Almost Timely News: Recipes vs. Principles in Generative AI

(2024-03-03)

Mar 03, 2024

Almost Timely News: Recipes vs. Principles in Generative AI (2024-03-03) :: View in Browser

👉 Missed my last Intro to Generative AI webinar? Catch the next one for free on March 19. Register here!

Content Authenticity Statement

95% of this week's newsletter was generated by me, the human. You'll see some outputs from Gemini in the opening section. Learn why this kind of disclosure is now legally required for anyone doing business in any capacity with the EU.

Watch This Newsletter On YouTube 📺

Click here for the video 📺 version of this newsletter on YouTube »

Click here for an MP3 audio 🎧 only version »

What's On My Mind: Recipes vs. Principles in Generative AI

Last week, we talked over principles-based prompt engineering.

And you were not impressed. In fact, last week's newsletter scored as one of the lowest issues in recent times (each issue has a one-click survey below the main article). And I have a hypothesis as to why. There's a big difference between recipe and concept, between principles-based prompt engineering and "here's a template, just copy and use this".

To build long-term success, you absolutely need to understand concepts and apply them. You start with the basics, you learn variations, and then you transcend the basics, a concept from the Japanese martial arts known as shu-ha-ri - learn the basics, vary the basics, transcend the basics. It's a process as old as professions themselves.

But that's for the long-term, for when you're trying to master a discipline over a period of years, perhaps even decades. When I go to the dojo on Saturday mornings, it's an expression of this principle in action.

The reality is, that's not most people's intent with generative AI, to have it be a discipline that you master over years. Why? Well, based on conversations I've heard in Analytics for Marketers and other forums, you're overwhelmed. Overwhelmed by generative AI, but overwhelmed and overworked in general. You're running without a full team, perhaps even a skeleton crew.

And that means your brain might not be receptive to investing a lot of time, the way you might study an art form. The analogy I often use is from cooking (huge surprise) is the difference between learning the principles of cooking versus following a recipe.

For example, a cooking principle is to always salt your tomatoes regardless of application. Tomatoes contain glutamic acid, which when combined with salt, form a natural kind of MSG, making them taste much better. That's the principle. Contrast that with a recipe which simply tells you to put salt on the tomatoes before serving in, say, a caprese salad. You don't know why you're doing it, but you do it if you're following the recipe and the outcome is still good.

The difference between principle and recipe is that the outcome for the specific recipe is the same whether you know the principle or not, but if you made another dish with tomatoes that had a different recipe, and you didn't understand the principle, then that recipe might not turn out as well if you omitted the salt.

I've been thinking quite a lot about this in the context of generative AI lately. There's no shortage of people hawking "TOP 50 POWER AWESOME CHATGPT PROMPTS" on LinkedIn and other places, and I've dug into some of those. They're essentially cookbooks with recipes, and those recipes are generally okay. (I haven't run into any that I was blown away by) And yet people LOVE these sorts of recipe collections.

Why? Because as much as the principles matter, sometimes you just need to get dinner on the table in 30 minutes or less. You don't care about the principle. You care about getting dinner on the table. At the end of the day, you're tired and you don't want to think too hard. You just want some directions to follow that aren't too hard.

And that's the generative AI equivalent of a canned prompt, a prompt you copy, tweak a little with your specifics, and then paste. You follow the instructions, as surely as you do on a box of pre-made cake mix, and you end up with a satisfactory result. Is it going to be the best result possible? No, probably not. Is it going to be good enough? Yes, probably.

Where you run into challenges is when you have something that doesn't fit an existing recipe. That's when principles come in handy. Let's take a look at this prompt situation suggested by my friend and colleague Ashley Faus on LinkedIn:

We have tiers for our product launches, ranging from a Tier 1 launch with all the activities (press, keynote mention, blog post, demo series, announcement email, product tour update, etc.) to Tier 4 (significantly less activities). It seems like there should be a combination of AI + automation that could help a marketer generate a launch plan and all the associated tickets and/or pages required. But... would the prompt be, "I'm doing a Tier 1 launch about [Product Name]. Generate the launch plan and associated tickets for the Creative team, Blog team, and Web team."? Or would the prompt be, "I have a launch coming up that meets these criteria: [Customer Impact], [Company Impact], [Other criteria as needed]. Please choose the relevant launch Tier, and generate a launch plan."? And then maybe a separate prompt to generate the tickets and pages? Like if we have a template for an announcement blog for a Tier 1 launch, would it generate the page with the template, or generate a draft of the launch blog itself, or...? Again, I think this is a mix of internal/external AI capabilities, automation rules, & human collab, but IDK the mix

Ashley is correct. This is something that generative AI can handle, at least partially - but I can guarantee that as of right now, there is no recipe for it.

The first principle we invoke is whether or not it's a task generative AI is even capable of handling. Building a launch plan? Yes. What about creating tickets - Ashley works for the software giant Atlassian, and their Jira ticketing system is well-known. Can a generative AI system talk to Jira? Perhaps not directly - but Jira can ingest spreadsheets like CSV files. Can a generative AI system generate CSV files? Yes it can.

You see what we're doing here, right? This isn't a recipe, but we are laying the foundation to create a recipe. Something that my partner and CEO Katie Robbert talks about ALL the time is user stories, short, punchy descriptions of what you're trying to do that helps build requirements for the project. In this case, a user story - or several - is what you need to create a recipe for generative AI.

Once you know what it is you're trying to do, and you've ascertained whether or not generative AI is capable of doing it, then you can build the recipe - and like any recipe, once you have it written down, you can follow it over and over again.

So, how would we turn Ashley's idea into a recipe? Well, watch this week's video as I step through the process from start to finish.

Recipes are absolutely a good idea, especially if you want to scale the use of generative AI within your company. But many recipes may not exist, or may not be sufficient in their original form to fit your needs. Like the culinary world, there are a lot of cooks but relatively few chefs, so work to identify the chefs in your organization or your network as quickly as you can, then work with them to start building your cookbook.

How Was This Issue?

Rate this week's newsletter issue with a single click. Your feedback over time helps me figure out what content to create for you.

Here's The Unsubscribe

It took me a while to find a convenient way to link it up, but here's how to get to the unsubscribe.

If you don't see anything, here's the text link to copy and paste:

https://almosttimely.substack.com/action/disable_email

Share With a Friend or Colleague

If you enjoy this newsletter and want to share it with a friend/colleague, please do. Send this URL to your friend/colleague:

https://www.christopherspenn.com/newsletter

For enrolled subscribers on Substack, there are referral rewards if you refer 100, 200, or 300 other readers. Visit the Leaderboard here.

ICYMI: In Case You Missed it

Besides the new Generative AI for Marketers course I'm relentlessly flogging, I recommend the livestream I did this week with John and Katie as we went over Matomo Analytics, the free, open source alternative to Google Analytics.

Skill Up With Classes

These are just a few of the classes I have available over at the Trust Insights website that you can take.

Premium

Free

Advertisement: Generative AI Workshops & Courses

Imagine a world where your marketing strategies are supercharged by the most cutting-edge technology available – Generative AI. Generative AI has the potential to save you incredible amounts of time and money, and you have the opportunity to be at the forefront. Get up to speed on using generative AI in your business in a thoughtful way with Trust Insights' new offering, Generative AI for Marketers, which comes in two flavors, workshops and a course.

Workshops: Offer the Generative AI for Marketers half and full day workshops at your company. These hands-on sessions are packed with exercises, resources and practical tips that you can implement immediately.

👉 Click/tap here to book a workshop

Course: We’ve turned our most popular full-day workshop into a self-paced course. The Generative AI for Marketers online course is now available and just updated this week! Use discount code ALMOSTTIMELY for $50 off the course tuition.

👉 Click/tap here to pre-register for the course

If you work at a company or organization that wants to do bulk licensing, let me know!

Get Back to Work

Folks who post jobs in the free Analytics for Marketers Slack community may have those jobs shared here, too. If you're looking for work, check out these recent open positions, and check out the Slack group for the comprehensive list.

What I'm Reading: Your Stuff

Let's look at the most interesting content from around the web on topics you care about, some of which you might have even written.

Social Media Marketing

Media and Content

SEO, Google, and Paid Media

Advertisement: Business Cameos

If you're familiar with the Cameo system - where people hire well-known folks for short video clips - then you'll totally get Thinkers One. Created by my friend Mitch Joel, Thinkers One lets you connect with the biggest thinkers for short videos on topics you care about. I've got a whole slew of Thinkers One Cameo-style topics for video clips you can use at internal company meetings, events, or even just for yourself. Want me to tell your boss that you need to be paying attention to generative AI right now?

📺 Pop on by my Thinkers One page today and grab a video now.

Tools, Machine Learning, and AI

Analytics, Stats, and Data Science

All Things IBM

Dealer's Choice : Random Stuff

How to Stay in Touch

Let's make sure we're connected in the places it suits you best. Here's where you can find different content:

My blog - daily videos, blog posts, and podcast episodes
My YouTube channel - daily videos, conference talks, and all things video
My company, Trust Insights - marketing analytics help
My podcast, Marketing over Coffee - weekly episodes of what's worth noting in marketing
My second podcast, In-Ear Insights - the Trust Insights weekly podcast focused on data and analytics
On Threads - random personal stuff and chaos
On LinkedIn - daily videos and news
On Instagram - personal photos and travels
My free Slack discussion forum, Analytics for Marketers - open conversations about marketing and analytics

Advertisement: Ukraine 🇺🇦 Humanitarian Fund

The war to free Ukraine continues. If you'd like to support humanitarian efforts in Ukraine, the Ukrainian government has set up a special portal, United24, to help make contributing easy. The effort to free Ukraine from Russia's illegal invasion needs your ongoing support.

👉 Donate today to the Ukraine Humanitarian Relief Fund »

Events I'll Be At

Here's where I'm speaking and attending. Say hi if you're at an event also:

MarketingProfs AI Series, Virtual, March 2024
Society for Marketing Professional Services, Boston, April 2024
Lab Products Association, Nashville, April 2024
Society for Marketing Professional Services, Los Angeles, May 2024
Australian Food and Grocery Council, Melbourne, May 2024
MAICON, Cleveland, September 2024
MarketingProfs B2B Forum, Boston, October 2024

Events marked with a physical location may become virtual if conditions and safety warrant it.

If you're an event organizer, let me help your event shine. Visit my speaking page for more details.

Can't be at an event? Stop by my private Slack group instead, Analytics for Marketers.

Required Disclosures

Events with links have purchased sponsorships in this newsletter and as a result, I receive direct financial compensation for promoting them.

Advertisements in this newsletter have paid to be promoted, and as a result, I receive direct financial compensation for promoting them.

My company, Trust Insights, maintains business partnerships with companies including, but not limited to, IBM, Cisco Systems, Amazon, Talkwalker, MarketingProfs, MarketMuse, Agorapulse, Hubspot, Informa, Demandbase, The Marketing AI Institute, and others. While links shared from partners are not explicit endorsements, nor do they directly financially benefit Trust Insights, a commercial relationship exists for which Trust Insights may receive indirect financial benefit, and thus I may receive indirect financial benefit from them as well.

Thank You

Thanks for subscribing and reading this far. I appreciate it. As always, thank you for your support, your attention, and your kindness.

See you next week,

Christopher S. Penn

Dave

Mar 3, 2024

First of all, thank you for your generosity in sharing your insights. It's great stuff that I enjoy and appreciate regularly. With regard to Priciples versus Recipes, I feel like you may be giving yourself a pass. Recipes are easy to share because they contain the seed of highly consumable content, they clearly indicate their relevance. As you point out, principles are pretty darn valuable but they are also harder to learn and apply. When a teacher wades into "principles", it's up to the teacher to make clear the relevance and therefore help the student to engage their mind, their interest, and find the applicability to their own world.

I think the prior Priciples post had some great thoughts but it somehow didn't manage to make the relevance clear and that'd be my theory why it didn't land as well as most of your awesome posts. Without that clarification of relevance, content can seem dry and academic. If you take a look at the opening paragraph of that post, pretend you are a reader asking themselves, "Why should I read the second paragraph?" I personally confess I didn't make it to the fifth paragraph the first time I read it and only went back to reread it after reading today's post. It wasn't until I got to the LAST paragraph and got this gem, "Principles-based prompt engineering also tends to work better across models; that is, if you know what's under the hood and how it works, your prompts will be more easily portable from one model to another. Understand how generative AI works under the hood, and you'll make everything you do more effective."

There it is... an explanation that makes me want to read the post.

Expand full comment

1 reply by Christopher S. Penn

BILL STANKIEWICZZ

trending topic here at GRIT SAVANNAH Conference!! Best regards,

Bill Stankiewicz

Member of Câmara Internacional da Indústria de Transportes (CIT) at The International Transportation Industry Chamber

Professor FORKLIFT SAFETY TRAINING/OSHA/ VR Virtual Reality Training

SAVANNAH TECHNICAL COLLEGE

Savannah Supply Chain-CEO

1 more comment...