Adam Sprague

@eliana_jordan Probably neither. X articles bury your value behind a click, and the feed punishes anything that makes people leave. Put the actual idea in the post. Let the article be the deep cut for people who already want more.

everyone says articles perform well on x. mine are often my worst performing posts. and i actually put effort into them. so now i’m wondering: am i writing things nobody cares about… or am i just posting them wrong? roast me

@etainos The "AI is for coders" framing is the biggest miss. Treating it like a system you delegate to and steer, not a magic button, is what separates leverage from slop.

Here's how I use AI day to day as a founder who lives mostly on the non-technical side. Too much of this conversation still assumes the best AI workflows are only for coders. They're not. What changed for me was using AI as a system I delegate to, review, and steer. Disclaimer: I'm optimizing for operating leverage and faster decision making that takes in a significant amount of data I otherwise would struggle to find time to analyze. What that looks like: - I run multiple streams at once in Claude Code and/or Hello Deck AI. One thread researches a market, another drafts messaging, another pressure-tests a product decision. I treat them like jobs I assign, redirect, and review when I'm ready vs. trying to do multiple things in a single session. - I ask for multiple iterations against one objective, framed as a day of work. For instance "run through 10 iterations of {ask}." I've found instead of stopping at the first answer, it drafts, critiques its own work, and tries again. The end result is far stronger than version one. - Combined with the above: I dispatch a team of agents at a single goal, each with a different job. One works the go-to-market angle, another acts as strategist, another as technical expert, and so on. Then I combine what's strong and discard the rest. It feels closer to managing a team than prompting a chatbot. In practice this means "dispatch a team of agents with agent a acting as gtm lead... each agent should run 10 iterations of {ask} which is the equivalent of one workday" - I memorialize what works into runbooks my agents build by looking back on the steps we took. Next time I want to repeat something, I point an agent at the runbook and it has everything it needs. No starting from scratch. - I use Granola for call notes because it integrates cleanly with Deck. That turns conversations into usable follow-up and context instead of notes I never read again. It's also super lightweight and I appreciate the templates and the new primer feature they released. - I set scheduled tasks in Deck to review product data, monitor Slack and HubSpot, and send me summaries of what actually needs my attention. It creates a recurring layer of review so I'm not manually checking every system for what changed. For what doesn't fit Deck, I run local cron jobs that ship data to my assistant by email. - I kick off one long research task before bed almost every night. We often say there aren't enough hours in the day. Now I wake up to real progress, because it spent hours structuring and refining while I was offline. - I maintain my corpus of context: family, product, GTM, voice. Better context is what makes the output consistently useful instead of randomly impressive. - I use plan mode religiously. A lot of bad AI usage is just bad task definition. When the plan is clear, the output gets better. For non-technical users this can improve output significantly. Planning improves writing and analysis as much as it improves code.

@clootybapper "Very advanced onboarding" is the most accurate description of the feed I've read all year. It doesn't show you what you like. It shows you what you can't stop reacting to. Very different things.

it is 2026 you liked one wet cat picture now your entire timeline is men discussing chin water you try to leave but the algorithm has already placed you in the enclosure very advanced onboarding

@victorbercaru Sharp. The hours, the editing, the reps all have a cost, it just never lands on an invoice. Treat organic like a budget and you stop posting for applause and start posting for return.

Organic reach isn't free traffic. It's paid media you forgot to measure.

@ruben_vdzr The receipt-as-email-marketing trap is everywhere. The money is in the flows nobody sets up.Post-purchase, winback, browse abandon. That's retention revenue sitting on the table while brands blast one newsletter a month.

Most ecom brands send a receipt and call that email marketing. I built email systems for brands before I ran my own. The ones doing real retention revenue weren't sending the most emails. They were sending the right one at the right moment. The most valuable email you'll ever send goes out 48 hours after someone buys, before the product even arrives. Not a discount. Not a review request. Just a message that makes the customer feel like they ordered from a real brand. That window between purchase and delivery is where people decide if they'll ever buy from you again. Most founders are already planning the next ad campaign while it sits open. Build the relationship before you try to monetize it a second time.

@Saturdaymatters That question is the whole sales process now. Founders aren't buying services, they're buying whether you'll really show up. Every founder who's been burned is asking one thing: are you going to disappear after the brief.

Got on an onboarding call with a founder two weeks ago. Before I could finish my intake questions, he stopped me. He wanted to know who I was. Whether I'd actually be involved or just hand off the brief. He'd been burned before. Last agency gave him copy that could've been written for anyone selling anything. Here's what this guy had already done before he ever hired us: > Spent a full month researching his avatar before he built the product > Compiled every word, phrase, and complaint his customers use on Reddit and Amazon into a document > Built the product around a gap (everyone else is selling gummies with sugar and food dye to kids who react to sugar and food dye) > Migrated subscription platforms specifically because the old ones' dunning system was producing bad churn data and he couldn't trust the numbers > Fulfills out of his home with his family because he trusts them most He didn't ask how our process worked. He asked whether the right person would be touching his brand. Most founders never think to ask that. They hire the agency, let whoever gets assigned do the work, then complain three months later that the copy doesn't sound like them. This guy treated the partnership like a hire. That level of intentionality doesn't just show up in how he vets vendors. It shows up in everything he's built.

@cedric_design Channels don't die. The copy-paste playbook does. Boring was always the thing on life support.

"SEO is dead." "Email is dead." "Cold outreach is dead." "Design is dead." "The funnel is dead." Nothing in B2B marketing is dead. Boring is dead. Generic is dead. The playbook every competitor copied is dead. SEO still works, when you actually have something worth ranking for. Email still works, when it doesn't read like Mailchimp crap. Cold outreach still works, when it's about them, not your Series B. The channel was never the problem. The mediocrity was. CMOs who keep blaming the tactic are just avoiding the harder truth: Their execution got lazy, not the channel. Stop attending funerals. Start raising the bar.

@ItsKieranDrew The weird stuff being a magnet is the whole secret. Most people sand it off trying to sound professional. Sand them down and you blend into everyone writing the same safe takes.

If you feel alone with your thoughts, start writing online. The things that make you weird become a magnet for like-minded people, and you stop feeling like you need to conform. Thinking differently is something to celebrate.

@kanekallaway Visuals don't replace the idea, but they decide if anyone gets to the idea. A weak thought with great design still flops. A great thought with no visual hierarchy never gets read. You need both doing their job.

Lots of people say "visuals don't matter" in content. This is nonsense. The people that say this are trying to convince you that raw yapping is the superior strategy (so they can sell you the yapping course). Premium visuals are the fastest way to differentiate a category-leading personal brand. This is because: 1. Visual context is richer: A picture's worth 1,000 words. A video's worth 1,000 pictures. 2. Visual communicate silently: 85% of people watch videos on silent. There was a false narrative that premium visuals were commoditized because "any editor can make them." Bad visuals are commoditzed. Great visuals are the easiest way to cut through and will be for a long-time. The only way to win without having a strong visual format is if you are: • Top .01% charismatic • Top .01% volume/articulation skill • Top .01% proof If you're not this, but have resources, visual differentiation is a helpful method to cut through faster. Not to say you have to have premium visuals to win (there are plenty of counterpoints), but it is the ultimate separator. There's a reason why all the top shows (TBPN, DOAC, Modern Wisdom, CHD, Rogan) have premium visuals. There is a premium paid for premium, both from brands and in the subconscious brain of the viewer.

@leonabboud Gold. You don't need case studies to start. You need to show your thinking out loud. Document how you solve the problem and the proof builds itself in public. The first case study is you, working.

How to build your personal brand when too have no case studies

@DanielSmidstrup 80% from one channel is the dream and the risk in the same number. Build in public clearly works for you. I'd start feeding a second track now while X is hot! One platform owning your traffic is great until the day it isn't.

X is driving 80% of ClimbX traffic. Google is second largest with 5%. 38 days in. Build in Public works!?

@palassis @bsheldonx Willing to put my first born on the line. 🤣 Substance and distribution are the whole game now. Timing just decides how fast it pays off.

@bsheldonx Organic reach got weird fast. Feels like you need substance, timing, distribution, and a minor blood sacrifice to get the right people in the room now 😂 followed for this.

Why has it become so incredibly challenging to get any kind of organic reach on X in the past year or two? Not claiming to be a cultural mover here by any means, but used to be able to engage in conversation with smart people if there was substance. Bots that bad even with an X sub? cc: @nikitabier

@MindXacademy This is the reframe nobody wants to hear. The algorithm isn't suppressing you, it just has nothing to work with. Give people a reason to stop and it stops being your enemy. Boring is the only thing it truly buries.

The algorithm did not kill organic reach. Boring content did. When you write something that makes someone stop, think, save, or share, the algorithm has no incentive to bury it. Chase engagement because you're saying something worth engaging with, not because you've found the hack. The hack changes. The quality doesn't have to.

Most of the work that builds a brand is invisible. Reps nobody sees. Posts that flop. Systems you rebuild twice. You keep going because you can see what it turns into.

I'd rather have 5 people who trust me than 5,000 who scroll past. Reach is rented. Trust is owned.

@msrshahisoft Clean schema makes you readable. It doesn't make you repeatable. The model trusts the brand telling one story everywhere, not just the one with the nicest JSON-LD.

If your marketing playbook is still built around "driving organic clicks," you’re fighting for real estate on a dying map. The shift from classic SEO (Search Engine Optimization) to GEO (Generative Engine Optimization) isn't a trend on the horizon. It is the baseline reality of 2026. When users consult ChatGPT, Perplexity, Gemini, or AI Overviews, they aren’t looking for a directory of URLs to click through. They expect a definitive, instantly summarized solution. For growth teams, this flips the traditional rule of digital marketing: Classic SEO focuses on winning a high placement. GEO focuses on embedding your brand into the LLM’s actual output as the verified authority. If an AI tool compiles a response and leaves your insights out of the loop, your organic reach effectively hits zero-no matter how dominant your keyword rankings used to be. The Architecture of an AI-Friendly Resource LLMs don't read web copy the way legacy search bots used to. They look for clean architecture, contextual depth, and data-backed authority rather than keyword repetition. To guarantee your content gets pulled into AI answers, it requires: Granular Semantic Framing: Intuitive heading logic (H1-H3) and targeted summary definitions that AI models can effortlessly parse and ingest. Flawless Schema Architecture: Comprehensive, hardcoded JSON-LD tags (such as Dataset and Product Schema) that serve as a direct guide for AI data harvesters. Exclusive Information Layers: Cookie-cutter articles fail here. Large language models prioritize original case studies, internal statistics, and localized insights (designed to feed secure, real-time RAG systems). Cross-LLM Readability: Text must be structured clearly enough to remain coherent and authoritative across the differing logic engines of Claude, Gemini, and OpenAI. Scaling Mass Production Safely The biggest challenge facing modern growth teams and digital agencies is volume. How do you scale topical authority and roll out complex, localized info hubs without relying on standard, low-value AI copywriting? The fix is abandoning basic, manual prompt engineering and upgrading to Autonomous AI Workflows baked directly into your backend CMS. By linking your company's proprietary data (PDF files, case studies, unique transcripts) straight to your content pipeline, you can seamlessly deploy massive amounts of original, clean, and highly structured pages that engage human readers while providing the exact technical framework that AI search platforms require. The future of brand discovery isn't about outsmarting the AI; it's about feeding it the primary sources it needs to function. Are you still optimizing for clicks, or are you optimizing for source citations? #GEO #SEO2026 #SearchGenerativeExperience #ContentMarketing #AIOps #ProgrammaticSEO #EnterpriseMarketing

Nobody claps when you're building the boring foundation. No likes for the systems, the docs, the late nights. That's the part that pays you later.

Pick one sentence that says who you are. Now say it the same way everywhere. That repetition is what teaches AI to recommend you.

You don't need more content. You need content a machine can read, trust, and repeat back to a buyer.

json runs your entire website and you've never once said thank you. 🤣