policy, and those portals often stalled because no one was accountable for keeping them current.
Should every department have its own data site?
Departmental or topical sites can work, but the guideline recommends a single central catalog that lists everything and links to specialized portals. Users should not need to know a city's organizational chart to find a dataset.
What features make a portal useful?
A searchable catalog tied to the data inventory, clear metadata for each dataset, bulk downloads and APIs, a feedback channel, indexing by search engines, and published usage statistics. Usage numbers help residents see what others find valuable and help agencies prioritize maintenance.
How quickly did cities typically launch portals after adopting a policy?
Timelines varied. Some policies set a short deadline, such as the 120-day requirement in Minneapolis's 2014 policy, while others relied on an implementation plan drafted after adoption. Short, specific deadlines tended to produce faster launches.
Policies in the index that reflect guideline 18
| Place | Adopted | Instrument | Guidelines |
|---|---|---|---|
| Washington, DC | Jun 12, 2006 | Executive memorandum | 1131822 |
| San Francisco, CA | Oct 21, 2009 | Executive directive | 11824 |
| San Francisco, CA | Oct 29, 2010 | Ordinance | 47811182526 |
| Cook County, IL | May 4, 2011 | Ordinance | 12378131825 |
| Raleigh, NC | Feb 7, 2012 | Council resolution | 381618 |
| New York, NY | Mar 7, 2012 | Local law (Local Law 11 of 2012) | 14578101118 |
| Providence, RI | Mar 19, 2012 | Resolution | 31824262729 |
| Philadelphia, PA | Apr 26, 2012 | Executive order (No. 1-12) | 34578182224 |
| Madison, WI | Aug 15, 2012 | Ordinance | 457810111318 |
| Chicago, IL | Dec 10, 2012 | Executive order | 34678101113 |
| Montgomery County, MD | Dec 17, 2012 | County legislation (Bill 23-12, Open Data Act) | 457811131822 |
| San Francisco, CA | Mar 28, 2013 | Ordinance | 34578101118 |
| South Bend, IN | Aug 22, 2013 | Executive order | 237810182528 |
| Austin, TX | Aug 26, 2013 | Administrative memo | 123457810 |
| Louisville, KY | Oct 11, 2013 | Executive order (No. 1, Series 2013) | 1234781018 |
| Oakland, CA | Oct 15, 2013 | Council resolution (No. 84659) | 2345781011 |
| Honolulu, HI | Nov 27, 2013 | Ordinance | 237810111820 |
| Los Angeles, CA | Dec 18, 2013 | Executive directive (No. 3) | 37818222630 |
| Pittsburgh, PA | Mar 19, 2014 | Ordinance | 12345678 |
| Hartford, CT | Mar 26, 2014 | Executive order | 34789182224 |
| Boston, MA | Apr 7, 2014 | Executive order | 3891318202425 |
| Nashville, TN | May 12, 2014 | Executive order (No. 043) | 2345781011 |
| Chattanooga, TN | May 31, 2014 | Executive order | 123467810 |
| Kansas City, MO | Jun 19, 2014 | Ordinance (No. 140443) | 81018242713 |
| Washington, DC | Jul 21, 2014 | Executive directive (Mayor's Order 2014-170) | 12356789 |
| Minneapolis, MN | Aug 1, 2014 | Council-adopted policy | 34568131819 |
| Houston, TX | Sep 18, 2014 | Administrative policy (AP 8-7) | 234567810 |
| San Diego, CA | Nov 24, 2014 | Administrative policy | 45818202427 |
| Denton, TX | Feb 9, 2015 | Administrative directive | 357811121618 |
| Detroit, MI | Feb 19, 2015 | Executive order | 818 |
| Suffolk County, NY | Jul 31, 2015 | Administrative policy | 78131820222426 |
| Jackson, MS | Sep 1, 2015 | Executive order (No. 2015-06) | 12345678 |
| Cambridge, MA | Sep 21, 2015 | Ordinance | 12345678 |
| Asheville, NC | Oct 13, 2015 | Council resolution | 13489101118 |
| Durham, NC | Nov 23, 2015 | Administrative policy | 7818192430 |
| Tulsa, OK | Dec 21, 2015 | Executive order | 123458910 |
| Seattle, WA | Feb 1, 2016 | Administrative policy (CTO policy) | 14678101112 |
| Baltimore, MD | Mar 7, 2016 | Ordinance | 1781113182426 |
| San Jose, CA | Apr 5, 2016 | Council policy | 1681113182230 |
| Las Vegas, NV | Apr 12, 2016 | Administrative policy (City Manager policy CM510) | 12345678 |
| Saint Paul, MN | Apr 27, 2016 | Council resolution | 12578101113 |
| Milwaukee, WI | Jul 14, 2016 | Council resolution | 12356789 |
| Bellevue, WA | Aug 1, 2016 | Management policy | 15678101118 |
| New Orleans, LA | Sep 1, 2016 | Executive order (MJL 16-01) | 124511182124 |
| Naperville, IL | Nov 15, 2016 | Council resolution | 45789101112 |
| Fort Collins, CO | Feb 7, 2017 | Council resolution | 12345678 |
Sources
Updated December 31, 2024.