Founder, P33 | Executive Thought Leadership Agency

Helped 50+ B2B tech companies build their Founder/CEO Brand and implement Executive Thought Leadership Programs for LinkedIn. Customers, results & process: project33.io

About me:

I'm Finn, 29, grew up in Bremen, Germany, studied Physics & Philosophy in Heidelberg for 3 semesters, dropped out of college in 2017 because I was doing it for other people, flew to New York with no plans, documented the journey on Linkedin, Instagram and Youtube, and started my first company (Project 33) on the back of having grown a small audience. We're a small remote agency that helps B2B founders & execs create thought leadership content for Linkedin.

In my free time, which I try to have as much as possible of, I read (fav author: Nassim N. Taleb), write, run, lift, code (currently building an app), cook, make Youtube videos, turn books into Instagram accounts, grapple (BJJ), and travel (lived in NYC, Cape Town, Bali, Minsk, CMDX, Lima & Antalya so far). In December 2020 I married the most beautiful woman in the world. In January 2023, I became dad to the most beautiful baby girl. We currently live in Berlin, Germany.

Finn Thormeier

Head of Growth @ Perspective / SaaS & Demand Growth

Niels Klement

Base44 hit $100M ARR.

And just like that, I questioned everything we're doing at Perspective.

These hyper-growth AI stories are everywhere right now. And every time I see one, there's this little voice going "are we doing something wrong?"

The reality is that Perspective has never had a crazy viral moment. Everything we've built to $10M+ ARR has been the result of having clear goals, going after them, and applying hard work and consistency over years.

And I think if you're in a similar position where you read these headlines and think "what the hell am I doing wrong?" here's what I remind myself of:

It's a very, very small percentage of companies that ever hit that type of escape velocity.

For the rest of us, it's perfectly fine to have a non-viral product.

To just keep showing up, keep building, keep improving, and still end up with a beautiful business that lets you live an amazing life working on things you thoroughly enjoy.

Oh well, maybe we'll still hit that escape velocity someday. Who knows, dreams die last.

Let me get this straight:

I used Sunday afternoon to build an app that saves us 4 to 6 hours every single week in media-buying ops

I shipped CRO tests on hardcoded checkouts by speaking to Wispr Flow.

I launched new funnels with Perspective AI duplication in 10 minutes

I built a campaign sheet once, and Claude drafts extremely on-point ad creatives, email sequences, SMS sequences, video scripts, VSLs, webinars, and literally one-tenth of the time

The golden days are here

I'm thinking of marketers with a Claude sub like Tony Stark, who opens up Claude Code and Wispr Flow in his garage and just creates a million-dollar campaign in a week

Everyone can do this now, which makes it even more significant for 10x marketers

lfg

I recently gave a detailed breakdown of how we spend $3.1M on ads (profitably) and ship 1,681 creatives a year with a team of three to a room of agency owners who paid to be there.

Basically the full creative system of how we run performance marketing at Perspective.

(Honestly might have shared too much, but here we are)

Here's what's in it:

- The one-word change in an ad script that tripled our scale overnight

- How we shipped 1,681 unique ad creatives this year with a team of three

- The iteration that turned a dying ad into our highest spender ever at $160K

- The exact 3-person team structure and 45-minute weekly meeting that runs the whole operation

- CBO testing vs ABO and why we switched after Andromeda

This is probably the most detailed breakdown I've ever done publicly. If you run ads or plan to, I think you'll get a lot out of it.

Want me to send you the recording?

Simply comment "creative" (we need to be connected) so I can send it over (no optin necessary).

🫶

My wild 2026 prediction:

Marketing will receive the budgets that engineering held for the last couple of years.

Think about it.

When fewer engineers can ship more, the next best dollar goes to distribution.

Am I crazy for thinking this?

I spent a weekend at the notus mastermind with some of the smartest organic marketers I know. My notes were endless. Here are my 6 core takeaways:

1) On outbound: We generate thousands of leads every month through paid acquisition and funnels. Almost all of them are left to self-serve the product. Nobody picks up the phone. Korbinian and I are building a sales layer on top of what already works. The leads are there. We just haven't been talking to them. This is what is going to turn organic content into revenue, plus I’m certain this will be a huge unlock for our paid lead flow as well (will report back).

2) On direct response: Our obsession with performance marketing (marketing that drives immediate growth and sales) is one of the key reasons we got past $10M ARR bootstrapped. I'm grateful we made that bet, but I’m convinced that one of the next step changes will be produced by bringing that same intensity to organic execution.

3) On content: Marvin showed me how his team thinks about content. It's exactly how we think about paid ads. Test fast, kill what doesn't work, double down on what does. His team publishes, measures, and iterates. If you do it right, organic can be just as performance-driven as most paid ads.

4) On team structure: We had a paid & creative team, and a content team. Now, I’m thinking about merging them into one social team. If the best organic content needs a performance mindset, and the best paid creative needs genuine consumption, running two separate teams doesn't make sense anymore. Hormozi said the same thing, so definitely worth exploring at Perspective as well.

5) On partnerships: I sat down with Suhit, who runs influencer and creator marketing for a $325M company with his team. Turns out they've been running the same playbook we've been building at Perspective, just at a completely different scale. Validating and humbling at the same time. Excited to push this channel as far as it can go with Alex.

6) On judgment: AI lets small teams ship at the volume of large ones. Doesn't mean the work is good. Everyone will have the same tools. The gap will be the quality of the humans operating them, and the proprietary data and pattern recognition they've built from years of actually doing the work.

---

Cool, that's it. Thanks for reading, and thanks to the notus team for having me. Sick weekend.

We changed one word in an ad script. The exact same creative started reaching 3x more people overnight (while performance stayed consistent):

The creative had been plateauing, solid conversion rate, but Meta was running out of people to show it to.

We'd been using "funnels" throughout the script. On a hunch, we swapped it for "landing pages."

(fyi, under Andromeda, Meta reads your creative deeply and uses every word to estimate who should see the ad. So the vocabulary you choose actively defines your audience)

You can see this in the hooks we tested side by side:

- "If you're using these tools to build quiz funnels for your clients..." reaches marketers already deep in that world, many of whom already knew us.

- "Stop wasting days building landing pages..." is how the broader market describes the exact same problem.

Same offer, completely different pool of people.

We now audit every script for this.

"Are we using language only our existing audience speaks?" vs "Are we speaking the language of the bigger market we're trying to reach?"

If your ads are plateauing and you've already tried new hooks, new formats, new targeting, look at the specific terms in your script. You might be narrowing your reach without realizing it. 🫶

Tell me I'm wrong?

In 2026, AI moves the marketing bottleneck from execution speed to the quality of what you know that Google doesn't.

Marketers are about to use tools like Claude Code to ship entire campaigns.

Different MCPs, same speed, same access.

So when everyone has the tools, who will be the ones that win and create the best outcome?

> General knowledge gets commoditized.
> Specific knowledge becomes the entire game.

The marketer with the unique insight (= the data, context on the market, taste and judgment, high agency) takes the cake.

This is the time to shine for 10x marketers. The gap between good and bad is going to spread because your ability to create something unique is heavily dependent your unique judgment.

That's why we have all hands on deck at Perspective:

Make the data (more than 10 million conversions a year) available to you guys to use in your marketing. More to come soon.

Spending $300 to decide if an ad works is the old playbook. We launch 20-40 new ads a week and let most of them get zero spend on purpose.

This came up on our last mastermind call and I think a lot of people running ads are still getting this wrong.

A founder was running $50/day, two ads in an ad set, waiting five days to hit his 2x target CPA threshold before making a call. Classic approach, nothing wrong with it in theory.

But after Andromeda, that's painfully slow.

Here's what we actually do at $250k/month in spend:

We load 20-40 fresh creatives into a CBO every week. One campaign, one ad set, all the ads in there.

Meta doesn't evenly distribute your budget across all of them (it never did, but now it's way more aggressive about it).

So out of those 40 creatives, maybe 4-5 get meaningful spend. The rest get almost nothing.

And that's the point.

Meta is pre-selecting winners before you ever have enough data to make a rational call yourself. It's mapping every creative against its prediction models, figuring out which ones are most likely to get you the result you're optimizing for, and funneling budget there.

So the ones that get zero spend, Meta already told you those won't be winners.

“But Niels, didn't you just waste time on 35 ads?”

No. You gave Meta 5x the creative surface to find winners. You just compressed what used to take five weeks of sequential testing into one week of parallel testing.

The old way: test 2 ads, wait a week, kill the loser, test 2 more.
The new way: test 40, let Meta pick 4-5, double down on those, launch the next batch.

Same budget with a wildly different velocity.

In the last 12 months, we spent $3.1M on ads.

That will turn into $12.3M over the average lifetime of those customers. The team behind it is three people.

I wrote up every piece of how we do this.

The creative system, the testing structure, the iteration playbook, the team we built (and in what order), and why we run 2-minute video ads when everyone says that's insane.

Some of what's in there:

- We shipped 1,681 ads this year. Only 129 produced any value. Only 20 were responsible for the majority of our growth. The entire system is built around those odds.

- Why we killed ABO testing and moved everything into one CBO campaign (and how it cut our testing waste in half).

- The single highest-leverage iteration we do: remixing hooks with bodies that don't even match. One remix took a $60K creative and turned it into a $160K creatives.

- How one word change in a script tripled the scale of our best-performing ad. Andromeda reads your script and decides who sees it based on the words you use.

- The 4-phase hiring order we followed to go from me-doing-everything to a team that ships 30-40 ads per week.

If you want the full breakdown read (and save this article):

In the last 12 months, we spent $3.1M on advertising. That will yield us $12.3M over the average lifetime of those customers. The team behind it is th

Open article: How to Build an 8-Figure Creative Engine With a Team of Three by Niels Klement • 9 min read

How to Build an 8-Figure Creative Engine With a Team of Three

After $3.1M in ad spend, the one creative format that builds brand, scales longest, and converts hardest is the same:

A 2+ minute educational video.

And yet, every platform tells you to keep it short (60-90 seconds max). For us, we've found the opposite.

A majority of our active ad account right now is video. And those videos are two, two and a half minutes long running in reels and stories.

Why this works (specifically after Andromeda):

1/ Signal density. A video gives the algorithm so much more data to optimize on compared to a static. Watch time, drop-off points, every word in the script is a new signal for reaching the right audience. That's why our videos consistently outlast and outscale statics in the account.

2/ Specificity as targeting. We use very deliberate language in scripts. "Ad agency owners," "retainers," "clients," terms that signal exactly who should see this ad. Under Andromeda, these words do more targeting work than your audience settings.

3/ Education builds brand in-feed. A 2-minute video that actually teaches something gets consumed in the feed. That means even people who don't click still walk away knowing who you are and what you do. We do zero brand campaigns. Our brand search is growing 6x faster than organic. All driven by ad creatives.

4/ Conviction converts. Because the video has time to educate, show proof, and make the offer, people who do click are already pre-sold. Higher quality leads, lower cost per qualified conversion.

One of these videos ran for five months, spent €160K, and accounted for €557K in conversion value.

We call this "brand performance" internally. Direct response principles (offer stacks, urgency, social proof) combined with genuinely valuable educational content.

Performance marketing that also builds the brand. One format, both outcomes.

Am I crazy for thinking most B2B companies are massively underinvesting in long-form video ads?

A profitable consulting firm asked me where to start with ads.

They had big-name clients, but zero experience with paid acquisition.

Naturally, they wanted to promote a white paper to generate leads. I told them that's the fastest way to waste your first $5,000 on ads.

When you run a white paper funnel, you're saying "hey, here's six free pages on how to implement AI into your business"

You'll get downloads from…

- people who want free information
- competitors doing research
- junior employees curious about AI

Maybe you get some decision-makers, but mostly people who are very far from spending $15-25K on your consulting.

And now you have a list of hundreds of leads that you need to somehow nurture into a sales conversation.

That's months of email sequences, follow-ups, content, all before a single dollar comes in (if you're lucky)

An appointment funnel is a completely different game.

You run an ad that speaks directly to your ICP.

"Manufacturing company with 100+ employees? Learn how to cut operational costs with AI in a free 30-minute consultation."

The people who book that call are already interested enough to give you their time. They have a problem they want solved. They're self-selecting as buyers.

Less volume, 100%, but every lead actually means something.

And if your sales team closes at 30-35% on calls, you can reverse-engineer exactly what a booked call is worth and scale from there.

The white paper route sounds safer because it's lower friction but when you're just starting with ads, you need to know, "Can I acquire customers profitably through this channel?" as fast as possible.

Start with the appointment funnel and get that working first

P.S. If you want a proven "Book a Call" funnel you can set up in an hour, like this post, comment "call" and I'll send you our template 🫡

We've been spending €250k/month on Meta ads for two years, and there's one thing I tell every founder who starts obsessing over their dashboard:

Meta's numbers are useful directionally, they tell you roughly what's working, but the moment you start making big budget decisions based on them alone, you're optimizing for a simulation.

We've had creatives Meta flagged as our best performers ever, pulled them into a third-party attribution tool, and they were driving basically nothing.

A simple question fixes this:

"How much did I spend? How much did I get back?"

That's your ground truth, keep it simple.  

The next layer is MTA data, or what your leads actually reported as their source, which is way closer to reality than anything inside Meta's dashboard.

"Should we start with a lead magnet or go straight to booking calls?"

This question comes up often and if you choose wrong you're almost certainly going to waste your first $5,000 on ads.

Talked through this with a consulting firm on our mastermind.

Profitable business, big-name clients, selling $15-25K engagements but zero ad experience.

Their instinct was to start soft:

- promote a white paper
- build a list
- nurture over time
- eventually sell

I get it - it feels less risky, but you're paying for leads and then hoping your email sequence does the heavy lifting over the next few months.

(The gap between "downloaded your PDF" and "ready to spend $15K" is massive)

With an appointment funnel you speak directly to your ICP, lead with the value of the call itself, and let people self-select. The ones who book are already problem-aware and interested enough to give you 30 minutes.

Less volume, but every lead is actually worth following up on.

And you get signal fast.

Within a few weeks you know your cost per booked call, your close rate, your customer acquisition cost. You can actually answer "do ads work for my business?" instead of staring at a list of PDF downloads wondering what to do with them.

So a book-a-call funnel definitely wins here imo.

P.S. If you want a proven "Book a Call" funnel you can set up in an hour, comment "call" and I'll send you our template 🫡

I've been thinking a lot about how AI will change marketing. Here are the 5 predictions that actually matter (some of them scare me):

1/ Hiring will shift towards marketing and creative

When everyone can build and implement technical things quickly with AI, the people who understand messaging, audience, and taste become the most valuable on the team.

AI can handle marketing operations, but the strategists who actually know what will resonate with a specific audience is about to crush it.

2/ Opinion becomes the edge

Knowledge has been commoditized since Google and AI makes it even more accessible.

But markets have always been drawn to individuals and companies with a clear point of view on how things should be done.

As knowledge gets cheaper, experience, proof of work, and strong opinions get more expensive.

3/ The top 1% becomes the top 0.1%

The barrier to entry drops, so the long tail grows (= more players, content, noise) but the share of attention concentrates even further at the top.

The internet already did this, AI will accelerate it even more.

4/ Only three things are truly defensible

When the smartest knowledge worker is available through a chat interface for $20/month, what are people still willing to pay for?

- Judgment: unique opinions and specific experience that people trust
- Data: proprietary datasets that help AI create better outcomes
- Infrastructure: the technical stuff that's still genuinely hard to build and maintain (hosting, page speed, reliability)

Everything else gets commoditized.

5/ The most authentic businesses win

We're experimenting with AI content and AI video at Perspective. But the businesses that will continue to get the most attention are the ones with a genuinely unique voice.

When feeds get flooded with AI-generated content (and they will), people will gravitate even harder towards voices they trust and faces they recognize.

---

If I zoom out on all of this, the pattern is pretty clear:

AI makes knowledge and execution free. What's left is taste, judgment, and the willingness to put yourself out there with a real opinion.

That actually makes me more confident in what we're building, not less.

lfg

P.S. Tell me if you disagree and think I'm wrong 🙏

Company email verification is now live in Perspective. 🥳

Here's why I'm implementing it immediately:

Lead quality is one of those things you continuously optimize for, and every time you do, you realize you should've done it sooner. The front-end impact is massive, and it all comes down to show-up rate.

The math is simple:

At 20% show rate, a €20 CPL costs you €100 per shown appointment. At 50% show rate, you could pay €50 CPL and land in the exact same spot.

That's why we built the verification toggle too.

You control whether a conversion registers when the lead is verified or not, so you're only counting the leads that actually matter.

I'm dropping this into one of our new flagship funnels next month (I already know it'll move the needle).

If you've been asking for this, it's live in your account now. 🫶

“We tried ads. Didn't work. Spent €200, got nothing.”

Me: “Okay, what's a customer worth to you?”

“About €500.”

“So you spent less than half of one customer's value and expected to validate an entire channel?”

“…”

Look, if you've never run ads before this might feel uncomfortable but if your customer is worth €500, you need to spend at least €500 to see if you can acquire one profitably.

(If not, you iterate, but don’t throw away the channel)

Anything less is just hoping to get lucky.

Ideally you're able to do this over 3 months to actually figure out what's working, fix what isn't and give the algorithm time to do its thing.

And if you’re able to go through the uncomfortable part of early experimentation, then you might reach the beautiful state where you have unlocked one of the most scalable, predictable channels to grow your business.

"Pausing a Meta campaign resets the algorithm and kills performance."

Algorithmically, there's something to it.

But in practice, the opposite is true often enough that the fear is overblown.

We've paused campaigns that had been running for months, strong performers with good history, reactivated them weeks later, and seen better performance than before.

(Obviously not always but often enough)

The real cost of not pausing is almost always higher than the algorithmic reset.
i.e., if you're away for three weeks and nobody is following up on leads, you're burning money on people you can't convert.

That's a guaranteed loss.

Pause when you're away and leads won't be nurtured, when you're at capacity, when you need to focus on delivery.

So yes, the algorithmic hit is real but it also recovers faster than you think.

P.S. Please let me know if you like these 'day to day' ad insights 🫶

I wrote a disclaimer for every new hire:

"I will be your toughest critic. I won't always be nice, but I'll be fair."

It's one of three things that changed how we build the team at Perspective. All three sound simple but took me way too long to figure out.

1/ I stopped hiring out of urgency

When you're overworked and the team is stretched thin, you make calls on people when you're still in doubt. I've done this multiple times. Now I only make the call when there's true conviction someone is a strong fit. Even if that means staying short-staffed longer.

2/ Freelance testing before any full-time offer

Work samples are easy to game. A 3-hour assignment can't test reliability, speed, or how someone solves problems that haven't been figured out yet. So everyone works with us for at least a month in freelance first.

3/ More hard conversations

I used to lean towards harmony. If feedback might be uncomfortable, I'd soften it, delay it, or skip it entirely. Now whenever I notice that feeling of "this could be a hard conversation," I take it as the signal to have it.

The disclaimer helps with this. When someone knows on day one that I'll be direct with them, there's no surprise when tough feedback comes. It's just how we operate.

These three changes are why we went from struggling to build the team to having the strongest one we've ever had.

P.S. We're hiring 🫶

Nobody really knows how business will look in three years.

Imo that's the best possible news for founders rn:

AI is fully rewriting the rules.

Not just for how we market or how we hire, but for what products get built, how companies operate, what skills actually matter.

At Perspective, we're living this right now. Fundamentally changing how we work, what we build, how we market. It's a new landscape we're learning to operate in real time.

The scary part:

Anything you've built up until now, any knowledge you've acquired, its value is trending toward zero. The landscape is just too different going forward.

Yet, someone is always rewriting the rules.

I truly believe high-agency people (founders, operators, people who figure things out fast) are exactly the ones who do.

lfg.

Founder & CEO @perspective | $0 to $10M+ ARR in 4 years bootstrapped & fully remote | We help you build your highly scalable customer acquisition & recruiting funnel

Nobody wants to admit they're scared of AI.

Especially founders. Especially publicly. So let me go first.

I've been building companies for a long time. I cofounded all of them with my best friend Marco. I've been through a lot of shifts. But nothing has felt like what's happened in the last few weeks.

There were days where I just didn't want it.

I avoided X because every AI post made me anxious. I felt resentment. Like, genuine resentment. I didn't want this change to happen.

I kept thinking about what AI means for our company, for our team, for our jobs.
What it means for humanity if AI comes for white-collar work and robotics comes for blue-collar.

It was heavy.

And I think it's completely human to feel all of that. The excitement and the fear at the same time.

At some point, I realized the resentment was just making me slower.

And the people who will come out of this stronger are the ones who find their curiosity again, who let AI make their work easier first, and then ride the momentum from there.

Chiara said something to me that stuck:

"Mikey, what if this is the moment we've been waiting for?"

I think she's totally right.

I don't have all the answers (I told the team that directly) but sitting in fear is a choice and building is also a choice.

The low-ticket funnels that actually scale all figured out the same thing: you don't need to be profitable on your $27 product.

An example from our mastermind calls:

Someone built a low-ticket funnel. $27 product, added a $97 upsell, couple of order bumps. They spent $1,500 on ads, got 15 customers, and made back maybe half.

And they're like, "This isn't working. Should I kill it?"

Almost always the answer is no.

You just acquired 15 people who gave you money!

That's a completely different relationship than someone who downloaded a free PDF. People who have paid you before are significantly more likely to pay you again. And the second purchase doesn't need to be $27.

What I suggested: get those 15 customers on a call. Not to upsell them aggressively, just to help them succeed with what they bought. And within that call, offer something bigger. $1,000, $2,000, whatever makes sense.

One of our customers does exactly this in the weight loss space.

They send a crazy amount of traffic to a $47 front-end product (barely breaking even) but it includes an onboarding call.

And on that onboarding call, they sell high-ticket coaching.

The conversion rate from $47 buyers to high-ticket is way higher than anything they've ever seen from cold leads (way higher).

The $47 product is the filter: It finds people willing to spend money, builds trust through a quick win, and then the real offer comes after.

So if you're sitting on a low-ticket funnel that's losing money on the front end, the question is "What are you selling those 15 buyers next?" instead of "How do I squeeze more conversion out of this low ticket sale?"

That's how it works at basically every scale I've seen.

taplio

The Battle

Finn Thormeier

Niels Klement

Followers

Posts

Engagements

Likes

Performance ScoreNiels Klement's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Finn Thormeier

Niels Klement

Followers

Posts

Engagements

Likes

Performance Score.css-goub6s{font-size:var(--chakra-fontSizes-md);font-weight:500;color:var(--chakra-colors-gray-600);display:block;margin-top:var(--chakra-space-1);}Niels Klement's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreNiels Klement's LinkedIn Performance