Currently accepting 2 clients for Q3 2026
See if we are a good fit
Growth

your AI is running on an outdated version of you

first off - there are 100+ of you on this list now. thank you. genuinely.

first off - there are 100+ of you on this list now.

thank you. genuinely. for every single one of you following along on this journey 🙏

if you get something out of these emails, do me a favor - forward this to a marketer, a tinkerer, or a friend who’d actually enjoy it. that’s how this little thing grows.

and one quick favor while i’ve got you - hit reply with ONE sentence on what you want to see more of from me.

(a bunch of you already asked for the google ads agent
 so i’m recording a full walkthrough loom so you can steal the whole setup. soon 👀)

ok. here’s your stupid-simple, high-ROI insight for the week.

i went to 4 events in less than 7 days.

i did not plan that.

if you read my last email, you know i was at SaaStanak - building live, pitching investors, getting cooked by the judges


that wrapped Wednesday.

a good friend and i went to Zagreb to chill before ECOM Architects
 that was the plan. chill.

then someone goes “hey, there’s an n8n event tomorrow, you in?”

soo we signed up. ofc.

and that tiny room kinda broke my brain.

10 people. no stage, no decks, no “thought leadership.”

just builders trading the actual nitty-gritty - their automations, how they land clients, their setups, which LLMs they run for what


and i’m sitting there like
 fck. i’m behind.

not because they were doing anything magic - they were just DOING it. openly, fast, while most of the industry sits on its hands.

(then the next day there was a crypto event. ~30 people. went to that too. don’t ask 😄 at that point i think i was just collecting events)

then finally ECOM Architects - the one i actually came for. good marketers, real conversations, probably opened the door to a couple new clients.

but the thing that stuck with me across all four wasn’t a tactic.

it was this:

most DTC and consumer SaaS brands aren’t using AI. like
 at all.

they’re still doing a stupid amount of manual work by hand. uploading ads. rebuilding reports. copy-pasting numbers into a friday deck so a client feels informed.

that’s not an AI problem.

that’s a “you never turned the boring repeatable stuff into a system” problem.

and here’s the part almost nobody fixes first


most of you don’t want to run OPENCLAW or HERMES or some custom agent. totally fine - you don’t need to.

but the LLM you’re already using - chatgpt, claude, whatever - is running on a half-outdated, half-empty picture of you and your business.

so it hands you generic answers
 because you fed it generic context.

fix THAT before you touch any fancy automation.

here’s how:

  1. drop the prompt below into your LLM of choice
  2. read what comes back - slowly. (you’ll be surprised how outdated the picture it has of you actually is)
  3. correct it. feed it your real numbers, your current frameworks, your beliefs, the way you actually work now - so every answer from today forward gets sharper

here’s your prompt 👇

Export all of my stored memories and any context you’ve learned about me from past conversations. Preserve my words verbatim where possible, especially for instructions and preferences.

## Categories (output in this order):

1. **Instructions**: Rules I’ve explicitly asked you to follow going forward — tone, format, style, “always do X”, “never do Y”, and corrections to your behavior. Only include rules from stored memories, not from conversations.

2. **Identity**: Name, age, location, education, family, relationships, languages, and personal interests.

3. **Career**: Current and past roles, companies, and general skill areas.

4. **Projects**: Projects I meaningfully built or committed to. Ideally ONE entry per project. Include what it does, current status, and any key decisions. Use the project name or a short descriptor as the first words of the entry.

5. **Preferences**: Opinions, tastes, and working-style preferences that apply broadly.

## Format: Use section headers for each category. Within each category, list one entry per line, sorted by oldest date first. Format each line as: [YYYY-MM-DD] - Entry content here. If no date is known, use [unknown] instead.

## Output: - Wrap the entire export in a single code block for easy copying. - After the code block, state whether this is the complete set or if more remain.

run it

takes 2 minutes and it’s the cheapest upgrade to your AI output you’ll make all week

Talk soon,

Kris

Next Letter, Next Wednesday

Want the next one in your inbox?

One letter, every week. Stupid simple, high-ROI insights for founders who need profit, not theories.