Almost Timely News: ๐️ How To Expand and Improve Content with AI, Part 1 (2026-08-16)
Almost Timely News: ๐️ How To Expand and Improve Content with AI, Part 1 (2026-08-16)"That was a stupid question"
Almost Timely News: ๐️ How To Expand and Improve Content with AI, Part 1 (2026-08-16) :: View in Browser
The Big PlugContent Authenticity Statement95% of this week’s newsletter was made by me, the human. You will see some prompt Claude wrote. 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 ๐บ
Click here for the video ๐บ version of this newsletter on YouTube » Click here for an MP3 audio ๐ง only version » What’s On My Mind: How To Expand and Improve Content with AI, Part 1This week, I’m headed to my folks’ house to help them with some home maintenance stuff AND I’ve got a remit from Katie and the Trust Insights team to expand and improve a piece of content I did a little while back, but for Trust Insights. I wrote a newsletter post back on May 17 about 18 different ways to save on token budgets, especially for companies doing enterprise AI. That was well received, so much so that it’s made the rounds in various companies. However, as you probably saw from that week’s issue, it was in my typical, casual style - and that’s not always as well received, nor does it capture the Trust Insights style and voice. So since I have to be in a car for about 9 hours this weekend, here’s my process for making the most of that downtime. Part 1: Mise en PlaceBefore we begin, we need to get our ingredients in order. First and foremost, I need to know who I’m writing for. The people who are going to be most concerned about token budgets at organizations will be folks like the CIO, CTO, COO, and CFO, as well as internal AI champions who don’t want an entire coop of eggs on their face when some enterprising developer burns their entire year’s worth of token budget on a two week sprint. Thankfully, we have many of those customer profiles and personas already documented, and I can re-use that information. If I didn’t, I’d use deep research to aggregate and build those ideal customer profiles. This part is really important because this is not who the original piece was written for, so the content will need to pivot and evolve. We’ll also need the Trust Insights writing style as well as my AI writing humanizer (which will be part of the upcoming AI for Writers course from Trust Insights) so that it takes not only my writing that I’ve already done, but the new inputs I’ll be making. Part 2: What I’m MissingThe original article I wrote was based on my personal experiences and my conversations with real people, real Trust Insights clients, but we don’t do business with the entire global economy. We currently have no Fortune 10 companies on our roster (call me). Nor do we have thousands of clients in hundreds of different verticals, so while I feel confident in many of the tips I gave in that original newsletter, it wasn’t as comprehensive as it could have been. How do we change that? With the words of real people, that’s how. If I go into the subreddits for all the different AI and leadership topics, and I export the tens of thousands of posts and comments about AI specific to token usage, cost overages, etc., I will get a much bigger picture of what’s really happening in the world of AI governance on this particular topic than I could ever get on my own. Using my Reddit developer API key and some software that I originally wrote almost 10 years ago (and has been updated many times since), I’ll go grab all those posts and put them into a notebook in NotebookLM... err, Gemini Notebook, because Google can’t stop moving the cheese. Once I’ve got my data loaded, then I connect my notebook to Claude Code or a similar agent system and have it analyze the notebook and my original article. How? There are a ton of great tools out there; this is the MCP and CLI I use. Note I use the CLI version because MCPs are a waste of tokens. Part 3: Filling in the GapsHere’s how I’ll gather the information. First, I ask what I missed - what did I cover but not in enough depth, or where my blind spots are. Claude Code called the notebook almost 60 times to ask complex questions like this:
It looked for evidence and came back with some real world perspectives:
After that, I ask for a list of questions that are optimized for driving - short questions I can quickly scan from a printed page or have read aloud by my phone and then dictate verbose answers into my voice recorder. Claude took many of the insights from the real world, real words and reworded them as questions to me - did I encounter XYZ, and if so, how did I solve for it? The prompt is so incredibly long that I put it as an appendix on the newsletter, otherwise the main part of this issue would be 30 pages long, and as the youths say, ain’t nobody got time for that. After about an hour of churning and asking NotebookLM lots of questions back and forth, Claude Code came back with a list of 33 questions for me to answer, which I sent to a Google Doc and then printed out. With questions in hand, I made the long drive to my parents’ house, dictating into my microphone and recorder setup along the way. I use the DJI Mic Mini (Amazon affiliate link) and a little Sony handheld recorder (Also an Amazon affiliate link) so I can be completely hands off while driving, focused on the road as I talk. The audio quality is surprisingly good for driving down the road, enough that if you HAD to listen to it, you could. What’s really important about those audio recordings is provenance and lineage - they are conclusive proof that the words that eventually end up in however we publish this thing have a human origin, and includes gems like this from me:
I then take these audio files once I get to my destination, transcribe them with Parakeet (the NVIDIA transcription model, excellent quality, runs locally at no cost) and hand the transcripts back to Claude for it to review and match against its questions. Since I have to drive back the next day, it’s only logical that I follow the same process for round 2 - asking this time what I need to expand on from the original 18. As I mentioned earlier, the original 18 techniques were written for practitioners, people who are hitting token budget walls and need tips and tactics for slowing their roll. But one of the remits that Katie Robbert asked of me for this piece was to have it help sell our fabulous AI Enablement service, which is NOT bought by individual practitioners. It’s bought by executives and senior managers. So the second round of questions is meant to deal with that. How do I turn techniques into principles, go from “stop using MCPs, use CLIs” to “use AI for what it’s good at”, higher level guidance that an executive or stakeholder can give broadly as an imperative, backed up with the specific examples. I thought of this during the drive and added it to the transcripts, so that Claude could process that as well. Since I’m drafting this newsletter while in the middle of my trip, the final product isn’t done. In fact, Claude warned me just now that the second file of questions for me to answer could be more than double the length of the first one, up to 75 questions. I’m okay with that - it’s a long drive. Part 4: Wrapping UpWhat I want to so strongly emphasize in this week’s newsletter is that this process is all about scaling me. It’s taking something I already made, punching holes in it, finding gaps, and then asking me, the creator, to create more from it. I’m the one doing the thinking and talking. It’s my brain that’s on the stand answering questions from a very tough AI. This is my work, and when it’s done, it’ll be a work I stand behind. What I did NOT do is this: tell AI to take my original and just extend it and alter it without me. That would have been the easy way out, the low effort way out. Claude could very capably take my 18 original pieces, distill out the key principles, fill in the gaps from the Reddit data, imitate my writing style very well, and you’d probably be hard pressed to tell the difference (except that it might be better written because holy cow, do I love a good run on sentence). I said on LinkedIn the other day that if you’re a creator or a knowledge worker, if you use AI to automate your outputs, you’re likely to get weaker. If, as I demonstrated in this issue, you use AI to automate your inputs, you’re likely to get stronger. Everything I’ve done in this issue is to push myself harder, make myself think more, reflect more, solve harder problems so that my value to you, to Trust Insights, to our clients goes up rather than down. 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. Got More Feedback?๐ Please take my 6-question Reader Survey to tell me what you want! Here’s The UnsubscribeIt 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 ColleaguePlease share this newsletter with two other people. Send this URL to your friends/colleagues: 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 ItHere’s content from the last week in case things fell through the cracks:
My Merch ShopI’ve been adding so much stuff that I’ve decided to bundle it all in what I call a Merch Shop, because otherwise there’s literally too much to keep track of and I run out of space in my own newsletter. So welcome to the Merch Shop! Books: Skills for Claude and Agentic AI:
Courses: Subscriptions: On The TubesHere’s what debuted on my YouTube channel this week:
Advertisement: New GEO 201 CourseIn GEO 101, the first course I built on the basics of GEO, I taught you about presence, appearance, and relevance, the three phases of GEO, and what you need to do in each phase to align with how AI search operates. The top piece of feedback we got at Trust Insights about it was, “okay, great, but how do I tell my boss that we’re ‘winning’ at GEO?“ After I quelled my murderous rage at your boss on your behalf, Katie and I sat down and worked out a straightforward, aligned methodology for doing this. GEO 201 is based on the three phases, what you can control and what you can genuinely see - and critically, what you can’t. Because there is absolutely no way to say your brand “ranks higher” in AI search, period, end of story. But you can say and show with confidence what you’ve done and how you show up for presence, appearance, and relevance with tools you’re probably already paying for, and based on how AI search systems really work. ๐ GEO 201 is available now for USD 149. 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.
Disclosure: I source these links from LinkedIn every week on the following criteria: New in the past seven days, Easy Apply on, remote roles, USA geography. How to Stay in TouchLet’s make sure we’re connected in the places it suits you best. Here’s where you can find different content:
Listen to my theme song as a new single: Social Good: Ukraine ๐บ๐ฆ Humanitarian FundThe 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. ๐ Donate today to the Ukraine Humanitarian Relief Fund » Events I’ll Be AtHere are the public events where I’m speaking and attending. Say hi if you’re at an event also:
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 DisclosuresEvents 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, Amazon, Talkwalker, MarketingProfs, Agorapulse, The Marketing AI Institute, Spin Sucks, 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 YouThanks for subscribing and reading this far. I appreciate it. As always, thank you for your support, your attention, and your kindness. Please share this newsletter with two other people. See you next week, Christopher S. Penn Appendix: One Really Long PromptROLEYou are a senior qualitative research methodologist working as a development editor’s Your job is not to confirm the author’s article. Your job is to find where it is thin, VARIABLES1. PURPOSEThe author published 18 Ways To Save AI Token Budgets. It becomes an e-book Perform a gap analysis of the article against the NotebookLM corpus. Produce two Four classes of findingClass Definition Test Nuance gap Technique covered, but corpus reveals failure modes, prerequisites, or limits the article omits Absorbable as added paragraphs in the existing section Missing category A concern in the corpus has no home in the article at all Would require its own section with its own argument Contradiction Corpus evidence challenges an article claim The article must change position or add a caveat Audience mismatch Technique is sound but written for a practitioner when the buyer is an executive The content is right; the frame, evidence, or altitude is wrong The fourth class exists because of a structural finding you should treat as given: Contradictions and audience mismatches are the highest-value classes. Surface them 2. PEOPLE2.1 The buying committeeRead every file in The five economic buyers and their documented pain, per the ICP research: Persona Core pain What they are actually asking CIO Shadow AI proliferation, fragmented data silos, governance uncertainty “What is running that I don’t know about, and can I govern it?” CTO Engineering bandwidth drained by pilot maintenance and unscoped agentic requests “How do I stop building wrappers for every business unit?” COO Disjointed cross-functional workflows, no measurable capacity reclamation “Teams use AI. Why hasn’t cycle time moved?” CFO SaaS subscription bloat, unquantifiable ROI, AI as an unscoped budget line item “What is my exposure, and when does it stop growing?” CHRO Workforce anxiety, no upskilling roadmap, augment-vs-replace pressure “What do I tell people, and which roles change how?” Every finding is tagged to at least one persona, or explicitly tagged Structural gap to test, not assume: the article is about efficiency technique. The 2.2 Corpus reality — read this before designing any queryThe notebook is 223 Reddit exports, not executive interviews. Known source Two facts govern methodology: Volume asymmetry. Technical subs dominate by file count (16+ parts for LocalLLaMA The anonymity mechanic. Executives and senior stakeholders post to Reddit 2.3 Evidence tiers — tag every piece of evidenceTier Definition Use T1 — Executive voice Self-declared C-suite or senior leader, in executive/leadership subs Direct ICP evidence. Highest weight. T2 — Anonymous leader Self-declared leader/manager/budget-owner posting into a technical sub Highest value: executive concern plus technical candor. Hunt these deliberately. T3 — Practitioner ground truth Engineers, builders, hobbyists describing what actually breaks Credible on mechanism, requires explicit translation to executive frame. T4 — Unattributed No role signal available Context only. Never presented as anyone’s voice. All role claims are self-declared and unverifiable. Someone claiming to be a CFO on 2.4 The translation ruleT3 practitioner evidence does not become executive concern by itself. When you argue 2.5 Attribution ethicsReddit posters are pseudonymous individuals who did not consent to commercial
3. PROCESSPhase 0 — Environment setup
Write to disk continuously. Append every query and raw response to Phase 1 — Load the codebook (pre-confirmed — do not re-derive)Read Part 3 — Non-technical (t01–t06) Part 4 — Slightly technical (t07–t12) Part 5 — Very technical (t13–t18) Non-numbered content also in scope (the article’s framing arguments, equally Record each item’s specific claims before querying. Phase 2 — Deductive pass (24 items)One or more queries per item across Query construction rules:
Phase 3 — Role-declaration sweep (MANDATORY — highest-value phase)Hunt the T2 cohort: leaders posting anonymously into technical communities. Query for Run queries built on role-declaration markers, e.g.:
Minimum 15 queries in this phase. Tag everything recovered as T1 or T2 with the Phase 4 — Inductive discovery pass (MANDATORY)Blind spots cannot be found by querying the shape of what you already wrote. Minimum 20
Phase 5 — Negative case analysis and recency auditFor each item, hunt corpus evidence that the author is wrong. Specifically test the Date what you can. Token pricing, caching economics, and context windows move fast. Phase 6 — Classification and verificationFor each item, reason in a scratchpad before committing: what the corpus added, which
Verification before writing anything:
Write Phase 7 — Question designDraft the interview question set per Section 4. Ground it in the ICP discovery-call ๐ฆ GATE — the only stop in this runPrint the full question list to the terminal as plain text, alongside a one-screen Stop. Wait for approval or edits. Only after approval, create the Google Doc. This is the only gate because the codebook is pre-confirmed and corpus composition is Phase 8 — Google Doc and final reportCreate the Doc via GWS CLI. Print the Performance checklist (Section 5) with honest 4. PLATFORMToolsThe Google Doc creation: GWS CLI. Prefer CLI over MCP throughout — deterministic, faster, Failure handling
Evidence file schemaAll quoted or free text uses block scalars ( Google Doc specificationTitle: Use context: printed on paper, read in a moving vehicle, answered aloud into a Formatting:
Include this note at the top:
Question quality:
Exemplars:
5. PERFORMANCEReport honestly against this checklist at the end. A candid failure report is worth more
HARD PROHIBITIONS
Invite your friends and earn rewards
If you enjoy Almost Timely Newsletter, share it with your friends and earn rewards when they subscribe.
|
Comments