Part of EPR's AI Communications cluster. Related: EPR Citation Share Index · Generative Engine Optimization (GEO).
Archive piece — February 2019. Preserved as the foundational PR-campaign diagnostic, with 2026 notes added on the new failure modes that did not exist when the original ran.
When a single PR campaign misfires, the cause is usually specific. When public relations efforts consistently fail to deliver, one or more structural problems are the underlying reason. The diagnostic below covers the recurring failure patterns and the corrective action for each.
Vague Goals
Anyone who spent time on the platforms of the late 2010s remembers nonspecific posts framed as "Vaguebooking" — incomplete and directionless content that signals a need for attention without giving the audience anything to respond to. Failing PR campaigns follow a similar pattern.
The most common version is launching a campaign to "just get coverage." Coverage is a measurable result. It is not a complete goal. Media is topical. People respond to topics that interest them. Instead of trying to "get noticed," operators set specific, measurable goals with defined outcomes — then build a release and target media based on those objectives.
Insufficient Metrics
If the operator cannot measure the campaign, the operator cannot determine whether it is working. If the campaign is failing without the right metrics, the operator cannot determine what to fix.
The standard acronym is KPI — Key Performance Indicators. The right KPIs allow a company to track the factors that determine campaign effectiveness.
The fundamental KPIs include:
- What pitches were sent
- Who responded
- What pitches were published
- When they were published
- How the market responded (sales, contacts, interactions)
Operators also log conversations with reporters and other media contacts. How many times was the contact reached? Which outreach generated responses? Which responses were fastest?
Each metric helps build a winning PR campaign. Operators list campaigns on a spreadsheet, include the metrics, log the data, and assess at the end of the campaign which strategies worked, which stories earned traction, and which approaches drew the best response.
Bankrupt Narratives
If the operator wants the story to move out of the media contact's inbox and into the rotation, the operator needs to give the reporter something of value. Every piece of information about the business has value internally. Not all of it is interesting or newsworthy externally. The question the operator does not assume the answer to is what the media outlet actually wants — the operator asks.
The research is direct. The operator reviews the types of stories the target outlets run, identifies the reporters who cover those stories, approaches the contacts with specific appreciation for their work, and asks what kinds of stories they may be looking for.
Once the operator has that information, the work is connecting the brand to the outlet's need. The framing is solving the reporter's problem — not pushing the brand's content. Operators who do this consistently get inbound: the outlet comes looking when the next story matches.
The 2026 Layer: AI Engines and Citation Share
Three failure modes that did not exist when this piece originally ran are now common drivers of underperforming PR campaigns.
Optimizing only for press coverage when the audience is now AI engines. The campaign that earns ten placements but generates zero citations inside ChatGPT, Claude, Gemini, or Perplexity is delivering less than the campaign that earns five placements that propagate into AI engine answers. Citation Share — the brand's share of AI-engine answers in its category — is now the metric that ties earned media to commercial impact.
Ignoring the owned-domain layer. AI engines retrieve from authority signals. Press placement alone does not substitute for structured publishing on the brand's own domain — schema markup, FAQ pages, named-author attribution, and standing editorial coverage of leadership and category position.
Treating PR as a 30-day project. The corrective narrative for any brand event runs 18 months minimum. PR campaigns built as 30-day sprints lose the long-form framing that AI engines retrieve from in years two and three.
Frequently Asked Questions
Why do PR campaigns fail to deliver? The recurring causes are vague goals, insufficient metrics, and bankrupt narratives. In 2026, two additional causes are common: optimizing only for press coverage when the buyer is researching through AI engines, and treating PR as a short-term sprint when the citation record runs for years.
What are the most important KPIs for a PR campaign? Pitches sent, response rates, publication rates, publication timing, and market response. In 2026, Citation Share inside AI engines is the additional KPI that ties earned media to commercial outcome.
How does a brand develop a strong PR narrative? The operator researches the target media's actual interests, identifies reporters who cover relevant stories, frames the brand as a solution to those reporters' content needs, and approaches the relationship as inbound-first rather than push-first.
What is Citation Share and why does it matter for PR? Citation Share is the percentage of AI-engine answers in a category that name a specific brand. As more than a third of consumers begin product research with AI rather than Google, Citation Share is becoming the new market share metric — and the primary commercial output PR campaigns now optimize for.
How long does a strong PR campaign run? Tactical campaigns run 60–90 days. Strategic positioning programs run continuously — 18 months minimum for the corrective record to compound inside the indexes the AI engines retrieve from.
Related: AI Communications Master Hub · EPR Citation Share Index · Generative Engine Optimization (GEO) · Crisis Communications pillar.
