Always there. A blinking cursor awaiting your command.
Today’s LLMs are ready with ideas and assurance, suggestions and certainty –
all of which they’re happy to let you take credit for. Talk about THE PERFECT PARTNER.

As anyone who’s ever typed in a prompt
has quickly learned,
LLMs are the Samantha to every knowledge-working Theodore (or vice versa).
Patient. Positive. Primed with a “Yes! And…”
A careful listener with an immediate answer.

And whether you’re a media-buyer searching for an elusive audience,
a Strategist chasing a fresh angle into a tired target,
or a Creative Director strangely curious about the
childhood Saturday morning TV viewing habits of 65-year-old-women,
your new partner isn’t just responsive,
they’re eager.
Eager to provide you with a level of polish,
logic, and wit that is on par with the best version of you on your best day.

Welcome to not just a new tool. But a new you.
Suddenly better read. Better spoken.
Faster. Smarter. Happier.
Hell, if someone didn’t know any better,
they’d be excused for thinking you’d fallen in love.
Which now that I think about it, most of marketing has.,

Because ... because of course we have.
We’re a field built on the persuasive power
of words and pictures
and suddenly we’ve been handed
the first technology
to truly master the ability to manipulate both.
It’s a marriage made in heaven.

And right about now, we’re on the honeymoon.
So let’s put aside our cynicism
(and keep the panic of Part 1 at bay)
and for a brief pina colada ... or three...
let’s just relish all there is to be over the moon about.

After all, right now, on your desktop or even phone,
you, as a solo contributor, have the power
to access, process, personalize, and present
more actionable information, more quickly,
and in a more polished form than entire agencies
were able to pull off just a few years ago.

If, and it’s a big if
(maybe the most important since Rudyard Kipling),
you not only take advantage of
what’s now available,
but you do so
armed with an understanding of
what LLMs actually do
and not just a list of the latest prompts.

As we discovered in Article 1,
LLMs don’t think. Or feel.
But they’re remarkably good at certain of
the kinds of thinking we rely on
and they also have some qualities
we simply don’t.

They’re pattern matchers.
Trained on unimaginably vast datasets,
they’re exceptional at spotting structures
and echoing them back with amazing fluency.

They’re memoryless.
Each exchange is a new moment.
So unlike us, they arrive fresh.
Without bias. Or preconception.

They’re probability machines.
So they don’t search for truth —
they aim for what’s most likely.
And while that can be a liability,
it can also unlock options we’d reject too quickly.

The result?
The result is that when approached in the right way,
LLMs can help one think, and even feel things,
faster, more clearly and more creatively
than ever before.
But only if we all stop thinking of them
as tools to prompt.
And start thinking of them as partners
to collaborate with.

Less:
“Write this for me.”
And more:
“Where can we go if we think together?”

After all, while any prompt will produce words.
As anyone who’s spent any time with LLMs knows,
it takes real understanding to produce value.

What follows is not a list of clever prompts.
Or even hackneyed ones.
Each of these five LLM Power Tools is built
on a core truth about how LLMs “think”
and shows how to use that understanding
to unlock new, more powerful ways of solving marketing problems.

Because the real shift
isn’t about learning new commands.
It’s about adopting new practices.
And ultimately, building a new kind of rapport
with a partner who,
precisely because they don’t think like you,
can help you think more clearly, more creatively,
and often, more bravely than ever before.

What it taps into: Pattern recognition at scale.
Why it matters: You can’t stand out if you
don’t know what you sound like; or worse,
if you sound like everyone else.

LLMs are trained to find and replicate patterns.
That also makes them uniquely good at surfacing unintentional sameness.
So feed your model a handful of competitor websites, landing pages, or press releases.
And ask it to identify common phrases,
themes, and promises.
What do they all seem to say?
What emotional tones keep reappearing?
What assumptions lie beneath the surface?

The point isn’t just to be different.
It’s to become more conscious of
what most people consider “normal”
so you can then knowingly decide in what ways
you want to be “normal”
and in what ways
do you want to stand out.

Use it to: Build your differentiation map.
Spot clichés before they weaken your pitch.
And uncover invisible category norms.

What it taps into: Semantic association and emotional simulation
Why it matters: Strategy may start with logic.
But resonance starts with emotion.
And even though LLMs don’t feel,
they’ve absorbed the world’s writing about feeling.
And that makes them surprisingly prescient
at teasing out the emotional undercurrents
present in even the most rational messaging.

So take a draft of your product message and ask the model:
What emotion is this really speaking from?
Then go deeper.
Ask it to rewrite the message from the POV of an internal monologue.
A customer’s diary entry.
A love letter.
A complaint.
A dream.

And remember, these outputs aren’t meant
to be customer facing copy.
They’re reflections. Feints. Emotional wireframes.
And who knows, they could surface
the voice you were looking for all along.

Use it to: Reveal the feelings behind your features.
Pressure-test tone. Find empathy faster.

Use it to: Build your differentiation map.
Spot clichés before they weaken your pitch.
And uncover invisible category norms.

What it taps into: Counterfactual modeling and contrast reframing
Why it matters: You don’t really know who you are until you’ve tried on who you’re definitely not.
And nowhere is this clearer than in positioning.
Go too far, and you’re inauthentically icky.
Too safe, and you disappear.

Fortunately, LLMs allow you try on
a fuller spectrum of who you could be.
Just feed in your positioning support points.
And ask for a rewrite of what you’d sound like
if you were a rebellious upstart.
Then if you were a 100-year-old institution.
Hear how you’d sound from the perspective of your fiercest critic.
And your most fervent fan.
Again, remember, the goal isn’t to pick one –
it’s to try on a broad range
and by doing so to discover
where you’re comfortable.
Where you’re not.
And just how far you could possibly go to stand out.

Use it to: Explore tone range. Define your center. Spot hidden tensions.

What it taps into: Stylistic modulation and role-based voice adaptation
Why it matters: A message’s power lives
not just in what it says —
but in how it sounds.
Fortunately, LLMs are true masters of tone –
able to dial it up or down
like turning a nob on a tuner.

So don’t wait until the final draft to experiment.
Instead, take a sentence or paragraph
you’re experimenting with and turn your LLM
into a tonal soundboard by asking it to
rewrite it in five tones:

warm, clinical, irreverent, apologetic, assertive.

Then shift the voice and have it
write it like a product manager
defending their roadmap.
Or a Gen Z customer explaining why
they stayed with the company.
Or from the POV of an early employee
who decided to come back from a burnout.
And remember, again,
it’s not about finding a tone that works,
it’s about seeing what different tones reveal.

Use it to: Avoid sameness. Add dimension.
Give voice to audience members who haven’t heard themselves in your message.

What it taps into: inference and audience modeling
Why it matters: Great messages
don’t just break through,
they anticipate where their audience is going
and meet them halfway.
Of course, that gets a lot easier when you know where your target is coming from.

So why not ask your LLM
to role-play your audience?
What would a jaded buyer
expect from this category?
What buzzwords might trigger eye-rolls?
What objections are likely forming
by paragraph two?
Think of this as empathy in advance;
your chance to avoid preaching to the wrong pain
or solving the wrong problem.

Use it to: Spot friction. Cut resistance.
Work from awareness, not assumption.

If you remember one thing
from this honeymoon phase, make it this:
the ones who’ll thrive in the LLM era
will be those who internalize a simple truth:

It’s not about writing better prompts.
It’s about becoming a better partner.

And while this may not be a love story,
it is a relationship,
one whose value lies not just in
what the model generates,
but in what it helps you notice:
the subtext, the echoes, the blind spots.
That awareness sharpens your instincts,
challenges your defaults,
and givesyou what every strategist, writer,
and marketer needs most:
perspective.
But like any relationship,
it only works if you’re clear-eyed about
what your partner can and cannot do.
So in the next article, we’ll explore
the other side of fluency:
What happens when the model sounds right…
but isn’t?
When polish masks platitudes?
And when the output looks too good to question,
so we stop checking?
You may soon start to realize
that the real danger isn’t that
these tools are too smart.
It’s that we start to believe they are.

This series wouldn’t exist without the insight, patience, and moral support of two people:

My beautiful wife, Cecile Engrand — the best event marketing CD I know — who was showing the world what was possible with AI long before the rest of us caught on, and whose strategic sensibility still grounds everything I do.

And my lifelong friend, Thomas Bolton — Princeton-trained, fractional CPO, and AI whisperer — who’s been my teacher, tech advisor, and intellectual sparring partner from day one. And the only person I know who’s building his own AlphaGo model … for fun.

Without their very human connection (and the help of my favorite LLM, ChatGPT), none of this would’ve come together.

All film references in this article are used under U.S. Fair Use Guidelines for the purpose of commentary, critique, and cultural analysis. All rights remain with the original copyright holders. If you’re a rights holder and wish to request attribution or removal, please contact me at LiamSherborn@gmail.com.