Quick answer
Earned media is publicity a brand gets without paying for the placement, such as press coverage, independent reviews, customer posts, backlinks and word of mouth. Unlike paid and owned media, you do not control it, and it is the hardest to attribute.
What is earned media?
In marketing, earned media is publicity a brand gets without paying for the placement: news coverage, reviews, social posts by customers, mentions in newsletters and podcasts, links from other sites, and word of mouth. Someone else decides to talk about you, in their own channel and their own words.
It is one of three media types. Paid media is space you buy (ads, sponsorships). Owned media is channels you control (your website, blog, app, email list, social accounts). Earned media is the one you cannot buy directly, only deserve or provoke. The differences between the three come down to two things: how much control you have and what each costs. They also differ in how measurable they are. Paid media comes with click and impression data; owned media sits on your own analytics. Earned media is the hardest to attribute, because the person who writes about you does not tag their links. You measure it indirectly, through referring domains, referral traffic and changes in branded search.
Paid, owned and earned media compared
| Paid media | Owned media | Earned media | |
|---|---|---|---|
| Who controls the message | You, within the platform's rules | You | The third party |
| Cost | Pay per click, impression or placement | Production and upkeep | No placement fee; PR, product and content effort to earn it |
| Examples | Paid search ads, social ads, sponsored posts | Website, blog, app, email newsletter, brand social accounts | Press articles, independent reviews, customer posts, backlinks, forum threads |
| Speed | Immediate, stops when spend stops | Builds over months | Spiky: a burst after coverage, then a long tail |
| Trust from the audience | Lowest; labelled as ads | Medium | Highest; it is someone else's opinion |
| How to measure | Platform clicks and UTM-tagged sessions | Your analytics: pages, events, goals | Referring domains, referral sessions, branded search impressions and clicks, mentions |
The lines blur in practice. When you pay an influencer, the post is paid media even though it lives on their account; in the US, the FTC's Endorsement Guides require the material connection to be disclosed. When you share a customer's review on your own blog, the review was earned and the blog post is owned. "Shared media" is sometimes added as a fourth type for social channels where all three mix.
Earned media examples
- Press coverage. A trade publication covers your product launch or quotes your founder.
- Independent reviews. A blogger, YouTuber or review site evaluates your product without payment.
- Customer social posts. A user posts a screenshot of your app or tags your brand.
- Backlinks. Another site links to your guide or tool because it is useful.
- Community mentions. Your product comes up in a forum thread, Q&A site or Slack community.
- Awards and lists. You appear in an editorially chosen list or win an industry award.
- Word of mouth. Someone tells a colleague, who later types your brand name into a search engine.
How to measure earned media
Earned media rarely arrives as a clean, tagged click. These are the signals that hold up:
- Referral sessions by referring domain. When someone clicks a link in an article, the browser sends the article's site as the referrer. Since most browsers default to the
strict-origin-when-cross-originpolicy, you usually see the domain, not the full article URL. A new domain at the top of your referrers report is the most direct evidence. See referral traffic for how referral is defined. - Branded search impressions and clicks. Many readers do not click; they search your name later. In Google Search Console, impressions and clicks on queries containing your brand rise after good coverage. Branded vs non-branded keywords explains how to split them.
- New referring sites and backlinks. Syndicated copies and follow-up articles add more referring domains over the following weeks.
- Conversions from those sessions. Referral and branded-search visitors who reach a goal tell you whether the coverage reached the right audience.
- Direct traffic, with caution. Direct visits often rise too, but direct is also where untagged email, apps and typed URLs land. Treat a direct uplift as supporting evidence, not as the coverage's result.
Some earned impact leaves no trace at all: a reader who sees your name in a newsletter and buys months later. Accept that you are measuring a lower bound.
Worked example: one article, 14 days before and after
Sample data: a trade publication runs a story about a SaaS product on day 0. The team compares the 14 days before (days −14 to −1) with the 14 days from publication (days 0 to 13), and uses the 14 days before that (days −28 to −15) to estimate normal growth.
| Signal | Days −28 to −15 | Days −14 to −1 | Days 0 to 13 | Raw change | Normal growth | Expected | Lift |
|---|---|---|---|---|---|---|---|
| Referral sessions from the publisher's domain | 0 | 0 | 1,260 | +1,260 | – | 0 | 1,260 |
| Branded search clicks (Search Console) | 2,300 | 2,380 | 3,710 | +1,330 | 3.5% | 2,463 | 1,247 |
| Branded search impressions | 9,500 | 9,800 | 15,600 | +5,800 | 3.2% | 10,114 | 5,486 |
| Direct sessions | 6,150 | 6,300 | 7,050 | +750 | 2.4% | 6,451 | 599 |
How the columns are computed:
- Normal growth = days −14 to −1 ÷ days −28 to −15 − 1. Branded clicks: 2,380 ÷ 2,300 − 1 = 3.5%.
- Expected = days −14 to −1 × (1 + normal growth). Branded clicks: 2,380 × 1.035 = 2,463.
- Lift = days 0 to 13 − expected. Branded clicks: 3,710 − 2,463 = 1,247.
What the team can claim: 1,260 referral sessions are directly traceable to the article, and about 1,247 extra branded-search clicks are a reasonable estimate of readers who searched instead of clicking. Branded click-through rate stayed flat (2,380 ÷ 9,800 = 24.3% before, 3,710 ÷ 15,600 = 23.8% after), which fits more people searching rather than a ranking change.
What it should not claim: the 599 extra direct sessions. Some are probably readers typing the URL, but the same window also contained a newsletter send, so direct is reported as "possibly related" and left out of the coverage total. The defensible figure is roughly 2,500 visits (1,260 + 1,247), not 3,100.
Common mistakes
- Crediting all direct traffic to the coverage. Direct is a catch-all. Use a baseline and keep it separate.
- Comparing with a quiet week. Compare equal windows and adjust for the normal trend, or a seasonal upswing looks like PR impact.
- Counting paid placements as earned. Sponsored articles and paid influencer posts are paid media, whatever they look like.
- Measuring only clicks. Many readers never click. In zero-click searches and in print or audio coverage, branded search is often the only trace.
- Stopping after a week. Syndication and follow-up links can keep sending referral traffic for weeks.
Measuring earned media in TapCub
Here is the example above, done in TapCub.
- Mark the publication date. In Configuration › Annotations, add a single-day annotation for day 0. It shows as a dashed line on Web analytics › Overview, so everyone reading the traffic chart sees when the story ran.
- Find the referring domain. In Web analytics › Sources, open the Referrers tab and set the date range to the 14 days after publication. The publisher's domain appears with its visitors, sessions, pageviews, bounce and duration. Click the row to filter the other reports to that source and see which pages those readers landed on. Switching the count to Every arrival also counts readers who came back from the article in the middle of a session.
- Read branded search. Connect Google Search Console or Bing Webmaster in Monitoring › Search performance, then compare clicks and impressions for your brand-name queries in the two windows. Search data usually lags 1 to 3 days, so wait before calling the result.
- Check conversions. Web analytics › Goals shows conversions by source for each goal, so you can see whether readers from the publisher's domain signed up, not just visited.
- Keep bots out of the baseline. TapCub counts bots and AI crawlers separately, so a crawler fetching the article's links does not inflate referral or direct counts.
TapCub does not monitor media mentions or count backlinks; it measures the visits and searches the coverage produces.
See also: What is owned media? Examples, vs paid and earned, and how to measure it.
See also: What is UGC? User-generated content meaning, types and how to measure it.
Sources
- Google Search Console Help, Performance report (Search results).
- MDN Web Docs, Referrer-Policy header.
- US eCFR, 16 CFR Part 255: Guides Concerning the Use of Endorsements and Testimonials in Advertising.
Was this explanation clear?
We use your answers to fix and expand this page.
Couldn’t send. Please try again in a minute.
Thanks for the feedback. We’ll use it to improve this page.You’ve already sent feedback on this page. Thanks.

Product Adoption Consultant
Explores customer onboarding, live conversations, and self-service, turning common usage questions into practical guidance and improvements.