If you're interested in creating an incredible product, training cutting-edge models, or building at-scale infrastructure, feel free to connect and send a message.

Archie Hollingsworth

One year with Harry.

Tonight he joined the team to talk about the road ahead.

It’s so early.

Speed is everything.

LFG.

Sub companies within companies. 

This is our approach: to hire all-star engineers, founders, ex-founders, and CTOs who can be the CEO of their area.

Engineering is changing so fast that we believe having experienced but technically remarkable team members leading small sub-companies within Fyxer's product is the way to build a magical human-like assistant. 

(And of course, each team is powered to the eye balls with hundreds of Opus agents and the like.) 

While separate teams’ magic with AI, the context gathered between each product/sub-company compounds the overall product experience for the user.

We've got some seriously incredible speakers in the Fyxer office at our weekly all-hands over the next few months. 

[Harry at Fyxer last week]

This is a sad headline, but if we put our heads down, in 10 years we could completely change the narrative around our country’s prospects. 

All my time in the US has shown me that we have the best talent in the world. 

It is belief that holds us back.

It starts with celebrating ambition.

Today in the UK, most people lead with skepticism. 

The best thing we can do for entrepreneurs is give them confidence that they can figure it out, and encourage them to think bigger.

What if it works?

Work is changing so fast.  

It's already happened in engineering. 

Why would you ever write code again? 

The same is coming for the rest of knowledge work.  

We're planning a fireside chat at Fyxer on how work is changing in 2026. 

This will focus on how knowledge workers actually want to spend their time versus how much time and money AI tools are really saving for teams. 

Who do you most want to hear from on this topic? 

If you'd be a great fit yourself, comment or DM me.

[Harry Stebbings is in this week for our last fireside]

No alternative text description for this image

In every category; voice, video, legal, email, European startups are crushing their US counterparts. 

We can build mega companies in Europe. 

20VC invested in Fyxer when it was 4 people and $1M in ARR. 

12 months later, it is $35M ARR and 63 people. 

This is our time. LFG. 🚀 

#Founder #funding #business #investing #vc #venturecapital #entrepreneur #startup

In the UK, we're taught that if something good happens, it's just a matter of time before something bad follows.

We don't have a finite number of wins.

As a culture, we put artificial limits on our ambition. 

Tell the average person your new idea and they meet it with nervousness and skepticism.

You spend your life fighting this narrative.

I want to change this. 

I want us all to fill each other with confidence.

What if it works?

Proud to be representing the incredible Fyxer team at Number 10 yesterday.

We are building the worlds best AI for relationship builders.

We've seen the explosion happen for systems builders who no longer write any code. 

This moment is coming for relationship builders.

Profitable growth in a post-identifier world | UGC & video ad creative production to drive ROAS & scale

I help mobile app companies–including gaming companies like Applovin, East Side Games, Jam City and Scopely, and apps like Bumble, Lingokids, Mojo and FitMind(among others) –acquire new users and achieve profitable, scalable growth in a Generative-AI world.

Offering both managed user acquisition services and done-for-you creative production for teams running their own ads, our clients have continued to not only grow after Apple’s privacy changes, and many have seen their ROAS improve by an avg. of 20-40%.

The secret to our clients’ success is the combination of our deep expertise from having produced 10k+ ads and managed over $100M in ad spend, and our ability to rapidly produce UGC and video ads at volume that can be used to dial in exactly what works, and discard what does not.

📹 HOW WE DO IT📹 

Leveraging a combination of both human and AI creators, we’re able to rapidly produce UGC & video ads that are based on rock solid strategy and research. By testing 50+ (or more) variants a month even at small budgets with our signature rapid testing and iteration process, we’re able to rapidly isolate key elements of winning UGC. This is essential, with our data showing that the tiniest changes can lead to outsized performance.

💻AI VIDEO YOU ASK?💻

Yes, AI video. Now not just AI video, we use both human and AI creators. With the advent of AI we’ve been able to rapidly scale and test in a way we could not before. While many in the industry worry that AI 
Creative appears robotic and doesn't invoke emotion, we’ve found it depends on the inputs you give and the strategy behind your ads. 

📈 RESULTS CREATED📈 

-70% ROAS uplift and 69% CPI improvement for Adventure Communist, an idle-clicking game.

-Tested 100+ creatives for Games2Win, drove 350K+ installs, and 100% increase in ROAS.

-Unlocked month on month growth for hampr, a hyperlocal laundry app while improving cost per purchaser by 55%.

🏎GET STARTED🏎

Reach out here on LinkedIn to get a free consultation on leverage AI-fueled creative production to supercharge your growth.

Email us at: hi@rocketshiphq.com

Shamanth M. Rao

How I'd 2x Headway’s creative velocity.

(after roasting 730 of their ads).

First, credit where it's due.
They are testing a TON. I identified 45 distinct angles across 730 ads.

That's what elite testing looks like.

But here's the problem. Almost every ad is built the same way:
↳ Stock image
↳ Bullet points/listicles
↳ Self-improvement command

It's leaving entire pockets of attention untouched.

𝗧𝗵𝗿𝗲𝗲 𝘁𝗵𝗶𝗻𝗴𝘀 𝗜'𝗱 𝗳𝗶𝘅 𝗶𝗺𝗺𝗲𝗱𝗶𝗮𝘁𝗲𝗹𝘆 to get them on a path to 2x-ing their creative velocity - while unlocking new audiences:

𝟭. 𝗡𝗲𝘄 𝗳𝗼𝗿𝗺𝗮𝘁𝘀
Wall-of-text. Meme-style. iMessage conversations.
Same messaging. Completely different structure.
That alone unlocks new audiences without building new angles.

𝟮. 𝗛𝘂𝗺𝗮𝗻𝗶𝘇𝗲 𝘁𝗵𝗲 𝗰𝗼𝗽𝘆
Some of the headlines read like a bad life coach shouting from a stage.
- "Don't be a weak man."
- "Stop rambling, lazy CEO."
Ask yourself: would anyone screenshot that and send it to a friend?
Headway's audience reads books.
Cringe has a ceiling.

Try something more nuanced, more human: "yesterday in a meeting, I connected the dots."

𝟯. 𝗦𝗵𝗼𝘄 𝘁𝗵𝗲 𝗮𝗰𝘁𝘂𝗮𝗹 𝗰𝗼𝗻𝘁𝗲𝗻𝘁
Headway has thousands of summarized books.
Each packed with real insights.
Yet most ads don't show any of it.

You get a bullet list that looks like a restaurant menu.

Instead, pull a contrarian idea from a real book.
That's THOUSANDS of new creatives sitting unused.

"3 ideas I learned from atomic habits"

---
The machine is humming now.
BUT it’s got a ton of other gears they arent touching.

45 angles got them here.

The next level comes from making the ads feel less like templates and more like something a real human would share.

Drop "ROAST" in the comments and I'll send you the full deck with the roast and recommendations.

🚀 20-40% ROAS increase for mobile apps in 60 days | AI-fueled UGC & video ad creative production 📹 | 3x Exits | $100m+ ad spend | Meta, Google, TikTok partner

At Zynga (acquired for $12.7B), we tested everything in India and Philippines.

If you're a subscription app founder spending $50k+/month: do the exact opposite.

I used to swear by Tier 3 testing(minimize your costs!) until I realized:
Cheap data isn't the same as useful data.

𝗪𝗵𝘆 𝗧𝗶𝗲𝗿 𝟯 𝘁𝗲𝘀𝘁𝗶𝗻𝗴 𝗯𝗿𝗲𝗮𝗸𝘀 𝗱𝗼𝘄𝗻 𝗮𝘁 𝘀𝗰𝗮𝗹𝗲:
𝗬𝗼𝘂𝗿 𝗮𝗹𝗴𝗼𝗿𝗶𝘁𝗵𝗺 𝗹𝗲𝗮𝗿𝗻𝘀 𝘁𝗵𝗲 𝘄𝗿𝗼𝗻𝗴 𝗹𝗲𝘀𝘀𝗼𝗻𝘀
You test in India.
Meta learns to find cheap converters in India.
That learning doesn't transfer to US campaigns.
Different country.
Different CPMs.
Different auction dynamics.
Different winning formats.
You're training the machine on data it can't use.

𝗕𝗮𝗱 𝗱𝗮𝘁𝗮 𝗰𝗼𝗺𝗽𝗼𝘂𝗻𝗱𝘀
Week 1's "winner" is a Tier 3 false positive.
Week 2 you build variations on that false positive.
Week 3 those variations compound the error.
The iteration cycle that should be your advantage?
Now it's working against you.
And "this crushed in the Philippines" builds false confidence that biases your entire US launch.

𝗬𝗼𝘂 𝗰𝗮𝗻'𝘁 𝘀𝗲𝗲 𝘄𝗵𝗮𝘁 𝗮𝗰𝘁𝘂𝗮𝗹𝗹𝘆 𝗺𝗮𝘁𝘁𝗲𝗿𝘀
Tier 3 can tell you what gets clicks.
It can't tell you what gets retention.
It can't tell you what gets paid conversions.
And at $50k+/month, the real question isn't "which creative gets installs."
It's which creative attracts users that pay back in 30 days vs 90 days.
That's the difference between cash flow positive and burning runway.
Tier 3 is completely blind to it.

𝗪𝗵𝗮𝘁 𝘁𝗼 𝗱𝗼 𝗶𝗻𝘀𝘁𝗲𝗮𝗱:
→ Test directly in the US: allocate 10-15% of monthly budget to creative testing
→ Run 3-5 day test cycles: kill the bottom 70% by day 3
→ Ship variations of winners the following week

𝗧𝗵𝗲 𝗼𝗻𝗲 𝗲𝘅𝗰𝗲𝗽𝘁𝗶𝗼𝗻:
Bootstrapped and genuinely can't afford US testing?
Tier 3 gives you directional signal. Better than nothing.
But past $50k/month, you're clinging to a habit that's actively slowing you down.
You're paying less per data point.
But the data points are lying to you.
Test where you sell.

How I'd 2x Ladder's creative velocity.

(after going through 69 of their ads and 500 of their reviews).

They are sitting on a TON of opportunity that can easily 2x their creative velocity - and add at LEAST 30-50% to their revenues:

They’ve 93% five-star reviews.
Nearly half their reviewers name a specific coach.
Not "great coaches." A specific person, by name.

That tells you something. The product works. People love it.

But here's what caught me off guard.

The reviews sound like a community. The ads dont really reflect this.

Customers keep talking about their team. Their streak. The feeling of belonging to something.
The ads? Coach in a gym. Over and over.

The most powerful thing about Ladder isn't visible in a single ad.

Three things I'd change to unlock their creative velocity - AND revenue:

1. Let your customers write the ads
- The reviews are full of lines that could be headlines
- "I finally feel like I belong to a fitness community" is a creative brief, not just a testimonial
- Show the team name. Show the check-in. Make the social layer real

2. Get into display
- Every ad in the library is video. Not a single static
- Their best hooks already work as text: calendar urgency, time promises, transformation arcs
- Repurpose what's proven. Different format, different audience, more scale

3. Give the algorithm something new to work with
- When most of your ads share the same visual feel, Meta treats them as one creative
- App screenshots. Review cards. Quote overlays. Text-only
- Each one breaks the pattern and reaches people the current ads can't

Here's what ties all three together.

Ladder has a strong, polished brand. That's intentional and it shows.
But that consistency is coming at the cost of experimentation.
One format. One visual style. One channel.

A strong brand is great until it starts capping your growth.

The product has real love behind it. The creative system just isn't keeping up. And until it does, Ladder is leaving scale on the table.

Drop "ROAST" in the comments and I'll send you the full deck.

taplio

The Battle

Archie Hollingsworth

Shamanth M. Rao

Followers

Posts

Engagements

Likes

Performance ScoreArchie Hollingsworth's LinkedIn Performance

Performance ScoreShamanth M. Rao's LinkedIn Performance

Ready to grow your LinkedIn presence?

taplio

The Battle

Archie Hollingsworth

Shamanth M. Rao

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);}Archie Hollingsworth's LinkedIn Performance

Performance ScoreShamanth M. Rao's LinkedIn Performance

Ready to grow your LinkedIn presence?

Performance ScoreArchie Hollingsworth's LinkedIn Performance