I think we've settled. Settled for the first way we were given to work with AI. The chat interface. Yes, it works. It's largely effective. But it has its flaws. It's one-dimensional. Send a prompt, wait for a response, send a follow-up, and so on until you get what you wanted. Then the session ends. Context is the biggest issue. So we all keep ever-growing project files to try and hold onto it. If you worked with a person like this, it would be ridiculous. Nobody would have a clue what was going on. So why settle? I'm not. That's why I built Context Board. I've been using it for the past week, and my daily workflow has changed more in that week than in the last year. My agent can now: - Pick up where we left off, because the plan lives on the board, not in a chat - Build a custom dashboard that tracks the metrics that actually matter, live - Map out an idea with me so we can both see it - Move clients through a pipeline and keep every next step up to date - Draft social posts and line them up on a content calendar - Work through an in-depth to-do list and tick each task off, instead of one massive session I may be biased, but I think this is the future of agentic workflows. contextbolt.com/board

A few weeks ago I had a great chat with @nickkirt for his Makers Series. We spoke about how I built @ContextBolt and my journey so far as a solo founder👇 I will continue to build in public and share updates/insights here on @X, so come along for the ride. 99helpers.com/makers/context…

Great way to start the day. 4,000 clicks from Google in the last 30 days. Six months ago I was struggling to hit 10 clicks a day. Most days it was 2. So what changed? Honestly, not much. Same fundamentals, just done every day. - Only go after keywords I can actually win. Check who ranks and how strong they are before writing a word. - Write the post that answers the search. Real steps, not a one-line answer Google can lift into an AI Overview. - One post a day. Every day. - Link every new post into the ones around it. - Every morning, fix old posts sitting just off page one with impressions and no clicks. - Change one thing at a time, check it two weeks later keep what worked. Boring works.

When I built my SEO tool I didn't know if it would be any good. I was launching my startup and didn't want to pay $100's for an SEO tool where I would spend hours stuck inside a dashboard. So I built the alternative, an SEO tool that was driven through MCP. At the time it wasn't really a thing, some legacy tools had MCPs but they were poorly constructed and under used. I wanted my tool to be truly AI-native, it didn't even have a dashboard for the first 5 months. 6 months later I am happy to report that my SEO tool is actually pretty good. In fact my first ever customer is still with me, paying $35 every month to run all their SEO work through my tool. This is my Search Console Graph. 👇 Every piece of content and SEO decision was done with my own tool. Pretty cool right?

SEO can often feel like it's going no where. Then you zoom out and see the progress. Last three months are looking strong!

Marc replaced beehiiv. The days of legacy products overpricing are over. They are no longer the only option. I built one product that replaces: Planable Notion Pipedrive Sunsama Every dashboard tab you open each morning It's called Context Board, it's completely customizable, connects to your APIs/MCPs and shows you the stuff that actually matters. It's also fully agentic. Everything you can do, so can your agent. contextbolt.com/board

Dashboards are ugly. Every individual tool has it's own. So I designed one for all your tools. Your agent builds it for you. Connect your APIs and MCP servers. Ask Claude to build you a custom widget. Pick a theme, or describe one and it designs it. contextbolt.com/board/

I pushed my Claude Max subscription to the limit. Almost $2K worth of usage in one day. The best part was, I was out all day and just checking in on my phone. 5.34B tokens and I was AFK 😂 What's the most tokens you've burned through in 24 hours?

I have exactly 60 paid users. That's 60 strangers on the internet who pay every month for something I built. The revenue is nice. Feedback is better. Most weeks, I ship at least one feature a user asked for. So I answer every email and every DM. It's how my products stay ahead.

My little startup just hit $3000 total revenue. 🥳 About a third of that came in the last month. About a hundred sales in and I still can't get used to the feeling when I make a sale. There's been a lot of ups and downs, but like most things if you keep at it you'll see progress. Excited for the future. New product launch coming soon 👀

Welp. Apparently my startup doesn't qualify for @claudeai's Startup Program. I wonder what metric determined whether or not you are eligible.

Thanks for teasing me @claudeai This morning I was accepted into the Claude Startups program. A few hours later my place was removed. Anthropic has 'underestimated demand for the program'. Has this happened to anyone else?

If you aren't optimizing your website content for AI search, you are leaving revenue on the table. Here's what's worked for me: Put the answer in the first two lines. Google quotes the top of the page most. Answer the questions people ask next. Google looks those up too and quotes from them. Share something only you have. Your own numbers, your own tests. Don't hide your page's preview from Google. No preview, no quote.

Here's my GSC Performance graph for the past 3 months. As you can see, clicks and impressions have almost tripled. Here's what worked for me: 1. Post every day: A new blog post every day. That's over 100 since July. More pages means more chances to show up. 2. Only write what I can win: Before I write a post, I look at who already ranks. If page one is all huge sites, we skip it. Plenty of searches are still answered by small blogs, and those are the ones we go after. 3. Fix the posts already on page one: A post at position 6 that nobody clicks is the easiest win on your site. Sharpen the title and answer the question in the first two lines. I fix one old post a day. 4. Update existing pages: Prices change, steps change and tools get renamed. When a post goes out of date, we update it and then change the date. This keeps Google happy. 5. Build clusters, not one offs: When a topic works, we write the posts around it and link them all together. Each new post gets help from the ones before it. 6. Write content on emerging topics: Everything is moving so fast you can target topics that don't even exist yet.

AI has made it 10x easier to produce content. But most sites are still shooting in the dark. They mass produce blogs without the data and insights that actually make them worthwhile. The simplest solution to this is to connect your AI to you Search Console. Then it can at least see what is working and what's not before it blindly publishes your next 10 blogs. Combine that with live SEO data and you're on to a winner.

500k Impressions in September! It's been a strong month for SEO. All time highs across all metrics. Time to hit a milli in October 📈