Idea Generation
Practical ways to find PR-worthy story angles for any brand
⌚ 11 min read · 2,530 words
proven methods
By Salva Jovells, Founder of Presslei
When I started Presslei, I had a problem I hadn't anticipated.
For 10 years at Hockerty and Sumissura, I analyzed other people's campaigns. I reverse-engineered what worked, catalogued patterns, built spreadsheets. I was the guy studying the game film, not the one on the pitch. Then I launched my own agency and suddenly I needed to generate ideas — not just study them.
That shift was harder than I expected. Studying 5,272 placements gives you a map of what works. It doesn't give you the instinct to spot a story before anyone else does. That part I had to build from scratch.
So I built a system. Nothing fancy — just a repeatable process for turning real signals into pitchable angles. After a few months of testing, refining, and occasionally failing, I noticed the same 7 idea sources kept appearing in every successful campaign I ran or studied.
This is that system.
In This Article
1. Google Trends — The Fastest Signal You're Not Using Properly
Most people open Google Trends, type a keyword, look at the line, and close the tab. That's not how it works.
Switch to “Rising” queries. Compare two or three terms. Check related topics. What you're looking for isn't the trend itself — it's the gap between the trend and what's been published about it.
Here's an example from my own work. In February 2025, DeepSeek had just exploded into public awareness. Everyone was comparing it to ChatGPT. I spotted the spike in Trends, combined it with a fashion hook — asking both AIs to name the best-dressed women in different countries and comparing the results — and pitched the discrepancies as a story.
That campaign landed Elle.fr, ABC.es, Gala.fr, and Vanitatis. No money changed hands. The story worked because it rode a genuine search spike and reframed it through a specific lens that fashion and lifestyle editors actually care about.
The pitch angle is never “this thing is trending.” It's “here's what this spike actually means for your readers.”
2. BuzzSumo Trending — What's Landing Right Now
BuzzSumo shows you what stories are getting traction today. Not what you think should get traction. Not what a brainstorm room decided sounds clever. What actual humans are sharing.
The move here is reframing. Take a trending story and add your data, your client's expertise, or a counter-angle. If a story about remote work costs is doing well, and your client sells office furniture, you have 30 minutes to get a quote out with one real number attached.
Don't try to compete with the original story. Add to it.
Pro Tip
Track everything. The difference between PR professionals who grow and those who stagnate is measurement. Know your pitch-to-placement rate and which angles convert.
3. TikTok Creative Center — Creator Signals That Editors Don't See Yet
This is the source most PR people ignore because they think TikTok is for 19-year-olds dancing. It's not. The TikTok Creative Center shows you rising hashtags and sounds by country, and those signals hit mainstream media 2-4 weeks later.
When I analyzed the 5,272 placements dataset, Instagram engagement data appeared in 266 campaigns. That number is going to shift toward TikTok. The brands that figure out how to translate creator trends into data stories will get coverage the rest are missing.
The angle that works: “why this trend converts (or doesn't) for [specific audience].” Editors want the business perspective that their lifestyle desk can't provide.
| Method | Time Needed | Cost | Best For |
|---|---|---|---|
| Google Trends | 15 min | Free | Trending topics |
| Competitor monitoring | 30 min | Free | Gap analysis |
| SEO tool data | 1 hour | $99/mo | Data studies |
| Industry surveys | 1-2 weeks | $0-5K | Original research |
4. Muck Rack Trends and Journalist Requests
This is where you flip the process. Instead of starting with your idea, start with what journalists are already asking for.
Muck Rack shows you who's writing about your theme right now. Qwoted, ResponseSource, SourceBottle — these are platforms where journalists post specific requests for sources and data.
From the placement data, I found that the most successful campaigns didn't push a story outward. They matched an existing demand. The journalist was already looking for an expert quote on AI regulation or a data point about housing prices — and the PR team showed up with exactly that, formatted to save the editor time.
Build a five-person micro-list of journalists covering your exact topic. Not 50 generic contacts. Five people whose last three articles match what you can provide.
“PR ideas do not come from brainstorms. They come from paying attention to what journalists are already covering and finding the gap your data can fill.”
— Salva Jovells, Presslei
5. Public Data — The Infinite Source Nobody Mines
Eurostat, ONS, Census data, local government portals. Free, credible, and almost nobody in PR touches it.
In the 5,272 placements I analyzed, geographic rankings using public data were the single highest-frequency campaign pattern. “Top 20 cities for X” indices built from government statistics generated consistent coverage because they combine three things editors want: a ranking (clickable), a methodology (credible), and local angles (relevant to specific audiences).
The trick is pairing two datasets that nobody has combined before. Commute times + coffee prices. Concert ticket costs + flight data. Property values + school ratings. Each combination is a potential story.
One spreadsheet, one chart, one methodology box. That's all you need.
Key Takeaway
PR is a long game. Individual campaigns matter less than building a reputation as a reliable, valuable source that journalists trust.
6. Seasonality and Calendar Hooks
February is the densest month in reactive PR. Valentine's Day, Oscars, Grammys, Super Bowl — all within weeks of each other. The data confirms it: February and summer peak periods generated roughly 140-143 placements each in the dataset.
But the obvious seasonal hooks are crowded. Everyone pitches Valentine's Day gift guides. The better play is timing a ranking or data story to a tentpole event rather than directly covering the event itself.
Tax season? Pitch a regional cost-of-living comparison. Back-to-school? Pitch a “hidden costs parents don't budget for” data story with local angles. The event provides the news hook. Your data provides the substance.
Map out your tentpole dates for the next quarter. For each one, ask: what's the data question nobody has answered around this event?
7. Competitor Formats — Steal Structures, Not Headlines
This is the one most people get wrong. They look at a competitor's successful campaign and try to do a version of it. That produces derivative work that editors have already seen.
What you should steal is the format, not the topic. If a casino brand got 50 placements with a “fictional character wardrobe valuation” campaign, don't do another fictional character wardrobe valuation. Take the format — “valuation of something people haven't thought to value” — and apply it to your category with fresh data.
In my dataset, 10 distinct patterns accounted for the vast majority of successful placements. Rankings. Expert tips tied to seasonal hooks. Price comparisons across regions. Social media engagement analyses. These are structural templates that work across verticals because they match how newsrooms think.
Scan the “best campaigns of 2026” roundups. Write down the structure of each winner. Ignore the topic. That structure is your starting point.
Pick One Lane Per Idea
Most campaigns that land fall into one of three lanes:
- Reactive — a speedy quote tied to a live story. You need it out in hours, not days.
- Data-led — an index or ranking built from a public dataset. This takes a day or two to build properly.
- Opinion/Explainer — a founder or analyst perspective on a platform shift or culture moment. Needs a real credential behind it.
Trying to be all three at once dilutes everything. Pick one lane. Execute it well.
Time-Boxed Ideation and AI Prompts
I use AI as a drafting tool, not a replacement for judgment. It's fast at structuring raw inputs into something pitchable. It's terrible at knowing whether a story is actually interesting.
Here's how I use it at three different time scales:
10 Minutes (You're Slammed)
Open Trends + one trade outlet. Draft three quotes and a subject line.
Prompt: “You are my PR desk editor. Based on these 5 headlines and these Trends notes, write 3 quotable lines (25 words max each), 1 clean stat (with likely source), and a subject line for [audience]. Headlines: [paste]. Trends: [paste].”
30 Minutes (Add Proof)
Pull a small public dataset and build a quick ranking.
Prompt: “Turn this dataset into a 6-bullet media note: 1 newsy headline, top 5 ranking, 1 methodology line (source/date), 1 skeptical caveat, and a spokesperson quote. Data: [paste table or bullet stats].”
1 Day (Reusable Asset)
Ship a mini index or micro-survey with a simple chart and local angles.
Prompt: “Create a press note from this survey (n=500). Output: headline, 3 key stats, 1 chart description, 2 local hooks (UK/US/DE variants), and 2 reporter-specific subject lines (business vs lifestyle). Survey: [paste].”
The key with all of these: you feed the AI your notes, your data, your angle. Never let it generate the idea. It structures what you already know — that's where it's useful.
25 Idea Starters (Pick Five This Month)
These are concrete. Print them out, stick them next to your screen, and force yourself to ship five this month.
- Compare “Rising” searches vs last year; explain the why.
- Turn a creator trend into a marketer's explainer with risk/ROI notes.
- Build a “Top 20 cities for X” index from Eurostat or open data; localize for each market.
- Scan “best campaigns of 2026” to borrow structures (not headlines).
- Run a 5-question micro-survey; one chart carries the story.
- Mine support tickets for the 10 most asked questions; quantify the change over time.
- Scrape product reviews to find “unknown deal-breakers”; make a PSA.
- Use Muck Rack Trends to see who covers your topic; pitch 5 humans, not 50.
- Analyze competitor link-earning formats and recreate with new data.
- Pair two datasets (prices + commute; flights + concerts) for a fresh cut.
- Explain a new platform feature and its practical impact in 3 lines.
- FOI/FOIA one local metric everyone assumes but nobody has published.
- Turn anonymized usage data into “how people really do X.”
- Re-angle a stale blog post as a ranked list with a method box.
- Founder POV on a polarizing ad — add one market stat for ballast.
- Summarize 5 takeaways from a niche conference YouTube playlist.
- Ask three respected creators one very specific question; compile the answers.
- Refresh a perennial topic with “what changed since last year.”
- Map “X myths vs facts” with one source link per fact.
- Time a ranking to a tentpole event (tax season, back-to-school).
- Do a “basket of prices” audit across 5 stores; document method.
- Package a regulation update into a 3-bullet explainer.
- Try a tasteful stunt that demonstrates product value, not just shock.
- End every piece with “what we'll watch next month” to seed your next reactive hit.
- Create a “toolbox” post (spreadsheets, checklists) and reuse it in pitches.
The Minimal Stack
You don't need 15 subscriptions. You need four functions covered:
Sense: Google Trends, BuzzSumo Trending, TikTok Creative Center. Where you spot signals.
Match: Muck Rack Trends (who's writing), Qwoted / ResponseSource / SourceBottle (live journalist requests). Where you find demand.
Prove: Eurostat, ONS, local data portals + a spreadsheet. One simple chart per story. Where you build credibility.
AI co-pilot: ChatGPT, Claude, whatever you prefer. Feed it your notes and data. Never let it run unsupervised.
The Only Rule That Matters
Three quotable lines (25 words max each). One clean stat with a source. Spokesperson availability.
If your pitch saves an editor time, it lands. If it creates work for them, it doesn't. Everything else is detail.
Related Reading
Keep Reading
Ready to earn links instead of buying them?
Get 8–14 editorial placements in top-tier publications. No contracts. No risk. Just results.
$3,000 per campaign · 8–14 links guaranteed · Zero penalty risk
Salva Jovells is the founder of Presslei, a reactive PR agency based in Zurich. After 12 years of ecommerce SEO and a decade of link buying for Hockerty and Sumissura, he reverse-engineered 5,272 earned media placements and built a PR agency around what actually works.
Keep Reading
- What 5,272 media placements taught us about what journalists want
- How Hockerty earned 2,296 placements through reactive PR
“The best PR ideas don’t come from brainstorming sessions. They come from systematically monitoring what journalists are already writing about and finding the gap where your expertise adds something they can’t get elsewhere.”
— Salva Jovells, Presslei
DO
- Use systematic methods (news monitoring, data mining, trend analysis) to find ideas
- Validate idea newsworthiness before investing in campaign production
- Connect every PR idea to a specific business goal before proceeding
- Keep a running list of ideas categorized by format and timing
- Test ideas with one or two journalists before full campaign launch
DON’T
- Rely on brainstorming sessions as your primary idea generation method
- Pursue ideas just because they’re creative without checking journalist demand
- Generate ideas in isolation from your available data and expertise
- Stockpile ideas without acting on them — timeliness matters
- Discard ideas that don’t work immediately — timing may make them relevant later
Frequently Asked Questions
How often should you look for new PR ideas?
Build a weekly 30-minute review into your Monday morning — scan news headlines, Google Trends, and your platform data. Teams with consistent output don’t wait for inspiration; they have a structured process for reviewing sources on a cadence.
Which method produces results fastest?
Google Trends. Search your topic, check “Related queries” for breakout tags — any breakout is a trend journalists haven’t fully covered yet. Cross-reference with your data, find the surprising angle, and you have a concept in under an hour.
Can AI tools help find PR ideas?
AI is useful for brainstorming and stress-testing whether an idea is newsworthy. Where it falls short is identifying truly original angles. Use AI to expand on ideas you’ve generated through trend-monitoring, not as a replacement for monitoring itself.
About the Author
Salvador Jovells
Founder of Presslei. 12+ years in ecommerce SEO across international markets. After a decade of link buying for Hockerty and Sumissura, I reverse-engineered 5,272 earned media placements and founded a reactive PR agency that builds authority through data-driven stories journalists actually want to publish. Based in Zurich.


