Welcome to the FIMS Lab newsletter

In this edition:

• FIMS Lab highlight: Your Black Friday promotions

• One Marketing Thing: Nobody wants to be told what to do

• 🤖 One AI Thing: Should AI companies compensate publishers

Total read time: ☕️ 3 minutes

FIMS Lab highlights:

Black Friday is NEXT WEEK! Amalie and several of your FIMS colleagues shared their promotional plans for the big weekend. (Passcode !@#^D2hV)

Here are five ideas from Amalie’s deck you can implement immediately:

• Highlight the benefits of a digital subscription - immediate access, app access, be the first to know

• Show how you made a difference in the community

• Personalize your staff (great example in the deck)

• Give thanks and appeal for support - Here are loads of examples from last year’s Fund for Journalism campaigns

Our friends in Maine said they tried the second and third suggestions last year but didn’t see a noticeable increase in subscriptions. (Timecode 13:15)

“We put a lot of effort in collecting stories to inspire giving and putting faces to each benefit, like our CEO and her love of puzzles,” they said, “and we didn’t have a very good year.”

This year, they’re going back to straight pricing. Their plan includes a deep focus on gift subscriptions, which do much better than sales.

They also plan to experiment with a flash sale offering targeted discounting.

Over at The Gazette, Liz Kennedy said their Black Friday push will begin in earnest as soon as their current sports offer ends. (Timecode 19:15)

“Parents who sign up during this period (of sports coverage) tend not to churn EVER,” she said.

“Parents who sign up during this period [of sports coverage] tend not to churn EVER,” she said.

Their promotion plan includes:

• Tease for Black Friday; November is the only month they run big sales.

• Offer a similar discount to the people who signed up last year and might churn. (They’re offering a 40 percent discount rather than 50 percent)

• Their first trial subscription test, timed for 2 weeks, which expires just before Iowa basketball starts.

• Segmented push alerts.

• Sale graphics across O&O.

• All staff with podcasts or newsletters will be encouraged to do personal asks to promote cyber sales.

• And more! 😅

Also hear examples from Anchorage (who worked with a GNI Lab coach), Wick, Shaw (A/B testing with last year’s subscribers), Sonoma and Georges Media on the recording. (Passcode above)

(Don’t miss the discussion about how a Reddit thread influenced decisions about the use of gift cards for subscriptions.)

Finally, here are additional considerations from my copywriting hero, Neville Medhora. He goes deep into his strategy in this Google Doc. (If you can’t access, lmk)

Help shape phase two of the FIMS Lab 💕 (THANK YOU to those who have submitted responses!)

One marketing thing: Don’t tell me what to do

During this time of year most people are ready to buy and don’t need to be convinced. This could tempt us into being a little aggressive (Buy Now! Don’t Wait!).

But people want to be the decision makers in their lives. They need to feel in control.

Studies show people are more likely to act when they feel a sense of autonomy. Author Bri Williams explains in the Nudge podcast how to avoid reactance, why offering more than one choice is essential and how to position your offer as a shared idea.

Listening and gaining their trust is the first step. In an example, when a restaurant “suggested” a tip on receipts, customers not only didn’t tip, they reported feeling negative about their dining experience and fewer of them returned.

When the restaurant explained the suggested tip amount was printed to save customers the effort of having to calculate it themselves, and was optional, customers left more tips and reported feeling happier.

How does this apply to your offers? The first step is always listening to their needs and reflecting back a responsive offer.

The 20-minute podcast episode is full of useful tips like this one! 

One AI thing: Should AI companies compensate news publishers for the use of their content?

Last week I attended a small gathering of academics, news professionals and tech founders interested in generative AI and the risks and opportunities for local news businesses.

It was sponsored by Medill, which will publish the recommendations early next year. While we spent two days covering a variety of topics, one particularly spicy discussion involved whether or not local news companies should be compensated for the use of their content.

I plugged my notes into ChatGPT and it produced the following summary, which has been edited for clarity and length. (FWIW I find the first two arguments against compensation particularly weak.)

The Case for Compensation 💰

1. Intellectual Property Rights: When AI companies use news content to train their models, they are leveraging the intellectual efforts and resources of news organizations. Just as any other form of intellectual property, this content deserves protection and fair use consideration.

2. Sustainability of Journalism: The news industry, particularly smaller publishers, often operates on thin margins. Paying a licensing fee could provide a vital revenue stream, supporting the diversity and health of the journalistic ecosystem.

3. Precedent in Other Industries: In industries like music and film, content creators are compensated when their work is used. Applying a similar model to news content could be seen as an extension of this established principle of fair compensation for content creators.

The Argument Against Compensation

1. Content in the Public Domain: Once content is publicly available, it should be free to use, especially if the use is transformative and not for direct commercial gain, i.e. search engines.

2. Benefit to Publishers: If AI tools drive traffic to news sites the increased visibility and traffic could be considered a form of compensation.

3. Technical and Practical Challenges: Determining the value of each piece of content, managing payments, and deciding which content is eligible for compensation would require a complex, potentially contentious framework.

Suggested solutions

We did not settle on any particular path, instead agreeing on the need for ongoing experimentation and multiple approaches.

1. Regulatory intervention: This could involve legislation that mandates compensation for content use or sets clear guidelines on what constitutes fair use in the context of AI training.

2. Collaborative Content Creation: AI companies could collaborate with news organizations in creating original content specifically for AI training.

3. Utilizing Open Source and Public Domain Content: Focusing on content that is explicitly in the public domain or released under open-source licenses.

4. Collective Bargaining Power: This collective approach could increase their bargaining power against AI companies, allowing them to negotiate terms more effectively.

5. Public Awareness and Advocacy Campaigns: Raising public awareness about the value of journalistic content and the challenges faced by publishers in the digital age could create pressure on AI companies to adopt more publisher-friendly practices.

6. International Collaboration and Standards: Given the global nature of the internet and AI, international collaboration to establish standards and best practices for content use and compensation could be beneficial.

Social media companies did not rely on news content for monetization, but AI large language models theoretically not only rely on news content but also could suffer economic harm if published news content is withheld or withdrawn.

More to come on this topic and others in generative AI!

Have thoughts about this topic? Send me an email!

Upcoming:

• TODAY at 2 p.m. Eastern time, Jeff Elgie from Village Media talks about hyperlocal community

• No newsletter next week! 🦃

• Nov. 28 FIMS group call: Jay reveals and discusses the results of the benefits benchmarking survey