TL;DR
Knowing what makes PR go viral is one thing. Building a campaign that actually hits those marks is another.
This guide gives you the specific tools, prompts and processes we use to go from blank page to pitchable campaign.
The goal of this phase isn’t to generate ideas. It’s to find angles that are already working, validate them with data, then build assets that make a journalist’s job as easy as possible.
Phase 1: Research
Most PR campaigns fail before a single pitch is sent. The problem isn’t the writing or the outreach – it’s that the angle was never based on what journalists are actually covering. Research fixes that.
The goal here is simple: find angles that are already working, then build something better or find the gap they’ve missed.
Tool 1: UK publication research
Before you generate a single idea, spend time inside the publications you want to land in. You’re not looking for inspiration – you’re looking for proof. Proof of what format works, what data types get used and what’s been done to death.
How to do it
1. Pick 3-5 target publications in your sector
Match the publication to your client’s industry – don’t just default to the nationals. Think about where your audience actually reads.
- Home/DIY – Daily Mirror, The Sun, Telegraph Property
- Finance – MoneySavingExpert, Guardian Money, Which?
- Health – Daily Mail Health, NHS website, regional Reach titles
2. Search each site for your topic over the last 6-12 months
Use Google’s site operator rather than the publication’s own search. It gives cleaner results and easier filtering.
Type this into Google: site:mirror.co.uk “home improvement” – swap the topic and domain to match your target publication.
3. Collect the top 15-20 headlines
Copy them into a doc or spreadsheet. Don’t analyse yet – just gather. You want enough volume to spot genuine patterns rather than coincidences.
4. Look for patterns
Once you have your headlines, ask yourself:
- What angles get repeated across multiple publications?
- What data types appear most often – cost breakdowns, rankings, percentages?
- Which experts or voices do they quote regularly?
- What’s conspicuously missing – the angle nobody has taken yet?
💡
Pro tip: The gap is often more valuable than the trend. If every article puts a cost figure on one behaviour, be the first to do it for a different one.
Tool 2: Network journalist mapping
Not all publications are equal. Some are standalone titles. Others are part of large networks – Reach PLC, Newsquest, National World – that syndicate content across dozens of sister titles. One placement at a network publication can become 20+ links overnight.
This tool is about identifying those high-leverage journalists before you pitch, not after.
The major UK publication networks
- Reach PLC – Mirror, Express, Manchester Evening News, Birmingham Mail and 100+ regional titles
- Newsquest – The National, regional dailies across Scotland and England
- National World – The Scotsman, Yorkshire Post and regional titles
How to map the journalists
1. Find journalists covering your topic within these networks
Search a network publication for your topic, then click through to the journalist’s profile page. Most network titles list a journalist’s full article archive – this is your research goldmine.
2. Analyse their last 20-30 articles
For each journalist you identify, look at their recent output and note:
- What topics do they cover most consistently?
- What data or expert comment do they rely on?
- Which of their stories have been syndicated across the network?
3. Build a priority target list
Rank journalists by syndication potential. A journalist whose work regularly runs across 15+ sister titles is worth 10x more effort than one writing only for a single local paper. Match their editorial preferences to your upcoming campaigns before you build anything.
💡
Key Insight: Targeting a journalist at a network publication means your story doesn’t just land once – it spreads. One well-placed pitch can generate 20+ links from a single conversation.
Tool 3: Google Trends
Publication research tells you what’s been covered. Google Trends tells you what people are actively searching for right now – which is the other half of the timing equation.
You’re not using Trends to generate ideas from scratch. You’re using it to validate timing, find the best moment to pitch and uncover the regional angles that make your story more publishable.
What to look for
- Seasonal spikes – when does interest in your topic peak each year?
- Rising queries – what related questions are trending upward?
- Regional interest – which cities or regions show the highest search volume?
- Related topics – what else are people searching for alongside your main term?
In the example above, DIY search interest peaks sharply in October – not spring, as most people assume. That single insight changes when you pitch a home improvement campaign and what angle you lead with.
Export the CSV from Google Trends and feed it into an AI prompt (see Tool 3 below). This turns raw index data into quotable insights and pitchable headline angles in minutes.
AI-powered analysis
Once you have your publication headlines and Trends data, AI is the fastest way to spot patterns, identify gaps and generate angles. But the quality of what you get back is entirely determined by the quality of what you put in.
Here are the two prompts we use at Bulldog. Copy them directly – they’re designed to produce actionable output, not generic suggestions.
Prompt 1: analyse publication coverage
Use this after collecting your 15-20 headlines. Paste them directly into the prompt or upload the articles as a file.
I’ve collected the top 20 headlines/articles from [publication names]about [topic] in the last year.
Headlines: [paste headlines here] OR upload the articles
Identify:
1. Common angles that get coverage
2. What data types appear most often
3. What’s being covered repeatedly vs. what’s missing
4. Suggest 3 PR angles I could pitch based on these patterns
Focus on gaps and sideways angles, not copying what’s already been done.
Prompt 2: analyse Google Trends data
Use this after exporting a CSV from Google Trends. Attach the file and run this prompt:
I’ve attached Google Trends data for “[search term]” in the UK.Turn it into PR insights for [target outlet, e.g. consumer lifestyle press].
Use only what is in the file. If something isn’t answerable from thisexport, say so and tell me which export would fill the gap.
Treat index values of 0 as suppressed low-volume weeks, not zero searches.Do not calculate percentages from a zero baseline.
Lead every finding with consumer-friendly language: percentages andmultipliers in plain English. Put the index in brackets as supportingevidence only, never as the headline number.
Deliver:
– The 3-5 strongest patterns, quantified with supporting index
– 4-6 quotable one-liners in the voice of the target outlet
– 3 headline angles, each labelled strong, defensible or speculative
– Additional Trends exports that would strengthen the story
Notice how the output labels each headline angle as strong, defensible or speculative. This is deliberate – it stops you pitching a speculative angle as if it were data-backed. Use it as a filter, not just a generator.
Phase 2: Creation
Research gives you the angle. Creation gives you the assets. The goal of this phase is to build something a journalist can use immediately – not a pitch they have to rewrite, but a story they can almost publish as-is.
There are three things to build:
➡️ A press release written for journalists
➡️ Data they can quote
➡️ A visual they can embed.
Writing the press release
The press release is not a company announcement. It’s a news story written in the third person, structured the way a journalist would write it. The moment it reads like marketing, it goes in the bin.
The structure that works
- Headline – the finding, not the company. ‘Manchester named UK’s best city for remote workers’, not ‘Company X releases new study’
- First paragraph – who, what, when, where, why. Lead with the stat or finding. Your brand comes second.
- Supporting paragraphs – more data, the methodology, context that makes the finding credible
- Quote – expert insight in plain English, not corporate language. Make it sound like a real person said it.
- Boilerplate – one short paragraph about who you are. Keep it factual, keep it brief.
The test: Read your headline back. If it could be a news headline on the Mirror or Express website, you’re on the right track. If it sounds like a press release, rewrite it.
Collecting your data
The data is what makes your story credible and quotable. Without it, you have an opinion. With it, you have a story. Here’s where to source it.
Original data (best)
Your own sources are always the most valuable because they’re exclusive. No other publication can match them.
- Customer or user data you already hold
- An industry survey commissioned specifically for the campaign
- Proprietary analysis of your own platform or product data
Aggregated data (also works)
Public data is free, widely available and often completely untapped. The trick is combining sources to produce an original analysis nobody else has done.
- Government datasets – ONS, Land Registry, Companies House
- FOI requests – underused and often produce genuinely exclusive findings
- Public APIs and open data portals
- Combining two public sources to produce a new metric
The example above combines attraction costs, free activities, public transport and parking data into a single weighted score. Each data point is public – the ranking is exclusive. That’s the aggregation approach in practice.
Making the data pitchable
Whatever your source, the data needs to be structured so it produces a clear, quotable finding. A few rules:
- Make it rankable – top 10 cities, best vs worst, fastest growing.
- Go in with a question – ‘which UK city is cheapest for X?’ – then answer it.
- Add the surprising finding – the result that makes a journalist stop scrolling.
- Keep the methodology simple enough to explain in one sentence.
Visualisation: making it easy to embed
Journalists are busy. If your data requires them to build a chart before they can use it, most won’t bother. The visual should be ready to embed the moment they open your email.
Choosing the right format
Different data tells a different story – and each story has a format that serves it best.
- Table – use for rankings and comparisons across multiple categories. Best when readers want to find their own city or industry.
- Bar chart – use for direct comparisons. ‘Which city is best/worst?’ lands immediately in bar form.
- Line chart – use for trends over time. ‘How has this changed over 5 years?’ is a line chart story.
- Map – use for regional data. If the story is about geography, a map earns its place.
- Infographic – use when the story has multiple stats that belong together. The big picture in one shareable image.
Notice that each of these visuals is self-explanatory. You don’t need to read the accompanying article to understand what they’re showing. That’s the standard to aim for.
What comes next?
Once you have your angle validated, your data collected, and your visual built, you’re ready to write the pitch and get it in front of the right journalists.
Related read: How to pitch and amplify your PR campaign
Start featuring in dream publications 🚀