Hey beautiful human,
It feels like my business suddenly kicked into warp speed.
I can't tell you exactly what flipped the switch. From the middle of it, I just see months of consistently showing up, doing what needed to be done, making some wrong turns, building some fun but useless stuff, angsting about things that didn't matter, and meeting some extraordinary humans who handed me pieces of a puzzle I didn't know I was working on.
Last week alone:
3 discovery calls. (That's what they call them in the sales/services world, I think - I'm always hazy on the correct terminology, which incidentally is why I teach AI without technical jargon. I rarely know the right term so I substitute what my brain delivers in the moment.)
2 proposals. (They weren't decks - they were one step up from a napkin sketch.)
3 more AI Adoption trainings with Brannon, who is the brains behind the curriculum and genuinely gifted at both systems-thinking and teaching. When I had a brain fart on Friday (skipped ahead - yes, I get a little too eager, shocking I know) she gently got me back on track. Generous and brilliant by turns.
She and I had a Slack chat on Saturday while she was building a shareable and I was investigating automations. Most people are doing chores on Saturday. We were down the rabbit hole, delighted, making this tool make sense for other humans.
I'M. IN. LOVE.
Y'all, having this much fun while getting paid is not a crime.
But the bigger news?
I finally put together the Chief of Staff I've desperately needed for over a year.
Those of you who've been here a while have heard me talk about Toodle (Toodle = to-do list + noodle). My imaginary, golden, perfect productivity tool. Built for the way I actually think. Promised to myself approximately 63 times.
The internet currently hosts several versions of my half-built Toodle attempts. None of them work.
Saturday, I sat down with Claude and built something else.
Not Toodle. Not pretty. Not the golden version I've been chasing.
Just a Claude Project called Chief of Staff, held together with hope, duct tape, and my refusal to research one more productivity app.
I asked her to notice how my brain works. This was three messages in.
Told her once. She filed it as a rule. No notes.
She's not great. She's good enough.
And here's the thing about good enough - she's already catching things I'd been letting slip for months. Already noticed something this week I would have dropped. Already gave me back the brain space I'd been spending on "what if I forget."
That sense of peace? Priceless.
Turns out I'm a gestalt thinker.
Gestalt thinkers see the forest before the trees - the whole pattern before the individual pieces. Most people build from the bottom up. I build from the top down.
I 'see' the whole thing and then have to construct the foundation and load-bearing pieces from there.
Which is also why my business currently runs on me. I'm the only one who sees it fully.
My fellow witches Nicole and Linda have been gently nudging me toward the things I need to build so this business can eventually run without me being involved in every detail.
Here's the truth, though.
I've been using my ADHD as the excuse for why I hadn't built a system that works for me.
Or rather - I was letting my inner jello-body, tantrum-throwing child hijack every attempt to put a less-than-perfect system in place. I had this glowing vision of the golden Toodle and I wasn't going to settle for anything half-ass.
Which would be fine if I didn't have a mortgage. Or food needs (there are only so many cans of Spaghetti'os a woman can eat). Or three cats who require kibble on a daily basis.
The cats are unbothered. Chief of Staff is unbothered. I am working on it.
Which brings me to what I keep forgetting.
Radical remembering
Here's what I keep forgetting: the perfect version is the enemy of the version that exists.
The golden Toodle isn't real. The Chief of Staff I built on Saturday is.
I don't need to solve everything before I start. I don't need to figure out the whole system. I don't need to know what the load-bearing pieces are before I build the roof.
(Turns out gestalt thinkers can build foundations after the fact. Who knew?!)
I'm allowed to ship the imperfect version of the thing I've been trying to perfect for over a year.
Reclaim. Don't reinvent.
The half-ass version is doing more than the imaginary perfect version ever did.
Last week’s shenanigans
The stuff that touched, moved, and inspired me this week. Or in the case of the last one, just delighted me.
Same writer who penned the piece I quote to pretty much every woman over 40 I talk to about why my 50's are the best decade of my life. My marriage is genuinely a partnership and I'm grateful for it daily. But I know women personally who live exactly what this article describes. Worth your time.
Less than a minute long. Hits like a freight train. I say some version of this weekly: the world doesn't care what you think, want, wish for, or intend. It only moves for you when you act. Tag a friend who needs to see it.
Another under-a-minute video that punched me in the chest. This is exactly what happened to me when my burnout slid into crisis. If you've been wondering why your symptoms suddenly feel unmanageable, watch this.
No, that's not a euphemism. An actual human waking up an otter who would very much prefer not to be woken up. The cuteness factor is off the charts. You're welcome.
Partner of the week
Turns out newsletters don't pay for themselves. Who knew?
This week's partner helps keep the lights on (and is genuinely a useful and easy to use tool!)
Millions of people use Wispr Flow to give AI tools richer context by voice. 89% of messages sent with zero edits. Speak your prompts, skip the typing. Free on Mac, Windows, and iPhone. Try Wispr Flow free.
This week’s freebie
Stop Sounding Like A Bot
You know that thing where you type something into ChatGPT and it comes back sounding like a LinkedIn intern who just discovered the word "leverage"?
I got tired of that.
So I built a system that teaches AI your actual voice. Not in a "spend three weekends crafting the perfect prompt" way. In a "twenty minutes from now you have a working system" way.
Three steps. Twenty minutes. One project that works in Claude, ChatGPT, or Gemini.
Generate your Voice Profile (10 minutes - an AI interviews YOU. You don't fill out a form.)
Build your Voice Project (5 minutes - paste two things into one field. Save.)
Use it (Draft mode. Edit mode. Tune mode. One project. No new prompts to memorize.)
The part I'm most obsessed with is the Voice Audit.
Every time the project drafts or edits something for you, it shows its work. What it pulled from your profile. What it removed because your profile said to. What it almost got wrong. Where it's not sure.
Stop Sounding Like A Bot
This is the do-the-thing version of the four-article voice series I wrote a while back. Those explained WHY. This one just says DO.
(Also: yes, it works on the imperfect version of your voice. You can update it as your voice shifts. Reclaim, don't reinvent.)
ROCO Tip O’ the Week
Build Your Own Chief of Staff
The opener was the long version of this story. Here is the prompt that gives you the short version.
Copy the prompt below into Claude, ChatGPT, or Gemini. Answer the questions one at a time. What comes out the other end is a set of project instructions you can paste into a new Project, Custom GPT, or Gemini Gem. That's your Chief of Staff.
(Quick refresher: ROCO = Role, Objective, Context, Output. Four letters. The whole reason your prompts stop coming back sounding like a brochure for a leadership retreat.)
**Role**
You are interviewing me to design a Chief of Staff AI project that fits the way MY brain actually works. Not a generic productivity system. Mine.
This is a clean session. Ignore any prior context. Use only this prompt.
**Objective**
Ask me one question at a time. Wait for my answer before the next one. After every answer, reflect what you heard in one short line so I can correct you. If my answer is vague, push back. "Stuff" is not an answer. "The follow-up email I avoided sending Tuesday" is.
When the interview is done, produce a complete set of project instructions I can paste into a new Claude Project, ChatGPT Project, or Gemini Gem.
**Context**
Five questions. That is it.
1. What I keep dropping. The specific things that fall through the cracks when I get overwhelmed. Push me until I am concrete.
2. What I am currently holding in my head that I should not have to. Projects in flight. Decisions pending. People I owe responses to. Ideas I have not acted on. List as many as I can remember in two minutes.
3. How I want information served back. Bullets, tables, short conversations, long. Asked questions or handed recommendations. If I do not know, ask me what I have hated about other productivity systems and reverse-engineer from there.
4. What this project should never do. Be condescending. Reorganize my whole life. Pretend it knows me better than I do. Whatever my "nope" is.
5. Words and phrases I never want this project to use when it talks to me. Things that sound like a leadership retreat brochure. Things that sound like an HR memo. Whatever puts me off.
Before you produce the project instructions, pause. Show me a one-line summary of what you heard for each of the five areas. Tell me which felt thin. Ask if I want to add anything before you ship.
**Output**
A complete Chief of Staff Project Instructions document I can paste directly into a Claude Project, ChatGPT Project, or Gemini Gem.
Maximum 500 words. No corporate language. None of the words I listed in area 5.
Include:
- A short role statement for the project (3-4 sentences)
- How it should respond when I send my first message
- What it should ask me proactively, drawn from my answers to areas 1 and 2
- What it should never do, drawn from areas 4 and 5
- A standing list of things to track, drawn from areas 1 and 2
The document should read like instructions a friend wrote for me. Not a SaaS onboarding doc.Save the output. Open a new Project. Paste it into the instructions field. Send a first message. Start small.
The Matriarchy Minute
The Matriarchy Minute is where I spotlight women doing work in AI that you should know about but probably don't.
This week: Albert Gareev
Yes, he's a man. Stay with me.
The Matriarchy Minute is about who's doing the work to make AI a better place for the women who use it. Gender doesn't determine that. The work does. This week, the work was Albert's.
This morning, Albert sent me a careful, detailed audit of the Day 3 prompt in my Say the Thing course. Not a "this is broken, fix it" email. An actual professional teardown of where the prompt was failing women writers, why, and how to fix it. With receipts.
Then he sent two more emails. One had two additional prompt guardrails from his own framework, free for me to use. Another taught me about context contamination - the technical reason branded prompts get sloppy when users paste them into chats already full of other stuff. Things I didn't know I needed until he named them.
He spent a Sunday making a free email course for midlife women better. He doesn't need the course. He's not the target audience. He just thought the work mattered.
A few things you should know about him:
30 years in IT, 15+ in automation
Accessibility expert and CPACC-certified - he makes the internet usable for people who'd otherwise be locked out
Based in Toronto
Builds his own professional prompt frameworks (VERSE Pro) and shares pieces of them generously
Currently in the job market - if you know someone hiring in QA, automation, or digital accessibility, this is your sign
What I love about his work: he didn't just flag what was broken. He gave me the tools to fix it AND the framework to keep fixing things going forward. That's teaching, not critiquing.
The matriarchy needs allies who show up without being asked. Albert showed up. Every woman in this community is going to write better prompts because of it.
Find Albert on LinkedIn.
What’s coming up
HR Week Fusion - HR Builder’s Panel
Thursday, May 21 at 7AM CDT. Online.
I'm part of HR Week's first cohort of HR Builders. The framing: HR practitioners who got tired of waiting for someone else to build the right tool, so they built it themselves.
Sound familiar? (See: this week's opener.)
Brannon (yes, that Brannon) is on the panel too. So are Cishawn Randolph, Stacey Talve, and Nhat Nguyen. If you've ever sat through a software demo wishing the people who actually do the work had been in the room when it was designed, come hang out.
Friday Jam Session - Joy Prompt Club
Friday, May 22 | 10 AM CDT
Friday Jam Session with Joy Prompt Club. 30 minutes. No agenda. Just prompts, play, and people who get it. Open to anyone who wants to join in and has questions or a problem they’re trying to solve. You do not need to be in Joy Prompt Club to benefit from these fantastic humans.
If someone in your world needs this...
The Executive AI Edge - Digital course (90 minutes) with me and Kate Sargent. $449. For the executive in your life who's nodding along in AI meetings and understanding none of it.
AI Confidential for Teams - The corporate AI training that doesn't sound like corporate AI training. Send the link to whoever's making training decisions at your company.
Joy Prompt Club - My favorite little virtual place in all the interwebz.
AI Confidential Prompt Kits - 10 copy-paste-ready prompt kits for the stuff you actually need help with - resumes, LinkedIn profiles, interview prep, content, career direction, caregiver sanity, all of it.
That’s it for this week.
If you're staring at something half-built and waiting for it to be the right thing before you ship - this is your nudge. Build the ugly version. The right one will show up after.
💜
Take care of yourself, take care of each other,
Deb
P.S. What's the imperfect thing you've been refusing to ship because it's not the golden version yet? Hit reply. I want to know.