Editorial Standards
These standards govern everything Civic Forecast publishes: hubs, data cards, stat tiles, maps, explainers, alerts, and any narrative supplied by a content partner. They describe how we source and verify material, how we separate record from interpretation, where automation ends and human review begins, how sponsorship is kept away from editorial judgment, and how to get something corrected, answered, or withdrawn. They supplement our Terms and Conditions, our Privacy Policy, and our Open APIs and Public Data policy.
Scope and purpose
In short: We publish derived civic signals with their sources attached, a person reviews narrative before it ships, and paid placement never buys editorial treatment.
These standards apply to civicforecast.com and the Civic Forecast application, both operated by Show Me Television, LLC ("Civic Forecast," "we," "us," or "our"), and to any material a content partner supplies for publication on our surfaces. Partner-supplied narrative is held to the same standards as material we produce ourselves; a partner's byline does not lower the bar it has to clear.
Civic Forecast's data and technology partner is API Automations, which publishes its own editorial standards covering the segments and data products it produces across its channels and partner publishing avenues, Civic Forecast among them. Both sets of standards apply to material API Automations produces for our surfaces. Where they differ, the stricter requirement governs, and nothing in an upstream standard reduces what we commit to here.
These standards are a statement of our editorial practice. They are not legal advice, they do not create a contractual right for any third party, and they do not modify the Terms and Conditions, which govern your use of the Service.
Our editorial mission
Education is the foundation of trust. Civic Forecast exists to make public records legible to people who are not specialists in them, and every editorial decision we make is measured against whether it leaves a reader better able to understand how policy, money, and power actually work.
As our library of hubs, signals, and explainers expands, we hold that work to accuracy, clarity, and ethical responsibility rather than to reach. We would rather publish less and be relied upon than publish more and be checked.
What Civic Forecast is, and is not
Civic Forecast is a civic-education service. We normalize public, licensed, and government data into aggregate, source-attributed civic signals, and we publish plain-language context alongside them.
We are not a news outlet and we do not republish source content as journalism. We are not a data reseller: we do not sell, redistribute, or paywall a provider's records, and every signal we show is a derived indicator rather than a copy of someone else's dataset. We say this here because it bounds what these standards can promise — we do not run a newsroom, and we do not hold ourselves out as one.
How we create content
- Research and verification. Every signal and explainer is grounded in primary public records, reputable sources, technical documentation, and first-hand domain experience.
- Expert input. Editorial work is done alongside the engineers, analysts, and subject-matter specialists who build and operate the pipeline, so that what a card says matches what the data actually supports.
- Clarity first. We simplify without oversimplifying. Material is written to serve readers, researchers, journalists, and decision-makers alike, and we do not trade precision for a cleaner sentence.
Sourcing and verification in practice
- We build on primary public records. Where a primary record exists, we use it in preference to a secondary description of it.
- Figures are verified against their originating source before publication, and the identifier that lets a reader retrieve the same record is carried with the figure wherever the source provides one.
- Every data card attributes its source and links back to it. Attribution stays visually present on cards, charts, maps, and briefs.
- When a number changes at the source, our surface reflects the update rather than preserving the figure we published first.
- Where a source publishes methodology, revision, or version metadata, we preserve it so a reader can judge freshness and provenance for themselves.
- We do not present a single source's estimate as settled fact where the underlying measure is genuinely contested, and we say when a figure is provisional, modeled, or subject to later revision.
Separating record from interpretation
Structured data and narrative context are kept separate by design and are visually distinguishable on every surface. A reader should never have to guess which part of a page is a public record and which part is our reading of it.
Editorial framing does not alter underlying figures. Narrative may explain, contextualize, or note the limits of a number; it does not restate the number differently from the source, and it does not silently exclude records that complicate the framing.
Forecasts, indices, scores, severity bands, and probabilistic estimates are analytical outputs, not statements of fact about what will happen. They are labeled as estimates wherever they appear.
Automation, AI, and the human review gate
In short: Automation prepares the draft. A person holds the review gate, and nothing publishes without passing it.
Our pipeline is automated end to end for acquisition, normalization, geographic alignment, and signal derivation. That automation produces the structured, aggregate indicators shown in the product. It does not create a substitute dataset and it does not enable record-level access.
Automated systems, including machine-learning and language models, may be used to prepare draft narrative, summaries, tags, and headlines. Every narrative or explainer asset is source-aware and release-checked before distribution: it carries its source and credit line, its release status is confirmed, and a human reviews it against the underlying record. Nothing publishes to the app or the site without passing that review. A person, not a model, is accountable for what appears.
Source and partner data is not used to train, test, or improve machine-learning, LLM, or AI systems outside the scope of its license.
Automation makes mistakes at a different rate and in a different shape than people do. If you find output that reads as machine-generated error — a confident figure with no support, a source that does not say what we claim it says — report it under the corrections process below and say so, and we will treat it as a review-gate failure rather than an isolated typo.
Nonpartisan framing
Civic Forecast does not endorse, oppose, rate, or recommend any candidate, party, campaign, ballot measure, or elected official, and we do not accept sponsorship from candidates, campaigns, parties, or political action committees for placement on our surfaces.
Explainers describe what public records show and the range of mainstream interpretations of them, without advocating a position. Covering a subject is not a judgment about that subject, and the prominence we give a record reflects civic significance rather than a view about who it favors.
Nonpartisan does not mean every claim gets equal weight. Where records plainly show something, we say so.
Editorial independence
In short: Sponsors can buy placement and visibility. They cannot buy coverage, framing, favorable treatment, or advance review.
Our editorial work operates independently from our sales and business development functions. No sponsor, partner, or advertiser dictates what we publish unless that involvement is disclosed transparently in the item itself.
What sponsorship does not buy
- Influence over which records, hubs, or signals we cover, or how prominently.
- Review of, or approval over, editorial content before publication.
- Removal, softening, or delay of unfavorable material, including material about the sponsor.
- Any representation that a sponsor endorses, or is endorsed by, the civic data shown near its placement.
Sponsorship of a hub does not exempt the sponsor from appearing in that hub's records. Where we publish about an organization that funds, sponsors, or partners with us, we disclose that relationship in the item itself.
Conflicts of interest
People who produce or review our editorial material disclose financial, professional, familial, and political interests that a reasonable reader would want weighed, and step back from material where such an interest would bear on the judgment being made. Where a conflict is disclosed but the material still merits publication, we publish the disclosure with it rather than quietly reassigning the work.
Partner, affiliate, and sponsorship disclosure
Civic Forecast is free to use, funded by the optional Premium upgrade rather than by advertising. Sponsorship and brand placement help keep it that way. All sponsored placements are clearly and visibly labeled at the point a reader encounters them, and are optional to engage with.
Some material may include affiliate, backlink, syndication, or sponsored links or mentions. Where it does, that relationship is disclosed in the item itself rather than only in this policy, and we include such a link only where the product or service meets the same technical and ethical standards we apply to our own work. A commercial relationship never converts into a recommendation we would not otherwise make.
We do not run surveillance advertising, we do not sell user-level data, and partner reporting is aggregate-only. We do not accept sponsorship from candidates, campaigns, parties, or political action committees for placement on our surfaces.
Corrections and content updates
We proactively review and update published material to reflect evolving records, regulation, and methodology. When we get something wrong, we fix it and say that we fixed it. We do not silently edit a published claim into a different one.
| Change type | When we use it | How it is marked |
|---|---|---|
| Update | The source revised, restated, or superseded its own record and our surface follows it | Timestamp refreshed |
| Clarification | What we published was accurate but the wording invited a materially wrong reading | Dated note |
| Correction | We published something factually wrong, or attributed a figure to the wrong source | Dated correction |
| Retraction | The basis for the item does not hold and context cannot repair it | Withdrawn and logged |
Routine refreshes as a source updates are not corrections and are reflected by the retrieval timestamp rather than by a note. Corrections and retractions stay attached to the item; we do not remove the record of having been wrong.
Reporting an error
Write to contact@civicforecast.com with the page or card, the figure or statement at issue, what is wrong, and the source that shows the correct information if you have it. We aim to acknowledge within five business days and to tell you what we did or why we did not act. Errors about a source's own data or attribution are also covered by our Open APIs and Public Data policy.
Right of reply, amendment, and unpublishing
If you are named in, or identifiable from, material we publish and you believe it is inaccurate or unfair, write to contact@civicforecast.com. We will review it on the merits, and where a response is warranted we will carry it near the material rather than only in correspondence.
Much of what we show is drawn from public records we did not create. We can change how a record is presented on our surfaces, and we can stop presenting it. We cannot amend the underlying government record; that has to be sought from the agency that published it.
We unpublish rarely, because a public record of what we said is part of being accountable for it. We will consider removal where material is inaccurate and cannot be corrected, where continued publication creates a concrete risk of harm to an identifiable person, or where a provider withdraws our right to use the material. Where we unpublish, we note that we did.
Feedback, complaints, and escalation
We value our readers, and feedback is how these standards stay honest. If you have questions, suggestions, or want to collaborate, write to contact@civicforecast.com — it helps us tailor what we publish to how people actually use it.
Editorial complaints go to the same address. Tell us what you are complaining about, what standard in this document you believe was not met, and what outcome you are asking for.
If our first response does not resolve it, say so and ask for escalation; the complaint is then reviewed by someone who was not responsible for the material. We will give you a substantive answer rather than a restatement of the first one.
Limits of what we publish
Civic Forecast is an information service. Nothing we publish is legal, financial, tax, electoral, or professional advice, and nothing we publish is an official communication of any government body.
For anything carrying a deadline or a legal consequence — voter registration, ballot status, filing requirements, or eligibility — verify with your state or local election official or the responsible agency before you act.
Changes and contact
We may revise these standards as the product and our partnerships change. Material changes are reflected in the "last updated" date at the top of this page.
Civic Forecast is operated by Show Me Television, LLC. Editorial questions, corrections, replies, and complaints: contact@civicforecast.com. Partnerships and sponsorship: partnerships@civicforecast.com.