Measuring your results
Last updated
Measure a press release with three things: pickups (how many outlets published), reach (their combined audience) and replies (journalists who responded). Report earned coverage, referral traffic and share of voice instead of AVE, track AI citations of your release as a new 2026 metric, and judge it against the goal you set before sending.
The three things that actually matter
Most measurement reports drown in numbers and say nothing. Strip it back to three signals and you can read a campaign in a glance. Everything else is detail layered on top of these.
- Pickups. How many outlets actually published or quoted your news. This is the cleanest proof the release worked, and it is the number a founder or board understands instantly.
- Reach. The combined audience of those outlets, their monthly visitors or print circulation added together. One trade title read by your exact buyer can beat a big generalist with no relevance.
- Replies. The journalists who responded, even to decline. Replies are relationship currency: a reporter who answered once is far likelier to open and cover your next story.
Each pickup is a piece of earned media: coverage you won through an editorial decision, not a payment. That is exactly what makes it credible, and exactly why you should count it carefully.
Why AVE is outdated, and what to report instead
For decades, agencies reported AVE (Advertising Value Equivalent): the price you would have paid to buy the same space as an advertisement. It is a comforting number and a misleading one. AMEC, the international association for measurement, and the Barcelona Principles both formally reject it. AVE treats earned coverage as if it were a paid ad, ignores whether the article was positive or negative, ignores whether your message even appeared, and can be inflated simply by picking a bigger rate card.
Report outcomes that map to reality instead:
- Earned coverage. The count and quality of outlets that published, weighted by relevance to your audience.
- Referral traffic. Visitors who arrived on your site from those articles, measured in your analytics.
- Share of voice. Your slice of total coverage in your category versus named competitors over the same window.
- Backlinks. Editorial links from covering outlets, which compound into long-term search authority.
AVE answers a question nobody should ask: what would this have cost as an ad? The honest question is whether the coverage moved the goal you set, and that is the only one worth reporting.On measurement that means something
Tracking AI citations: the new 2026 metric
A growing share of your audience never visits a publisher. They ask ChatGPT, Perplexity, Claude or read a Google AI Overview, and the assistant answers using sources it trusts. Being named or linked there is the newest form of earned visibility, and it sits at the heart of GEO (Generative Engine Optimization). A release that gets cited by an assistant keeps working long after the news cycle ends.
To track it without any special tooling:
- List the five questions your news genuinely answers, in plain language a buyer would type.
- Ask each question in ChatGPT, Perplexity, Claude and Google AI Overviews, and note whether your company, release or newsroom is named or linked.
- Log which prompts surface you and which do not, then repeat the same set every month to see the trend.
- Watch your referral analytics for sessions from sources like chatgpt.com and perplexity.ai, which signal assistants are sending real people.
A hosted, crawlable newsroom is what makes a release citable in the first place, since assistants need a stable public URL to point at. Distributing through PressPilot gives every release exactly that, and the dashboard records opens, clicks and coverage in one place, see the pricing page.
The modern KPIs, in one table
Pick the two or three rows that match your goal rather than reporting all of them. The table tells you what each metric reveals and where to read it.
| Metric | What it tells you | How to track it |
|---|---|---|
| Pickups | Whether the release earned coverage at all | Manual log, Google News, media monitoring |
| Reach | The size of the audience exposed | Sum each outlet's monthly visitors or circulation |
| Replies | Relationship strength with journalists | Your inbox and outreach tool |
| Referral traffic | Real people the coverage sent you | Web analytics, source and medium |
| Backlinks | Long-term search authority gained | Search Console, a backlink checker |
| Share of voice | Your visibility versus competitors | Media monitoring filtered by competitor |
| AI citations | Visibility inside assistant answers | Manual prompt checks, AI referral traffic |
One dashboard, not ten
You do not need an enterprise platform. A single sheet with one row per release and a column per metric beats a tool nobody opens. Capture the date, the pickups, the combined reach, the reply count, referral traffic at 30 days, any backlinks, and a short note on AI citations. Update it once a week. Over a year that sheet becomes the most honest picture of what your press work actually delivers.
Defining success per goal
The same numbers can mean a triumph or a flop depending on why you sent the release. Set the target before you hit send, then judge against it:
- Awareness. Reach and share of voice are what count. Ten relevant trade pickups beat one buried mention in a giant outlet.
- Links and SEO. Backlinks and referral traffic matter most. Two editorial links from authoritative sites can outvalue a hundred no-link reprints.
- Hiring. Watch applications and the quality of inbound candidates in the weeks after coverage, not the pickup count.
- Fundraising. Track investor replies and warm introductions sparked by the news, since a single right reader can be the whole point.
With your metrics defined, the last chapter shows them in action. Move on to chapter 7, the Carimmat case study, where a real campaign is broken down outlet by outlet, with the results behind it.
Frequently asked questions
- How do I measure press release ROI?
- Compare the result to your goal, not to a dollar figure. Track pickups (how many outlets published), reach (their combined audience), referral traffic to your site, and the concrete outcome you wanted: links, signups, applicants or investor replies. Divide the value of that outcome by what the campaign cost you. Avoid AVE, which invents a number that does not map to real impact.
- What are good PR KPIs in 2026?
- The KPIs that hold up are earned coverage (number and quality of outlets that published), combined reach, referral traffic and resulting backlinks, share of voice against competitors, journalist reply rate, and AI citations of your release in ChatGPT, Perplexity and Google AI Overviews. Pick the two or three that match your goal rather than tracking all of them.
- Is AVE still used to measure PR?
- It still appears in legacy reports, but the industry has formally moved away from it. AMEC, the international measurement body, and the Barcelona Principles both reject AVE because it equates earned coverage with paid advertising, ignores sentiment and message, and invents a number you can inflate at will. Report earned coverage, reach, referral traffic and share of voice instead.
- How do I track AI citations of my press release?
- Ask the assistants directly. Query ChatGPT, Perplexity, Claude and Google AI Overviews with the questions your news answers, then check whether your company, release or hosted newsroom is named or linked. Repeat monthly, log which prompts surface you, and watch your referral analytics for traffic from chatgpt.com, perplexity.ai and similar sources. A hosted, crawlable newsroom makes you far easier to cite.
- How long should I wait before measuring a press release?
- Read pickups and reach within 48 to 72 hours, when the news cycle peaks. Give referral traffic, backlinks and AI citations two to four weeks, since indexing and assistant retraining lag. Lock your final numbers at 30 days, then compare them to the goal you set before you hit send.