Almost Timely News: 🗞️ 3 Phases of Optimizing for AI (2025-11-23)

AI optimization isn't one thing

Almost Timely News: 🗞️ 3 Phases of Optimizing for AI (2025-11-23) :: View in Browser

The Big Plug

🚨 Watch my latest keynote, How to Successfully Apply AI in Financial Aid, from MASFAA 2025.

Content Authenticity Statement

100% of this week’s newsletter was generated by me, the human. You will see Gemini’s output about pizza and activation sequences in the first section. Learn why this kind of disclosure is a good idea and might be required for anyone doing business in any capacity with the EU in the near future.

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: 3 Phases of Optimizing for AI

This week’s newsletter is brought to you by a rant I had at MarketingProfs B2B Forum where tons of people were talking about the alphabet soup that is optimizing your business content for AI, and very few of those conversations were grounded in actual reality. It made me cranky.

So let’s take the rant I had several times and put it in words that are somewhat less profane and more helpful.

Part 1: Definitions and Prerequisites

Let’s start with some table setting. There’s a ton of alphabet soup out there:

• GEO: generative engine optimization, quite possibly the stupidest of them all because AI models are not engines in the same sense as search engines

• AIO: AI optimization, which is a little better

• GAIO: generative AI optimization, which is more specific but sounds like hair product

• AEO: Answer engine optimization because GEO isn’t obtuse enough (seriously, who comes up with these?)

• SEO: Search Everywhere Optimization, coined by Ashley Liddell of Deviate, which is a pretty good summary

All of these are intended to mean pretty much the same thing, which is optimizing your brand and content to be found by AI. Why so many variations? Blame marketers for that, as individual agencies and people try to establish something they can brand. I lean towards using AIO if I have to use something, but more often than not, I just call it AI optimization.

Next, one of the most important things to remember:

No one other than AI companies knows what actual people type into generative AI tools.

Louder for the back:

NO ONE OTHER THAN AI COMPANIES KNOWS WHAT ACTUAL PEOPLE TYPE INTO GENERATIVE AI TOOLS.

Any vendor who proudly proclaims they know what people type into ChatGPT or similar tools is either:

• Lying by varying degrees, OR

• Stealing information from AI companies

Neither is a good situation.

On top of that, even for companies attempting to approximate presence inside of a language model (often by running thousands of synthetic prompts/queries), you’re still not going to get an accurate read on what results people get. Here are a couple of simple examples as to why that’s largely a fool’s errand.

First, all AI tools are probabilistic. Type the same prompt into ChatGPT in different, new chats, and you’ll get conceptually identical answers but specifics WILL vary because that’s the nature of generative AI. Type in “What are the top SEO tools for optimizing your website?” and you will get wildly different results every time. Sometimes you get Semrush or AHREFS. Sometimes you get Google Search Console. Sometimes you get waffling.

Here is the exact same question with no previous context with memory turned off using the exact same model seconds apart in Chat GPT. Wildly different answers.

Can you factually claim that XYZ brand ranks top in ChatGPT? Absolutely not.

Not only are the tools probabilistic and unable to replicate an answer, but even a single token variation - because of how AI works - can create dramatically different answers. “What are the best SEO tools for optimizing your website” and “What are the top SEO tools for optimizing your website” will yield different answers, even though to us they should be identical. To us, they pretty much mean the same thing. To AI models, they are different. They will generate different results when you ask that question with the one word difference.

On top of that, AI tools now often have model switching built into them. Tools like Microsoft Copilot and ChatGPT will automatically select a model based on their estimated difficulty of the task, and depending on where in the conversation you ask for something, you will get different models answering your question. ChatGPT has 6 different models under the hood - GPT-5.1, 5.1-mini, 5.1-nano, 5.1-chat-latest, 5.1-codex-high, 5.1-codex-mini.

Which model is answering the user’s question? You don’t know. Neither do I. And when you ask these individual models, they give different answers to the same question. Asking a question of 5.1-mini versus 5.1-chat-latest, you’re going to get very different answers. Some even have different technical capabilities than others.

Even when the user is given a choice, there are variations in how the models work. Google Gemini Flash works differently and gives different answers than Gemini Pro. Claude Sonnet 4.5 gives different answers if you have Extended Thinking turned on or off.

And this doesn’t even begin to touch on customization and memory. Features that have been introduced to many AI tools that allow you to personalize and customize them, from giving them cute nicknames to telling them to never use em-dashes, etc. These memories and customizations become part of the overall system prompt, which means that you can’t even reproduce a single sentence questions results from generative AI from user to user because every user is going to have a different set of system instructions and customizations.

When you tell Chat GPT, “hey, remember this about me”, it writes that in its memory, and then that becomes part of every prompt. So that one word difference in the SEO question now could be an 8 paragraph difference from user to user because they all have different memories stored in their ChatGPT or Gemini or Claude. All the major tools provide a support memory, and that is going to completely screw up any ability you have to predict what people are typing in these tools.

So before we proceed, let’s cement this particular piece of knowledge: no one knows what users are typing into AI tools, and no one is likely to approximate it well with any level of credibility.

So what can we know?

Part 2: The Three Phases of AI Results

What we can know with a good degree of reliability is the underlying architecture in general of AI tools and what’s happening behind the scenes. There are three major components and phases:

• The model itself

• Tool handling

• Ranking

Let’s walk through each of these phases.

Phase 1: Model

The first phase is the model itself. Every AI model in the world is trained on massive amounts of data. Today’s models are trained on somewhere around 15-20 trillion tokens, which is the equivalent of a bookshelf of books that goes around the equator of the planet 12 times. It’s a vast amount of information. Days models are also trained on more than text. They’re trained on audio, they’re trained on video, they’re trained on transcripts, they’re trained on PDFs.

Where do AI model makers get the information? Mostly from scraping the web and all the data sources on it, plus whatever they can beg, borrow, or steal. As evidenced by recently settled lawsuits like Anthropic’s, model makers have infringed on creators’ intellectual property rights.

The model is the source of first knowledge for AI systems and tools like ChatGPT. When you type in something like “What are the best SEO tools for optimizing your website”, it immediately begins to create probabilistic associations inside itself, drawing on all the knowledge it’s been exposed to. It hears the term “best SEO tools” and starts looking around for other probabilities, words and phrases and topics and concepts associated with that question and all the words in it.

Here’s the catch: model training is VERY expensive, very time consuming, and very resource intensive. Model makers generally have between 6-18 MONTH lags from the time they start training a model to the time they release it. For example, Google’s brand new Gemini 3 has a knowledge cutoff date of January 2025. A model released last week has knowledge that stops at the beginning of the year, 11 months ago.

We don’t know what’s in the exact recipe for ingredients with each model, but we do know that it’s most publicly available information. We also know YouTube is a huge part of it, as are things like podcasts and other forms of media files that scrapers can quickly obtain and process.

If you doubt this, go to your YouTube channel’s settings. Go to channel settings, go to advanced, and scroll down and look at third-party AI training and look at all the companies that you can give permission to to scrape and train off your YouTube channel. And by the way, if you’re in marketing, when you have marketing content on your YouTube channel, say yes to all of them.

We do know that in general, walled data that isn’t accessible to the public is typically not in AI models either. The general rule of thumb is that if content is accessible to traditional search crawlers, it’s likely accessible to AI.

We also know that model makers like Meta are beginning to use extensive libraries of synthetic data, data that they manufacture that is known to be high quality, to help train their models as well.

Once the model has been built and deployed, it starts fielding questions and queries from users. When someone asks a question like, “What are the best SEO tools for optimizing your website”, the model can answer it, albeit with old data.

Because the data is old, it presents a less helpful solution. A lot of people appropriately ding AI for having hallucinations, presenting incorrect data, out-of-date data, conflicting data, or wrong data. AI toolmakers, to improve the usefulness of AI and continue its adoption, had to come up with a solution for this.

Phase 2: Tool Handling and Search

What is that solution? For the last 18 months, model makers have been developing the concept of “tool handling” in models, an awareness that a model can have multiple tools available to it to do work. These tools are capabilities outside of what AI models themselves can do.

For example, a tool might be a local filesystem read/write tool, something that allows a model to read and write files on your computer. Anthropic’s Claude has this built into their desktop app. The most relevant tool to our discussion is a simple, well-known one:

Web search.

Every model and system today has some kind of web search built in. AI models write their own search queries, search the web, and assess the results.

This fixes the recency problem, where a model might know about a topic but only know about it up to the cutoff of its training data. For things like breaking news or new innovations or new publications, models know to default to web search to deal with gaps in their recent knowledge.

Where this gets interesting is how each model performs web searches. They all write fairly dense, multiword searches that typically exceed what humans search for. People, when they search, tend to write searches between 1-5 words. AI tools generally write between 5-10 word searches, making them more robust AND harder to replicate. When a model is writing an eight to ten word search query, it’s very hard to predict what the model’s gonna search for.

You can see this for yourself, especially if you use command-line AI tools like OpenAI Codex, Gemini Code Assist, or Claude Code. You can see them writing search queries and parsing the results.

So how does a model know when to search the web and when not to? That depends on the model and how it’s built. As a broad generalization, every model creates a list of probabilities in its work. When it’s running, it’s generating not just outputs, but how confident it is in those outputs.

When it produces a result that’s below a certain confidence threshold - and it’s not clear what that threshold is - it will trigger a tool handling call that will invoke a web search. Previous versions of Google’s documentation, for example, cited a confidence threshold of 70% - meaning if the model wasn’t 70% confident in its answer, it would trigger a web search.

OpenAI hasn’t disclosed this for ChatGPT, but their documentation obliquely references ReAct architecture, meaning that the model itself tries to interpret whether a web search is contextually a good idea or not. It does this regardless of probabilities or confidence levels. It says, “this conversation snippet deserves an actual web search”.

But that’s not all. What comes back from a tool handling call is typically a big dataset along with probabilities and other metadata. That raw data isn’t displayed to the end user. Instead, that data has to be used by the AI software client, which brings us to...

Phase 3: AI Ranking

Those raw results need to be processed. Every platform does this differently; when the tools come back with data, it has to be processed and included or excluded in the answer. If you use things like deep research tools, you will even see this in reports. Sources used in the report, sources not used in the report.

As with search, every platform handles this differently. ChatGPT in its web interface has an internal ranker that tries to prioritize results based on relevance - not every web result is relevant. As of GPT-5.1, its own internal agentic structure makes these determinations. Google does this using a system it developed back in 2022 called LaMDA, and that has carried over into Gemini.

Unsurprisingly, Google’s approach very much mirrors the way regular search works, with things like credibility, authority, expertise, etc. being part of the model’s assessment.

What’s critical to note here is that the entire web page isn’t returned for ranking within an AI tool. Instead, snippets are returned to AI tools, and they use that for judging relevance. Where do those snippets come from? From the search engines or from thin slices of the page, typically the first few hundred words. Today’s AI tools will use a lightweight crawler to extract a few hundred words from the page if a relevant snippet was not returned from the search engine.

Part 3: What You Can Control

Now that we understand the overall architecture of how AI takes user input and the three stages it goes through, we have a better sense of how to approach AI optimization.

The first and most important thing we have to realize is that there is no one size fits all strategy for optimizing for AI systems. The three phases that search queries go through have three different strategic approaches. Saying that you’re optimizing for AI means addressing each phase individually, and your strategies for each are going to vary.

Beware of anyone trying to sell you a single AI optimization strategy and they claim this will solve all your AI optimization problems. Each phase is different, and what they’re selling should be clear about which phase they are addressing, and hopefully it’s more than one of them.

Optimizing for Phase 1

Optimizing for phase one means getting knowledge about yourself, your company, and your brands on the internet in as many places as possible. This is all about creating lots of training data for models to consume and train on so that they understand who you are. If you do not exist as a named entity associated with a specific topic or concept in the model itself, it won’t know to invoke you in its answers. A model who has no idea who you are will not execute a web search about you.

For example, if I never published anything online, then Christopher Penn, the marketing AI person, would not exist. You would still have Christopher Penn, the actor, of course. But if I never published anything online, then I would be absent from that training database. And that in turn means that the models would never bring up my name, no matter the level of confidence, when someone asked “who is a marketing AI expert?”.

So optimizing for phase one means becoming known, building your brand, creating a corpus of work that AI model makers can consume and train on. It should be obvious at this point that you don’t want your content to be behind a wall of any kind that AI can’t see if you want AI to recommend you.

As I say in my keynote talk on this topic, you have to be everywhere you can possibly be. You have to create as much content as you practically can and make sure that it is accessible to generative AI model makers.

Create content on YouTube, create podcasts, create blog posts, create newsletters, send out press releases, get coverage in the local newspaper. Be everywhere. Say yes to any podcast that will have you on as a guest, no matter how big or small it is, as long as they publish transcripts and put them online, or even better, put recordings on YouTube and Spotify.

This runs contrary to typical content marketing advice about focusing on creating great content in a few places and being really good at it. That works well for humans and you should do that. But if you are trying to optimize for the way generative AI models work today, you need to be everywhere and not just a few places.

When you optimize for phase one, you need to advise your stakeholders that you will not see results probably for a year because of the long lag time it takes for model makers to ingest data and train models on it. Six months is the bare minimum for smaller models. For big models, foundation models like GPT-5 or Gemini 3, you’re looking at a year before these efforts begin to pay off.

Optimizing for Phase 2

Optimizing for phase two is straightforward and everyone should be good at it. Because optimizing for phase two is optimizing for traditional SEO.

Oops, did you fire your SEO firm because they didn’t have enough shiny buzzwords about GEO/AEO/AIO? Worse, did a slick talking consultant persuade you to fire them?

Get out your knife and fork and prepare to enjoy a lovely crow dinner. Put your hat in hand; if your SEO agency delivered great results in the past and they are up to date on modern SEO, rehire them.

Optimizing for phase two is entirely about optimizing for SEO because web search results are dependent on a good SEO strategy. This part hasn’t changed.

Optimizing for Phase 3

Optimizing for phase three is what happens when results come back from search engines or page crawlers to the models themselves and their interfaces and how those interfaces handle processing that data. This is part SEO and part website management strategy.

If you open up your website to a page in a screen reader or text-based web browser, how many words do you have to get through on the page to get to the actual content? How many navigation bars and menus stand in the way between the reader and the content? Here’s a classic example from CNN:

I kid you not, the first four paragraphs on the CNN homepage, when you look at it in a text browser, asks you about how relevant the ad is to you. You don’t even get to the menus until halfway down the page. The news, the content you came to that page to get is like four pages down. This is a perfect example of how your website should not look.

Putting aside AI for a moment, this is a miserable experience for anyone using a screen reader. If you optimize your website to work in a text-based screen reader like this, you will get an immediate 5% lift in things like conversions, not from AI, but from low-vision people who are suddenly like, “oh, I can use your website now and I can actually buy things from you”.

When a generative AI model extracts a small portion of the page to judge its relevance, what do you think it thinks about when it sees this kind of garbage text like on CNN’s site? How relevant is this kind of text to the conversation? Unless your conversation is about ads, the chances are this is not relevant at all. Which means that even if CNN manages to make it into the raw search results from phase two, once the AI model gets a hold of it in phase three and reads the first few hundred words on the page, it will discard it as irrelevant.

Here’s a guideline that I’ve been giving about optimizing for AI for the last five years: optimize so that your site is a first-class experience for a screen reader or other low-vision assistive device. If you do that well and you present a first class experience where the content is right there, where a person doesn’t have to work terribly hard to get to the point, then you are automatically optimizing for AI to do the same thing.

And as AI agents become more prevalent and in greater use, an AI agent will have a much easier time using a site that is optimized for a screen reader because that’s essentially what it is - a really skilled screen reader.

Next, in terms of content structure, those folks who remember their journalism days remember the inverted pyramid strategy, a.k.a. not burying the lead. This is putting the most important, relevant information up front so that the reader, be it human or machine, gets the point quickly. They don’t have to scroll and page down fifty-two times to get to the content. They can get what they came for right up front.

One of the simplest and easiest tactics you can take is to use your own AI tools to create a one-paragraph summary of any article or blog post that goes at the very top of the post so that when an AI crawler or search bot comes by, it gets the point quickly.

Part 4: Wrapping Up

There’s a tremendous amount of snake oil right now in the AI optimization field, in part because no one’s really sure how to handle these new technologies. No one’s really sure how the technologies work, especially in the marketing space because very few marketers are highly technical folks. And a lot of agencies are struggling to make ends meet and will go to some lengths to retain or win new business, even if they’re not particularly truthful.

When it comes to vetting vendors and agencies, ask them how their AI optimization solution addresses each of the three phases of AI optimization and ask for documentation and proof that they tested their solutions against the technologies themselves. All the information in this issue of the newsletter comes from various developer blogs and official documentation from the AI companies. I don’t have to speculate on any of this because I have the receipts - and so can you.

I have an entire folder of probably close to two hundred academic papers and patents from AI companies documenting all their systems. I just used the deep research tools to gather them. You can do the same thing too.

Be on the lookout for obvious lies and fabrications. “We mined thousands of real queries from ChatGPT!” Oh really? So you violated OpenAI’s Terms of Service and breached their security, which is a felony offense? Or you just made up a bunch of stuff?

Ask for receipts. Ask for proof that isn’t anecdotal. Ask for high quality studies that would pass peer review. One of the things that I am on the lookout for right now is any of these AI optimization vendors to publish a paper submitted to a conference that has been peer reviewed by a peer jury and passes peer review.

So far, the number of vendors who have done that is zero.

The first vendor that publishes a paper that shows how they’re able to effectively execute on the three phases of AI optimization, I will happily recommend once I have seen the peer reviewed paper and it passes evaluation as real.

Finally, if you’d like a completely free of financial cost guide to AI optimization, you can download the Trust Insights 8-page guide here. There’s a form to fill out, but that’s it. It is based on and grounded in all of these academic papers, research papers, and patents that AI companies have put out. That’s where we get our information from. We go to the sources themselves. We go to the receipts, we read the receipts, and we publish our content based upon them.

As you can imagine, my rants at MarketingProfs were considerably more profane and involved a lot of banging my head on the table, so you get the benefit of the much more refined version. I hope it was helpful.

How Was This Issue?

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

• 😄 Good

• 😐 Neutral

• 😦 Bad

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.

[](https://almosttimely.substack.com/action/disable_email?utm_source=almost-timely-newsletter&utm_medium=email&utm_campaign=almost-timely-2024- 11-23)

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.

Advertisement: The Unofficial LinkedIn Algorithm Guide

If you’re wondering whether the LinkedIn ‘algorithm’ has changed, the entire system has changed.

I refreshed the Trust Insights Unofficial LinkedIn Algorithm Guide with the latest technical papers, blog posts, and data from LinkedIn Engineering.

The big news is that not only has the system changed since our last version of the paper (back in May), it’s changed MASSIVELY. It behaves very differently now because there’s all new technology under the hood that’s very clever but focuses much more heavily on relevance than recency, courtesy of a custom-tuned LLM under the hood.

In the updated guide, you’ll learn what the system is, how it works, and most important, what you should do with your profile, content, and engagement to align with the technical aspects of the system, derived from LinkedIn’s own engineering content.

👉 Here’s where to get it, free of financial cost (but with a form fill)

ICYMI: In Case You Missed It

Here’s content from the last week in case things fell through the cracks:

• Almost Timely News: 🗞️ Cultivating an AI Mindset, Part 2 (2025-11-16)

• Live From Boston at MarketingProfs B2B!

• In-Ear Insights: Teaching Workshops and Professional Development

On The Tubes

Here’s what debuted on my YouTube channel this week:

• You Ask, I Answer: How to Remove Email Bot Clicks?

• You Ask, I Answer: Can AI Tabulate Data from PDFs

• You Ask, I Answer: How to Measure PR Beyond AVEs

• You Ask, I Answer: Career Skills for the AI Era

• You Ask, I Answer: How Foundation AI Models Differ

• You Ask, I Answer: How to Disclose AI Use to Clients?

• You Ask, I Answer: Can AI Generate Great Logos?

• You Ask, I Answer: How to Choose AI Tools for Kids?

• You Ask, I Answer: AI Real-Time Vision Capabilities?

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

• 🧭 New! The AI-Ready Strategist

• 💥 Generative AI Use Cases for Marketers

• 👉 Mastering Prompt Engineering for Marketers

• 🦾 Generative AI for Marketers

• 👉 Google Analytics 4 for Marketers

• 👉 Google Search Console for Marketers (🚨 just updated with AI SEO stuff! 🚨)

Free

• 👉 New! From Text to Video in Seconds, a session on AI video generation!

• Never Think Alone: How AI Has Changed Marketing Forever (AMA 2025)

• Generative AI for Tourism and Destination Marketing

• Powering Up Your LinkedIn Profile (For Job Hunters) 2023 Edition

• Building the Data-Driven, AI-Powered Customer Journey for Retail and Ecommerce, 2024 Edition

• The Marketing Singularity: How Generative AI Means the End of Marketing As We Knew It

Advertisement: New AI Book!

In Almost Timeless, generative AI expert Christopher Penn provides the definitive playbook. Drawing on 18 months of in-the-trenches work and insights from thousands of real-world questions, Penn distills the noise into 48 foundational principles—durable mental models that give you a more permanent, strategic understanding of this transformative technology.

In this book, you will learn to:

• Master the Machine: Finally understand why AI acts like a “brilliant but forgetful intern” and turn its quirks into your greatest strength.

• Deploy the Playbook: Move from theory to practice with frameworks for driving real, measurable business value with AI.

• Secure Your Human Advantage: Discover why your creativity, judgment, and ethics are more valuable than ever—and how to leverage them to win.

Stop feeling overwhelmed. Start leading with confidence. By the time you finish Almost Timeless, you won’t just know what to do; you will understand why you are doing it. And in an age of constant change, that understanding is the only real competitive advantage.

👉 Order your copy of Almost Timeless: 48 Foundation Principles of Generative AI today!

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.

• Avp, Generative Ai at Old Republic Specialty Insurance Group

• Chief Ai Officer at Elevation Capital

• Chief Data And Ai Officer at Empiric

• Chief Data Officer at Confidential - Leading HealthTech & Financial Wellness Platform

• Chief Marketing Officer at NU Advisory Partners

• Global Svp, Data & Ai (North America) at Endava

• Growth Analyst / Marketing Data Analyst – Google Ads & Facebook Ads Strategist at Sterling & Co SB

• Head Of Data, Ai & Intelligence at Bizee

• Marketing Director at Connecticut Innovations

• Senior Growth & Lifecycle Marketing Manager at RapidWorks

• Senior Manager, Data Analytics at SimplePractice

I don’t know if you’ve noticed this or not, but when I search for marketing jobs now, an awful lot of them have AI in the job title. That’s an indicator, if there ever was one, that this is broadly a skill set that you can’t ignore.

Advertisement: New AI Strategy Course

Almost every AI course is the same, conceptually. They show you how to prompt, how to set things up - the cooking equivalents of how to use a blender or how to cook a dish. These are foundation skills, and while they’re good and important, you know what’s missing from all of them? How to run a restaurant successfully. That’s the big miss. We’re so focused on the how that we completely lose sight of the why and the what.

This is why our new course, the AI-Ready Strategist, is different. It’s not a collection of prompting techniques or a set of recipes; it’s about why we do things with AI. AI strategy has nothing to do with prompting or the shiny object of the day — it has everything to do with extracting value from AI and avoiding preventable disasters. This course is for everyone in a decision-making capacity because it answers the questions almost every AI hype artist ignores: Why are you even considering AI in the first place? What will you do with it? If your AI strategy is the equivalent of obsessing over blenders while your steakhouse goes out of business, this is the course to get you back on course.

👉 Take the course now!

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 Bluesky - 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

Listen to my theme song as a new single:

• Apple Music

• Spotify

• YouTube

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 are the public events where I’m speaking and attending. Say hi if you’re at an event also:

• Social Media Marketing World, Anaheim, April 2026

There are also private events that aren’t open to the public.

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