The Skims Refund and What It Teaches About Brand Backlash
Skims offered refunds after its $148 advent calendar drew a TikTok backlash. Rick Pendrick on spotting chatter early and when a refund beats a statement.

Most companies that ask me how to measure PR results (and what counts as real PR results) have already been burned once. They paid an agency, got a monthly report full of big numbers, and still could not say whether anything changed for the business. Impressions went up. Sales did not move.
The problem is rarely the work. It is that nobody agreed on what success meant before the work started. PRSA’s guide to measuring your PR success says there is no one-size-fits-all approach to measuring impact, and that you should begin by learning which key performance indicators (KPIs) your organization’s decision makers already use.
I have run my own agency for eight years, and before that I spent more than 20 years in corporate America and at top New York agencies. I also built Prism’s own AI tools for PR, including media monitoring that checks every story we earn. So I have seen measurement done badly in a lot of ways. Here is how we measure PR results at Prism, step by step, with the metrics that actually matter.
PR is earned, not bought. A reporter decides whether to write your story, so you cannot guarantee a specific outcome the way you can with an ad. That makes measurement feel soft, and it leaves room for vanity numbers.
The most famous vanity number is the advertising value equivalent, or AVE. It tries to put a dollar figure on a story by asking what the same space would have cost as an ad. The measurement body AMEC runs a Say No to AVEs campaign, and agency networks that joined it describe AVEs as limited to measuring only certain components of a communications campaign. Its Barcelona Principles 4.0 call for measuring and evaluating PR in a meaningful, transparent and outcome-driven way.
Outcome-driven is the phrase to hold onto. An output is something your PR team did or got: a pitch sent, a story published. An outcome is something that changed because of it: a visitor who found you, a lead who called, a ranking that moved. Good measurement tracks both and never confuses them.
We never open a report template first. We start by asking what the company needs in the next 6 to 12 months. The answer usually falls into one of these buckets:
Pick one primary goal and one backup. Everything else we measure supports it. If a metric cannot be tied back to the goal, it does not go in the report.
This is the same logic PRSA recommends: define communications goals, then define KPIs that relate to things you can control and analyze, such as business leads and unique audience visits.
You cannot show growth without a starting point. Before any campaign begins, we record:
A one-page baseline of your PR results takes a few hours and saves months of arguing later about whether PR worked.
Outputs are the early PR results. They tell you the engine is running, even before business results show up. We track:
Outputs are necessary but not sufficient. Fifty stories on sites nobody reads is a weak month. Five stories on strong outlets, each with a backlink, can be a very strong one.
If I could pick only one number for PR results, it would be editorial backlinks. Of all PR results, they are the hardest to fake. A backlink from a real news outlet is a public, verifiable vote of confidence. You can check it yourself, and so can Google.
For each backlink we record the story, the outlet, its Domain Rating, whether the link is do-follow (it passes ranking credit) or no-follow, and the date it went live. Over time this shows whether the campaign is lifting your site’s authority or just producing mentions.
It is also why we built a daily check. Links disappear, pages change, and reporters sometimes publish a story with no link at all. We look at every story for a missing backlink, and our unlinked mentions guide covers how to win those back.
Across all clients, Prism has placed 25,000 stories and earned 20,000 backlinks, about 20 backlinks per campaign. Those numbers are useful because every one of them can be checked on a live page.
This is where measuring PR results has changed the most in the last two years. People now get answers from AI tools as well as from Google’s blue links, and those tools lean heavily on earned media. Muck Rack’s What Is AI Reading? report analyzed more than 25 million links in answers from ChatGPT, Claude and Gemini. It found that about 84% of the citations came from earned media, a group that includes journalism, academic research and government sources.
So we add two checks to every report:
AI answers vary from one run to the next, so we ask each question several times and look at the pattern, not a single answer. We never promise a specific AI result, because no one controls it. We track whether earned coverage correlates with the client appearing more often. For more on why this works, read our guide to SEO, AEO and GEO.
Now the PR results that your CEO or CFO cares about. These depend on your goal from Step 1:
PRSA points to engagements, sentiment and time spent on your website as meaningful metrics, and says the right ones depend on the KPIs you defined up front.
Be honest about attribution. PR rarely produces a clean last-click sale. A prospect reads a story, searches your name a week later, and fills out a form from a Google ad. That is still a PR result. We ask every new lead how they heard of the company, and we read branded search trends alongside coverage dates.
The cleanest way to see PR results measured in action is a campaign with a long track record. For Seniorly, a senior living referral marketplace, Prism produced a newsworthy, data-driven industry study every month. In plain words: we analyzed the latest federal data, found the national trend, ranked every state, and pitched national, trade and local media. The Seniorly case study shows what that produced in 2025: 317 stories with editorial backlinks at an average DA of 76.2, including CNBC, Fortune, Newsweek and AARP. Seniorly was acquired by CareScout, a subsidiary of Genworth Financial, in October 2025.
Measured the way I described above, that campaign reads in layers: outputs (stories and outlets), backlinks (count and strength), then the business result (an acquisition we are proud to have supported). No single number tells the story of PR results. The set does.
Here’s a live dashboard for our client CareScout showing what we’ve secured over the last 9 months, which was a full-service campaign, including these monthly studies. We give every client one like it, so no one has to wait for a monthly PDF to see what ran.
A good report on PR results fits on one screen. Ours leads with the headline numbers for the month, then the stories, each with its outlet, Domain Rating, audience and backlink status. Below that: what is coming next, and what we would change.
Four rules keep it honest:
For the sourcing and pitching side behind those numbers, see how we build a media list and newsjack a breaking story.
PR results compound. A single month of PR results can mislead you in either direction. We review the report monthly, to catch problems early, and judge the strategy quarterly, once enough stories and links have landed to see a trend. Our guide on how long PR takes to work explains why the first 90 days are about momentum and not final results.
At each quarterly review we ask three questions: Did we move toward the goal from Step 1? Which angles earned the strongest outlets and backlinks? What should we double down on or drop?
Start with a business goal, record a baseline, then track outputs (stories and outlets), backlinks and Domain Rating, search and AI visibility, and business outcomes like leads and branded search. Review monthly and judge the strategy quarterly.
Common PR KPIs include stories published, outlet quality, backlinks earned, referral traffic, branded search growth, inbound leads and sentiment. Choose the few that tie to your primary goal rather than tracking everything.
No. AMEC’s Say No to AVEs campaign argues against it, and agency networks backing the campaign say AVEs measure only certain components of a communications campaign. Track outcomes such as backlinks, traffic and leads instead.
Compare the cost of the campaign against business results you can trace: leads and sales that mention coverage, referral traffic, and the backlinks and rankings that support your search visibility. Perfect attribution is rare, so ask new leads how they found you.
Monthly for the PR results, quarterly for the strategy. A live dashboard lets you check progress any time between reports.
Good PR results measurement takes a clear goal, a baseline, the right outputs, verified backlinks, search and AI checks, business outcomes and an honest monthly report. Then it takes doing it again next month.
At Prism, an award-winning, full-service PR agency and a digital PR and linkbuilding firm, every client gets reporting built this way as part of our full-service PR and SEO and AEO linkbuilding. You can see what it looks like in our case studies.
But if you don’t want to deal with all this, let us do it for you. Contact us and tell us about your company, or start with our pay-for-performance pilot.
Keep Reading
Skims offered refunds after its $148 advent calendar drew a TikTok backlash. Rick Pendrick on spotting chatter early and when a refund beats a statement.
Poppi's founders are launching Jetty after a $1.95 billion exit. Rick Pendrick on product launch PR: a successful exit buys one free pass, so nail…
How long does PR take? A realistic timeline from groundwork to first stories, backlinks, and compounding results, plus what speeds it up or slows it…
Get Started
Tell us what you want to be known for. We will show you how earned media gets you there.