There are thousands of AI tools with a landing page and a waitlist. Your business needs
about eight of them, and probably only two at a time. This lesson is the honest map: what each
major tool actually does, what it costs this month, and the two-tool starter stack built
for your exact line of work.
Updated August 1, 2026Refreshed monthlySources: Anthropic · OpenAI · Google · Perplexity · Midjourney · Gamma
Core concept 01
Three roles, not thirty tabs. Everything else is a spoke.
The tool landscape looks like a flood because it is sold as one. Sort it by the role a
tool plays and it collapses into three. A generalist is your home base, the place
you open first for almost anything: Claude, ChatGPT, or Gemini, one of them, your choice. A
specialist does one job better than the generalist ever will: Perplexity for research
with sources attached, Midjourney for images, Gamma for decks. And the newest layer, the
agent, runs a whole multi-step task instead of answering once, mostly bolted onto the
generalist you already pay for as Claude Cowork or ChatGPT's agent mode.
That is the durable map, and it will still be true when this month's model names have all
changed. The specific prices below move constantly, which is exactly why they live in a dated
table you can scan and not in the teaching. Read the roles once; check the table whenever you
are about to pay for something.
Go deeper: the open-source and self-hosted layer
Below the tools with friendly apps sits a developer layer: open-source agents you run
yourself, like Nous Research's Hermes Agent, a real project with persistent memory and
dozens of tools. It is genuinely capable and genuinely not for a non-technical owner yet,
because it runs from a terminal, not a tab.
Why put it on your radar anyway: the capabilities that debut in the developer layer tend
to arrive in the friendly apps six to twelve months later. Watching it is how you see what
your generalist will be able to do next, without having to install anything.
The tools, by role · entry prices as of August 2026 · always confirm on the vendor's page
Tool
Role
Free tier
Where paid starts
Claude
Generalist
Yes
Pro $20/mo ($17 annual); Max from $100/mo
ChatGPT
Generalist
Yes
Go $8/mo; Plus $20/mo; Pro $100–$200/mo
Gemini
Generalist
Yes
AI Plus $4.99/mo; AI Pro $19.99/mo; Ultra $99.99–$199.99/mo
Perplexity
Specialist · research
Yes
Pro $20/mo
Midjourney
Specialist · images
No
Basic $10/mo up to Mega $120/mo
Gamma
Specialist · decks
Yes
Plus $10/mo; Pro $20/mo
Core concept 02
Pick by job, not by hype. The assistant is the hub, everyone else is a spoke.
One of these tools should be your home base: the place you open first, where your instructions and
files live, and where most of a normal day gets handled. That's Claude, ChatGPT, or Gemini, your
choice, and the choice matters less than picking one and sticking with it. Everything else is a
spoke you visit for one job it does better than the hub: Perplexity when the answer needs current,
checkable sources; Midjourney when the deliverable is an image; Gamma when the deliverable is a
deck someone else will look at.
Good engineers don't reach for the most powerful tool on the shelf, they reach for the right one.
Asking Midjourney to write your invoice email or asking Claude to render a photorealistic product
shot both work badly for the same reason: wrong tool for the job. The skill worth building here
isn't mastering eight interfaces, it's the two-second habit of asking "what is this task, really"
before you open a tab.
Price the tool against your hour, not against zero.
Six free tabs feel thrifty and quietly cost more than two paid tools. Free tiers throttle you
mid-task, forget your instructions between sessions, and drop you to a weaker model exactly
when the work gets hard. The honest comparison for a paid tool is never "$20 versus free." It
is "$20 versus the hours it hands back, valued at what your hour is actually worth."
Run it as real math. A maker on ChatGPT Plus ($20) plus Midjourney Basic ($10) spends $30 a
month on listing copy and seasonal mockups that used to eat a weekend each. A consultant on
Claude Pro ($20) plus Perplexity Pro ($20) spends $40 and gets sourced research before every
prospect call. If either setup buys back even two hours a month, and it buys back far more,
the subscription is the cheapest line in the business. Fee-anxiety is aimed at the wrong
number; the expensive thing is the hour you did not reclaim.
Price against your hour · the subscription is the small barCore concept 04
Pick one generalist and go deep. Switching has a real cost.
The instinct with a crowded shelf is to keep sampling. It is the wrong instinct for the
generalist, because your leverage there compounds in one place: the standing instructions you
tuned, the files it has learned, the memory it has built, and your own muscle memory for how it
responds. All of that resets to zero when you hop to a different app for the novelty of a new
name. Depth in one generalist beats a shallow lap around three.
This is the same judgment engineers use in Lesson 01, aimed at your toolbox: the powerful new
option is often the wrong option, because the switching cost outweighs the gain. Specialists
are different, swap those freely by job, since you are not building a home there. But the hub
is a commitment on purpose. Choose it deliberately, then let it get to know you, which is the
thing a free-tab scavenger hunt never allows.
Go deeper: what "lock-in" really costs, and what it doesn't
Real switching cost is your accumulated setup, not your data, since a well-written
standing brief is a plain text file you can paste into any generalist in thirty seconds.
Keep your brief and your best examples in a document you own, not trapped in one app's
memory, and the cost of switching drops to re-tuning, not rebuilding.
The practical rule: commit to one hub for depth, but keep the portable version of your
setup outside it. That way the commitment is a choice you can revisit, never a cage.
Which tool for which job · Claude, ChatGPT, Gemini, Perplexity, Midjourney, GammaFresh from the lab
What changed this month
ChatGPT can now finish the whole project, not just draft it. On July 9, OpenAI launched
ChatGPT Work, an agent that takes on a real assignment instead of answering one prompt at a
time. Give it a goal, like "put together the Q3 client deck," and it gathers what it needs from
your connected apps, breaks the job into steps, and keeps working for hours to hand back a
finished doc, spreadsheet, deck, or simple website. It rolled out first on Pro, Enterprise, and
Edu plans, with Plus and Business access following soon, alongside a new ChatGPT for Small
Business program bundling free trials from Shopify, Dropbox, and Intuit.
Claude Cowork left your desktop, so a task can keep running while you sleep. On July 7,
Anthropic moved Cowork onto the web at claude.ai and onto mobile for iOS and Android, in beta.
The real change is under the hood: sessions now run on Anthropic's own servers, so a scheduled
task keeps going even after you close the laptop or lock your phone. It's rolling out first to
Claude Max subscribers, with other plans to follow in the coming weeks.
Midjourney's new default model is a real taste upgrade, and it costs the same as last
month. Version 8.2 became Midjourney's default model on July 24. Midjourney describes the
images as more creative, bold, and polished, with far fewer of the flat, low-quality misses
earlier versions occasionally produced, and the Personalization feature now reads your taste
more accurately once you've rated a decent chunk of your gallery. Every plan from Basic on up
gets the new default automatically, no upgrade and no new charge required.
These show up on every tool's site
before you even sign up. Knowing them makes comparison shopping five minutes instead of an
afternoon.
Multimodal
A model that handles more than text: drop in a photo, a screenshot, or a PDF and it reads
it in the same conversation. Claude, ChatGPT, and Gemini all do this now.
Image generation
Turning a written prompt into a picture. Midjourney's whole job, and a bolted-on feature
inside ChatGPT and Gemini too, usually with fewer controls.
Freemium
A free tier that's a real, usable product, not a 14-day trial, with a paid tier above it
for higher limits. Every tool in this lesson runs this model.
API
How one piece of software talks to a model directly, priced per token instead of per
month. You won't touch this yourself yet, but it's what runs under Lesson 04's automations.
Model
Which brain is actually answering you: Claude vs. ChatGPT vs. Gemini, and which version
inside each, like Sonnet vs. Opus. The model picker in your settings is worth a look.
Pick your lane, see the exact two tools
that fit your craft and your actual pain, and set both up properly this week. No six-tab
scavenger hunt. A free Claude, ChatGPT, or Gemini account is enough to start; upgrade the piece
that's actually working.
For female founders
I'm the CEO, the marketing team, and the bookkeeper before 9am, which means
the actual building waits for everyone else to log off.
Claude for the brain, Perplexity for the ground truth
Tools: Claude Pro ($20/mo) + Perplexity Pro ($20/mo)
Choose the stack: Claude as your daily hub, Perplexity for anything that needs a
real, current, checkable answer, competitors, market size, a founder you're about to meet.
Set up Claude properly: a Project with instructions covering who you are, who you
sell to, and how you sound, so drafts start close instead of generic.
Run one real pain task: your next investor update or competitive scan, start to
finish, Perplexity for the facts, Claude for the writing and the framing.
Decide the split: both stay paid if you're fundraising or shipping weekly; drop
Perplexity to free the weeks research isn't the bottleneck.
Calendar the 30-day check: one recurring reminder to ask "did this pair earn its
$40 this month," and adjust without guilt.
The payoff: one hub, one ground-truth tool, and a founder who isn't
running six subscriptions to avoid running one business.
For artists
I make the work, and then grant applications and artist statements quietly
eat the studio hours the work actually needed.
Gemini for the writing, Perplexity for the opportunities
Tools: Gemini AI Pro ($19.99/mo) + Perplexity Pro ($20/mo)
Choose the stack: Gemini as your hub for statements and applications (it plugs
neatly into Docs if that's already your workflow), Perplexity to actually find and vet
open calls, residencies, and grants with real deadlines.
Set up Gemini properly: save your voice, your CV, and a plain-language paragraph
about your practice as a saved instruction or reusable prompt.
Run one real pain task: the application sitting in your tabs right now, sourced
and drafted in one sitting instead of a lost studio day.
Decide the split: Perplexity Pro during application season, back to free between
cycles; Gemini stays paid if you're writing weekly.
Calendar the 30-day check: a recurring reminder tied to your next deadline, not
the calendar date, so the check actually lands when it matters.
The payoff: applications stop being a lost studio day and become an
edit session with real deadlines already found for you.
For actors
Between the survival job and the self-tapes, the paperwork of the career
crowds out the actual acting.
ChatGPT for the drafts, Perplexity for the breakdown research
Tools: ChatGPT Plus ($20/mo) + Perplexity Pro ($20/mo)
Choose the stack: ChatGPT for cover notes, slate scripts, and scene prep, Perplexity
for looking up the casting director, the project, and past work before you submit.
Set up ChatGPT properly: a custom GPT with your resume, your bio in three lengths,
and instructions to write warm and specific, never desperate.
Run one real pain task: the next submission end to end, Perplexity for context on
the project, ChatGPT for the tailored cover note and credit to lead with.
Decide the split: ChatGPT Plus stays on if you're submitting weekly; Perplexity can
flex to free between audition seasons.
Calendar the 30-day check: review what you booked against what these tools touched,
and keep only what's earning its seat.
The payoff: submissions in minutes instead of an evening, with real
context on who's on the other side of the breakdown.
For musicians
Nobody hired me to be my own manager, but I'm the one booking gigs, chasing
deposits, and answering wedding inquiries at midnight.
ChatGPT for the words, Gamma for the press kit
Tools: ChatGPT Plus ($20/mo) + Gamma Plus ($10/mo)
Choose the stack: ChatGPT for pitch emails, captions, and inquiry replies in your
voice, Gamma for the EPK and one-sheet you keep meaning to redo.
Set up ChatGPT properly: a custom GPT loaded with your bio, past press, and the
five details fans always ask about.
Run one real pain task: rebuild your EPK in Gamma from what ChatGPT already knows
about you, then send it to the one venue that's been waiting.
Decide the split: Gamma Plus for release season and EPK refreshes, downgrade to
free the rest of the year; ChatGPT stays on if inquiries are weekly.
Calendar the 30-day check: after your next release cycle, note what got used and
cut what didn't.
The payoff: release week promo drops from days to an afternoon, and
the EPK finally matches the music.
For fitness pros
By the time class lets out I still owe posts, DMs, and the renewal texts to
three people who went quiet.
Claude for the programming, Midjourney for the studio graphics
Tools: Claude Pro ($20/mo) + Midjourney Basic ($10/mo)
Choose the stack: Claude for programming, check-ins, and content drafts, Midjourney
for the challenge graphics and studio flyers you'd otherwise pay a designer for.
Set up Claude properly: a Project with your method, movement standards, and
check-in style, instructed to never invent exercises you don't coach.
Run one real pain task: this week's programming plus one Midjourney graphic for
the challenge you keep meaning to launch.
Decide the split: Midjourney Basic is enough for occasional graphics; upgrade only
if you're running a launch every month.
Calendar the 30-day check: Sunday programming day is a built-in reminder, use it.
The payoff: Sunday admin shrinks to one focused hour, and the studio
finally looks as sharp as the coaching.
For coaches
I sell transformation, and then my actual week disappears into scheduling,
notes, and the follow-ups I keep meaning to send.
Gemini for the day-to-day, Gamma for the program decks
Tools: Gemini AI Pro ($19.99/mo) + Gamma Plus ($10/mo)
Choose the stack: Gemini for recaps and follow-ups (handy if your calendar and
docs already live in Google), Gamma for the program overview deck new clients ask for.
Set up Gemini properly: feed it your niche, your method, and two pieces of content
that actually performed with your audience.
Run one real pain task: build the client onboarding deck in Gamma from the program
outline Gemini already helped you write.
Decide the split: Gamma only needs to be paid the months you're actually enrolling;
Gemini stays on for the weekly writing.
Calendar the 30-day check: tie it to your enrollment cycle so the review lands
when you can actually act on it.
The payoff: marketing and onboarding happen every week, in the margins,
instead of never, in the someday.
For therapists
I trained to help people, and somewhere along the way I became the office
manager too.
Claude for the practice, Gamma for the referral kit (with a bright red line)
Tools: Claude Pro ($20/mo) + Gamma Plus ($10/mo)
Draw the line first: no client information goes into either tool, period. This
stack handles the practice, never the people. (Clinical notes need a HIPAA-compliant tool
with a signed BAA, that's a different aisle entirely.)
Choose the stack: Claude for FAQ replies, waitlist emails, and directory profile
drafts, Gamma for the referral one-pager and any CEU workshop slides.
Set up Claude properly: a Project with your modalities, fees, policies, and a
warm-but-clear tone, nothing client-specific ever.
Run one real pain task: the referral one-pager for colleagues, drafted in Gamma
from copy Claude already wrote about your practice.
Decide the split and calendar the check: both tools can flex to free between
active projects; review in 30 days whether the red line held and the admin actually shrank.
The payoff: the admin hour between clients goes back to notes, lunch, or
nothing at all, and the privacy line stays intact.
For authors
I wanted to write the next book. Instead I run launches, newsletters, and
reader email like an accidental small publisher.
Gemini for the platform writing, Perplexity for the pitch research
Tools: Gemini AI Pro ($19.99/mo) + Perplexity Pro ($20/mo)
Choose the stack: Gemini for newsletters, synopsis variants, and event copy,
Perplexity to find and vet podcasts, book clubs, and comp titles worth pitching.
Set up Gemini properly: load your synopsis, a sample chapter, reviews, and the
way you actually talk to readers.
Run one real pain task: five real podcast pitch targets from Perplexity, five
tailored pitch emails from Gemini, sent by Friday.
Decide the split: Perplexity earns its keep during launch windows; let it lapse
to free between books if pitching goes quiet.
Calendar the 30-day check: count replies, not sends, and adjust the stack to
whichever tool is actually moving the needle.
The payoff: platform upkeep drops to an hour a week, aimed at outlets
that were actually worth pitching.
For chefs
Menus, allergy notes, and quotes all live in my head and my texts until the
day something finally slips.
Gemini for the quotes, Midjourney for the menu art
Tools: Gemini AI Pro ($19.99/mo) + Midjourney Basic ($10/mo)
Choose the stack: Gemini for menus, catering quotes, and specials posts in your
voice, Midjourney for seasonal menu cards and event flyer art.
Set up Gemini properly: your signature dishes, sourcing philosophy, price points,
and per-head costs, once, so quotes assemble themselves after.
Run one real pain task: your next seasonal menu, quote email from Gemini, printable
menu card art from Midjourney.
Decide the split: Midjourney Basic is plenty for occasional menu art; Gemini stays
on if you're quoting weekly.
Calendar the 30-day check: after the season turns, check whether the specials bank
is actually saving you the after-midnight shift.
The payoff: the after-midnight admin shift gets cut from the schedule,
and the menu still sounds like you.
For consultants
Billable work eats the whole day, so my own pipeline, proposals, and
invoices always get pushed to tomorrow.
Claude for the delivery, Perplexity for the research
Tools: Claude Pro ($20/mo) + Perplexity Pro ($20/mo)
Choose the stack: Claude for proposals, recaps, and thought-leadership drafts,
Perplexity for pre-call research on the prospect, their market, and their competitors.
Set up Claude properly: your positioning, the three problems you solve, and a
winning proposal, so drafts start from your actual track record.
Run one real pain task: your next discovery call, Perplexity research beforehand,
Claude recap and next-step draft right after, before the coffee's cold.
Decide the split: both stay paid while you're actively prospecting; if you're
heads-down in delivery for a stretch, let Perplexity lapse.
Calendar the 30-day check: a recurring Friday pipeline review is the natural
moment to ask if the $40 is earning itself back.
The payoff: the pipeline stays warm while you're billing, which is the
whole trick of consulting.
For health advocates
Every case is a crisis and the paperwork never stops, so growing the
practice always loses to today's emergency.
ChatGPT for the templates, Gamma for the community decks (privacy first)
Tools: ChatGPT Plus ($20/mo) + Gamma Plus ($10/mo)
Set the privacy rule first: no client health information in either tool, ever.
Work in templates and blanks; the person's details get added by you, after, in your own
documents.
Choose the stack: ChatGPT for appeal-letter and intake-letter templates with
[BRACKETS] where the client goes, Gamma for workshop and community-education decks.
Set up ChatGPT properly: your service descriptions and de-identified skeletons of
your best letters, nothing case-specific.
Run one real pain task: one new template this week (appeal, records request, or
intake), plus one workshop deck families actually need.
Decide the split and calendar the check: Gamma can flex to free between workshops;
review in 30 days whether the template library actually grew.
The payoff: the paperwork mountain becomes a fill-in-the-blanks library,
and client hours go to clients.
For makers
I make it, list it, ship it, answer every message about it, and somehow run
out of time to actually make more of it.
ChatGPT for the listings, Midjourney for the seasonal mockups
Tools: ChatGPT Plus ($20/mo) + Midjourney Standard ($30/mo)
Choose the stack: ChatGPT for listings, wholesale pitches, and market applications,
Midjourney for seasonal collection mockups before you commit materials.
Set up ChatGPT properly: your materials, process, price ranges, and three of your
best existing listings as the voice reference.
Run one real pain task: mock up next season's collection in Midjourney, then write
the launch listings in ChatGPT from the same concepts.
Decide the split: Midjourney Standard earns its 15 fast hours around seasonal
drops; drop to Basic or pause between collections.
Calendar the 30-day check: line it up with your restock cadence so the review is
also a planning session.
The payoff: listing night becomes listing hour, and the studio gets
its maker back.
For nonprofit leaders
I'm the grant writer, the thank-you-note writer, and the program, all at
once, and the mission is stuck behind my inbox.
ChatGPT for the writing, Perplexity for finding funders
Tools: ChatGPT Plus ($20/mo) + Perplexity Pro ($20/mo)
Choose the stack: ChatGPT for grant boilerplate and donor communications,
Perplexity for finding funders whose actual guidelines match your mission.
Set up ChatGPT properly: your mission and vision, programs, impact numbers, and
one funded proposal as the template.
Run one real pain task: three new prospective funders from Perplexity, one tailored
LOI per funder drafted in ChatGPT the same afternoon.
Decide the split: keep Perplexity paid during active prospecting seasons, let it
lapse when you're mid-cycle on existing grants.
Calendar the 30-day check: tie it to your grant calendar so the review doubles as
deadline planning.
The payoff: development stops eating the mission's hours, and funders
still hear a human.
For photographers
I shoot weekends, edit all week, and the inquiries and gallery follow-ups
fall through whatever cracks are left.
Claude for inquiries, Gamma for the client pricing guide
Tools: Claude Pro ($20/mo) + Gamma Plus ($10/mo)
Choose the stack: Claude for same-day inquiry replies and gallery delivery emails,
Gamma for a real pricing and package guide clients can click through instead of a PDF.
Set up Claude properly: your packages, pricing, turnaround, and three inquiry
replies that felt right as the voice reference.
Run one real pain task: build the pricing guide in Gamma from what Claude already
knows about your packages, then send it on your next inquiry.
Decide the split: Gamma stays paid once the guide exists and needs updates each
season; Claude runs daily either way.
Calendar the 30-day check: compare inquiry-to-booking rate before and after the
guide went live.
The payoff: inquiry response goes same-day without a VA, and the
pricing guide closes bookings while you're behind the lens.
For realtors
My old clients are a graveyard in my CRM, and every new listing turns into
a scramble of copy and coordination.
Gemini for the nurture, Midjourney for the listing graphics
Tools: Gemini AI Pro ($19.99/mo) + Midjourney Basic ($10/mo)
Choose the stack: Gemini for listing descriptions and nurture emails (a natural
fit if your inbox and calendar already run on Google), Midjourney for just-listed and
open-house social graphics.
Set up Gemini properly: your farm area, niche, recent sales, and the way you
actually talk to clients, texts included.
Run one real pain task: segment your database into three groups, draft one
re-connection email per group, pair with one Midjourney graphic for the next listing.
Decide the split: Midjourney Basic covers occasional graphics; upgrade only during
a heavy listing season.
Calendar the 30-day check: a monthly planning hour is the natural moment, tie the
review to it.
The payoff: the graveyard becomes a referral engine, on one planning
hour a month.
For stylists
I'm behind the chair all day, so rebooking texts and content pile up, and
the clients I forget just don't come back.
ChatGPT for the rebooking texts, Midjourney for the mood boards
Tools: ChatGPT Plus ($20/mo) + Midjourney Basic ($10/mo)
Choose the stack: ChatGPT for rebooking nudges and captions between clients,
Midjourney for mood boards you can show a client before you touch a single strand.
Set up ChatGPT properly: your services, prices, rebooking windows, and the
products you actually believe in.
Run one real pain task: this week's rebooking list, nudge texts from ChatGPT, plus
one Midjourney mood board for tomorrow's consult.
Decide the split: Midjourney Basic's 3.3 fast hours cover a mood board or two a
week comfortably; ChatGPT runs daily either way.
Calendar the 30-day check: your day off is the natural review moment, since it's
the thing this stack is trying to protect.
The payoff: the day off goes back to being a day off, and the chair
stays booked.
The house rule
Lauren's line, applied to the toolbox: what belongs to you and keeps your humanity, you keep.
What frees you to be more human away from your laptop, you hand off, to one assistant and one
specialist, not six. More tabs is not more power. It's more logins between you and the work.