Ana Staples

There's so much talk about using AI to generate content, especially now that tools like ChatGPT make it look so easy. But is that coming at a cost to the reader experience? I'm a huge fan of (properly using) AI. It's been great at summarizing meetings. I use it sometimes to generate headline ideas or to generate a first draft of social media copy. I'm legitimately amazed at how fast this technology is developing. But as someone who has been "generating content" for newspapers, websites, blogs and social media since MySpace was the dominant social platform, I see far too many companies rushing to adopt AI for content generation without question. It seems so easy — because it is. Upload some reputable, trustworthy data sources (at least, please tell me you're doing this), click a couple buttons and you've got a story ready to go! Unfortunately, going this route ignores a lot of success factors. - Is it going to rank on Google? (No — the algorithm wants human experience and expertise that AI can't magically generate.) - Is it something that will add value for your customers? (People don't really want to read awkward AI copy, and will just bounce.) - Is it going to read well? (Unless you're also a robot ... no.) Instead, use AI to *assist* human writers. It's great at summarizing interviews, gathering key data points and helping you brainstorm. It can generate a first draft that gives you a structure to work with, if you're stuck. AI is a wonderful tool as an assistant, but it's not a long-term content strategy. While the technology changes over time, one thing doesn't — people want to trust that the article they're reading will help them out. Seeing how often people add "Reddit" on the end of their Google searches just confirms this. If you really want to stand out among your competitors and get long-lasting results, don't use AI as a content generator. Use it to help your human writers and editors do their job better.

4

A great test to see if your copy is too formal is the "comma Mary” test. The test comes from Bruce DeSilva, a thriller author and former journalist. The test: put ", Mary" at the end of a statement. Then read it aloud. If it doesn't sound right—namely, too formal—turn the word-choice dial toward less specialized, more conversational until it feels right. Example: “Undergoing regular cancer screenings is the expert-recommended approach to discovering cancer early enough to optimize treatment options and improve health outcomes, Mary." or "Regular cancer screenings help find cancer early, which is why experts recommend them. Treatment works best then and can help patients live longer, Mary." Which one are you most likely to say to a person? Now, it's no silver bullet; you'll want to go deeper and look at your writing decisions—sentence length along with word choice and order. But the "comma Mary" test is a great tool to judge your copy's readability. And to make changes toward making it more human—if that's what it needs, Mary. Alexandra Howson and I talked shop about voice last year. #writing #storytelling #medicalwriting

81

29 Comments

A great test to see if your copy is too formal is the "comma Mary” test. The test comes from Bruce DeSilva, a thriller author and former journalist. The test: put ", Mary" at the end of a statement. Then read it aloud. If it doesn't sound right—namely, too formal—turn the word-choice dial toward less specialized, more conversational until it feels right. Example: “Undergoing regular cancer screenings is the expert-recommended approach to discovering cancer early enough to optimize treatment options and improve health outcomes, Mary." or "Regular cancer screenings help find cancer early, which is why experts recommend them. Treatment works best then and can help patients live longer, Mary." Which one are you most likely to say to a person? Now, it's no silver bullet; you'll want to go deeper and look at your writing decisions—sentence length along with word choice and order. But the "comma Mary" test is a great tool to judge your copy's readability. And to make changes toward making it more human—if that's what it needs, Mary. My friend Alexandra and I talked shop about voice last year. #writing #storytelling #healthcare

20

6 Comments

You may groan when you read this, buuuuut trust: You need to write your holiday blogs now, not in November. If you have (or work for) a company that sells anything, especially more things around the holidays, get those blogs done now + pinned on Pinterest. You’ll thank me when that search volume ratchets up in Q3-4. It’s also exactly what I did when I wrote + pinned this blog: https://lnkd.in/gpmSRCGA. #contentwriting #holidaycontent #companyblog #holidayblog #organicsearch

4

Take a big step forward in embracing accessibility and inclusion best practices by making your PDFs accessible using this actionable 8-step guide!

4

Are you diversifying your "visibility" streams? Three takes: 1) Social media is no longer an organic gift of reach 2) The media are searching hourly for experts like you 3) You don't have to wait for "when you're more significant for PR Like this from Qwoted, which arrived just now in my email inbox. Where are my NY-based career experts; this one is for YOU. The amount of time we post/scroll/comment on social media means we can effectively source the journalist and media sights for the right fit for their stories. In addition to proactive story pitching, I teach my clients to create profiles and monitor journo requests from > Qwoted > Featured. > SourceofSources (formerly HARO) > SourceBottle (great for AUS-based businesses) Among other places to monitor the request that can deliver value to the media. I'll be detailing more of this soon in my just-launched newsletter. The Standout Women Newsletter just launched (FINALLY) yesterday. It's Your Moment, Michelle P.S. Want to submit for this request? Sign up for a free account at Qwoted and ensure your media bio and other assets are well dialed into your "expertise area."

22

21 Comments

Every paragraph in this article about headline writing is terrific. Not a word more or less than needs to be said. It's a testament to clear, actionable writing. So read it for the sake of understanding what a great article looks like, but also read it for the takeaways about headline writing for journalists...because the insights are also applicable to the corporate journos[1] among us. In short: Recent research suggests that keeping headlines straightforward can significantly boost reader interaction and comprehension. 1. Simplicity Wins: Headlines with simple, common words and a narrative style receive more clicks. For example, using "job" instead of "occupation" can increase engagement. 2. Reader vs. Journalist Preferences: Typical news readers prefer simple headlines, while journalists do not. (This is often *my* problem.) 3. Engagement and Recall: Simple headlines attract more clicks and improve reader recall and engagement with the article's content. https://lnkd.in/gwMKEsic

28

1 Comment

💡How are newsrooms using AI?💡  🌟 As AI continues to evolve, marketers and PR pros need to know how it affects journalism. Understanding how AI software influences story discovery and content creation enables teams to refine their pitching and press release strategies. 📰🤖 🔍 Consider these recent examples of AI projects supported by AP's Local News AI initiative and the John S. and James L. Knight Foundation: 🎙️ Michigan Radio, a University of Michigan NPR affiliate, uses an innovative tool that scrapes city council meetings from video sharing services, transcribes them, and generates tailored alerts. The result? Better visibility into public meetings and proactive notifications aligned with journalists’ beats and interests. 📺 Television station WFMZ-TV in Allentown, Pa., uses AI software that sorts and prioritizes news releases and story pitches from emails and other outside sources. And then, it enters the information into a content management system. 🤔 Thoughts? 👇 🔗 Learn more about the AI projects mentioned at https://lnkd.in/ep9H9QCM J.P. Escobar Jason Fink R. Kurt Osenlund Isha Kamani Hollis Daniels Melanie Shaw Chris Hayes #ArtificialIntelligence #Journalism #MarketingStrategy #PR #AIinJournalism #Innovation #FutureOfNews #TechAdvancements #Collaboration

11

Friday morning editing tips: To tighten your writing, trim your... Prepositions "Followed behind" —> "followed" "Stood up" —> "stood" "Knelt down" —> "knelt" Implied nouns "Nodded their head" —> "nodded" "Blinked their eyes" —> "blinked" Redundant adverbs "Scowled irritably" —> "scowled" "Grinned joyfully" —> "grinned" #Writing #Editing

13

The court of #ContentMarketing appeal is nearly in session. As a judge for Content Marketing Institute's Marketing Awards, I can't be bribed, but I can be dazzled by great work. If you've got successful strategies and exciting campaigns to share, enter by May 19, 2024.

7

2 Comments

Digital News Report, the Reuters Institute for the Study of Journalism flagship annual study, is out this week. Big-picture findings won’t come as a surprise, but it’s still a must-read. For The Fix Media, I picked a subjective list of three interesting insights from the report: 📍Video-first platforms are rising, and traditional media are struggling to compete with individual news creators. YouTube is a key platform for news, and TikTok has overtaken Twitter in news consumption for the first time this year. Unlike on Facebook & Twitter, traditional journalists are lagging behind individual news creators. Interesting examples from the report – Hugo Décrypte in France (who, across all platforms, gets more mentions than Le Monde or BFMTV) and TLDR News in the UK. 📍Paying for news isn’t becoming more widespread – and even existing numbers are skewed by heavy discounts. Last year 17% of people across 20 richer countries paid for online news. This year it’s still exactly 17%. One interesting addition in the 2024 report is the look at how heavy discounting might distort the numbers. “[W]ell over half of those in the US who are paying for digital news report paying less than the median cost of a main subscription ($16), often much less”. In Poland, over three quarters of respondents pay less than full price for their subscription. 📍Elections have had little effect on interest in news so far, but we’ll see full effect in 2025 Interest in news has fallen across much of Europe and other countries. It increased in the US but only slightly. But the surveys for this report were conducted at the beginning of the year. It will be interesting to watch the impact of elections throughout 2024, especially in the United States.

11

Effective corporate apologies don't sound like they were written by a copywriter, or written out of obligation rather than sincerity, or watered down via multiple approvers. Here’s a start: 1. Say you're sorry. 2. For what you did. 3. Show you understand why it was bad. 4. Only explain if you need to; don't make excuses. 5. Say why it won't happen again. 6. Offer to make up for it.

109

23 Comments

Considering what's happening in the markets, it's unsurprising that so many friends in the media are engaging financial planners to get some thoughts and opinions for their stories. It's so gratifying that so many of them are using FPA MediaSource to get that perspective from FPA members! In just the past 24 hours, the system has sent 22 queries to #CFPpro members of FPA, including The Wall Street Journal, CBS News, Barron's, Investopedia, InvestmentNews, ThinkAdvisor, CNBC, MarketWatch, Fortune, and many others. If you're not taking part, you're missing out! https://lnkd.in/eMJeNS-F Financial Planning Association (FPA) #financialplanning #media

28

I keep seeing PR agencies make the mistake of hiring prompt engineers with no experience actually doing PR work. Whyyy? 1) Prompt engineers are no longer relevant - LLMs can write their own prompts better than any human can. 2) Unless you deeply understand the use-case, what is the point of someone who knows how to prompt AI to get a broad range of results. It's AI + a narrow focus that delivers value not AI + general use case humans. 3) Just hire smart, capable, problem solvers that are also PR consultants.

12

1 Comment

A client's spokesperson recently used ChatGPT to help them write responses to a reporter's questions. Here's why that's a no-go (and why I asked them to re-write before I sent it!). ➡ Reporters are REALLY good at Googling and doing online research. They *literally* do this professionally. They don't really need generic quotes from you about information that's already publicly available online (and if it's in a public LLM, likely years old). Plus, in many cases, using AI-generated content violates their editorial or ethical guidelines. If you're stumped at how to respond to a reporter's questions, here are some great ideas for what you might bring to the table that AI cannot: 🌟 Anecdotes/stories that reflect your personal experience with the topic/subject of the interview. 🌟 Unique insights about what's happening behind the scenes - what don't people know or understand about how your industry works? What aren't people paying attention to right now that's going to be huge? 🌟 Stories or examples about your customers - how does this topic impact them? What have you heard from them? 🌟 Advice on who else they should connect with or other potential sources of information that you are aware of as an industry insider. Offering value that goes beyond promoting yourself is a great way to build a strong relationship! I don't believe this spokesperson intended to "cheat" on their media interview - they simply thought they were being more efficient. But just to be safe... I now always use a tool that checks for AI-generated content before I send quotes to a reporter 😁 #PR #journalism #publicrelations #mediarelations

29

6 Comments

🤔 What makes a PR pro truly *great*? Here’s my take based on 20+ years in tech PR: 1. Stellar writing. ✍️ Good writers are plentiful. GREAT writers are unicorns. In PR, great writers are also *versatile* writers. They effortlessly toggle between crafting killer press releases to pithy social media posts to SWOT analysis to byline articles. Great writing is my #1 requirement for PR pros. Even more so in the age of AI. Full stop. 🫳 🎤 2. Ability to process info quickly. 🧠 Great PR pros digest large amounts of info and connect the dots to current trends and storytelling opps. Lots of PR folks have knowledge. But *knowledge isn’t wisdom*, and you need wisdom to find creative pathways that others can’t see. That’s how you guide clients to higher ground. 3. Charm. 😁 Yes, charm! Look, PR is as much about sales as it is writing and strategy. The old adage that “people buy from people they like” is absolutely true in public relations. Charming = interesting. Charming = energetic. Charming = persuasive. Can you be a solid PR pro without being charming? Of course. I’m just telling you what I think. 4. Aptitude for failure. 😬 Say what? PR people who aren’t trying new things aren’t going to move the needle. Trying new things *necessarily means failing* sometimes. The best PR people aren’t afraid to take calculated risks, and know how to fail quickly and pivot to better ideas. PR people who balk at doing things differently might be competent, but they rarely *inspire* clients. 5. A bias for empathy 🥺 Great PR people have respect for their audiences. They respect journalists. They can *read the freaking room* and act accordingly. They feel things, ya know? This list might not work for you. But for Crackle PR? This is the way ✊ #publicrelations

211

23 Comments

This month on the Web Writing Advice blog I'm speaking to my friends who work in public relations, and to the experts/sources who help make my stories interesting and insightful! As a writer working in content marketing and journalism, I WILL go to your website to get more information. What information? That's what I'm chatting about in my latest article. Take a peek here! https://lnkd.in/gcc_TMd6 Enjoy! ~ Angela #WebWritingAdvice #Sourcing #PublicRelations #Experts #Writing

5

2 Comments

Before you craft your next AI narrative or piece of content, listen to the recent episode of #SaaS Half Full with Jake Doll. In this episode, Jake tackles which AI storylines are trending, and which ones are complete duds. And yes, both Jake and I say "please" and "thank you" when entering genAI prompts...how about you? #Saasmarketing #PRagency #podcast

6

2 Comments

Is LinkedIn Premium just a giant AI-data scraping exercise that I’m paying for? Its clutter of inept features seem to serve no other purpose than to gather data. They’re of absolutely no use, other than as a reminder of how useless generic AI is. And, possibly, how we’re being used to improve it. 1 Useless ‘top applicant’ job recommendations Here are three jobs LinkedIn Premium thinks I’m going to be a top applicant for. - Bulgarian-speaking copywriter.  - Freelance copy writer – musical instrument industry.  - And finally, as if it’s doing this to spite me and my 25 years in writing and journalism, Tech writer – intern. Oddly, I didn’t get recommended for the Japanese-speaking copywriter at the place wanting the Bulgarian speaker. 2 Those suggestions at the bottom of people’s posts which initially look like they might be helpful, but then you quickly realise it’s just an AI-driven search asking questions no-one needs the answers to. Three I’ve just seen: - 'Top news from Hay-on-Wye?' (AI’s answer: Hay-on-Wye is currently in the news for its literary significance and unique position on the border between England and Wales.) - 'What impact does gratitude have on professional settings?' (AI’s answer: It fosters a positive work environment, leading to higher employee engagement.) - 'How can a cocktail shaker be useful?' (AI’s answer: a cocktail shaker is used to make drinks.) 3 Rewriting everything with AI Seriously, it’s not good enough. Not even for people who aren’t confident writing. In fact, it’s doing more harm than good. If you trust it to ‘tailor your CV to this job’ or ‘rewrite this message with AI,’ you’re damaging your chances of getting the job or getting your message taken seriously. It makes mistakes, uses cloying, predcitable language and it will not represent you properly. I’d rather read writing with mistakes in it by a human than from the grammatically correct, dead hand of AI. 

6

1 Comment