so there's no flash<br>of the wrong palette before React hydrates. Mirrors the useTheme hook<br>(storage key "rw-theme"; classes rw-theme-dark|light|ascii). -->

RuntimeWire launches a live dashboard with auditable readership metrics - RuntimeWire

RuntimeWire

You're browsing RuntimeWire with JavaScript disabled. Articles and<br>navigation work fully. Interactive features — search, comments,<br>and newsletter signup — require JavaScript.

Why it matters

Traffic claims are easy to inflate and difficult to compare. Publishing definitions, exclusions and failure states gives readers a way to judge the measurement itself.

A public metric you cannot audit is marketing. Today, I opened RuntimeWire's live transparency dashboard so readers can inspect how this publication measures readership, subscribers, story performance and outside citations.

The dashboard pulls continuously from RuntimeWire's source-of-truth database. The figures are not copied into a presentation layer, adjusted for a launch announcement or frozen when they look flattering. Each metric carries its own definition and methodology because a number without either forces the reader to trust the publisher's interpretation.

I founded RuntimeWire around a simple problem: AI has made it much easier to build software, while distribution and credible coverage remain scarce. That same technology has also made publishing at scale cheaper. It has become easier for media companies to produce more material and harder for readers to judge the claims attached to it. Showing our counting rules is one small way to narrow that trust gap.

What counts as a read

RuntimeWire records reads from first-party pageview logs. We de-duplicate them using a per-view token and require a minimum amount of time on the page. That reduces the noise from refreshes, accidental loads and automated traffic.

It does not create perfect bot filtering, and we will not claim that it does. We also exclude anomaly windows when the logs show a bot storm or an isolated surge that would distort the underlying trend. A lucky spike should not be presented as durable audience growth.

The dashboard buckets readership by Central Time days, matching the reporting period RuntimeWire uses internally. The public numbers and the newsroom numbers come from the same underlying system.

What counts as a subscriber

The subscriber total includes confirmed email addresses after double opt-in. Pending registrations do not count.

That rule produces a smaller number than counting every submitted form, but it answers the question that matters: how many people completed the subscription process and agreed to receive RuntimeWire. An unconfirmed address is an attempt, not a subscriber.

What counts as a press mention

RuntimeWire logs independent mentions that have been reviewed and verified by an editor. Our own pages, copied listings and generic directories do not qualify.

The standard is earned citation. If another publication, aggregator or independent community references RuntimeWire's reporting, it can appear. If we control the page or the mention exists solely because a database ingested our name, it stays out.

Why story data arrives late

The dashboard shows which stories drove readership, but story-level performance appears after a deliberate delay. A transparency page should not become a live assignment feed for competitors, sources or even our own newsroom.

That delay also reduces the temptation to chase an early burst of traffic before the readership pattern has settled. The goal is to explain what happened, rather than turn every hour of audience activity into an editorial instruction.

What happens when a metric breaks

Public dashboards usually fail in one of two ways: they display a zero that looks real, or they quietly keep serving an old value without saying so.

RuntimeWire does neither. If a metric cannot be recomputed, the dashboard preserves the last known good value and flags it as stale. The system also alerts us so the underlying job can be repaired. Readers should know when a number has stopped updating instead of being asked to interpret a technical failure as performance.

This dashboard will change as RuntimeWire changes. The rule will remain fixed: if we cannot define a metric, trace its inputs and recompute it, it does not belong on the page.

Come look at the dashboard and tell us what is missing.

Reader comments

Conversation for this story loads after sign-in.