Scale & Strategy
together with
This is Scale & Strategy, your friendly neighborhood ‘business information sponge’ (squeeze us out daily → get smarter about BizOps).
Here’s what we got for you today:
- Nvidia Isn’t Competing With AI Labs — It’s Expanding the Market
- Retention Isn’t a Feature — It’s a System
Nvidia Isn’t Competing With AI Labs — It’s Expanding the Market
If it feels weird that Nvidia is building open models that compete with OpenAI and Anthropic, that’s because most people are looking at it like a software play.
It’s not.
It’s a demand-generation strategy for hardware.
Take Nemotron 3 Super.
120B parameters, mixture-of-experts, massive context window. It’s not a toy model. It’s built to sit at the same table as frontier systems. And they’ve already said a version ~4x larger is coming.
But the real move isn’t the model itself.
It’s how open it is.
Weights, datasets, training recipes. Not just “open weights” like most so-called open-source models. Closer to full-stack transparency.
That’s rare at this level.
So why would the company selling the picks and shovels start building tools that compete with its biggest customers?
Because the model isn’t the business.
Nvidia doesn’t make money when you use AI. It makes money when you run it.
And the fastest way to increase demand is to make it easier for more people to build.
There are three angles here, and they’re all pretty calculated.
First, control the optimization loop.
If you’re building the models yourself, you’re not waiting on external labs to see what workloads look like. You can design hardware and software together, tighten the feedback loop, and squeeze more performance out of your stack.
That’s a compounding advantage.
Second, don’t let your customers own you.
If the frontier model market consolidates into a handful of players, those players gain leverage. Fewer buyers placing massive orders means pricing pressure goes one way.
By pushing open models, Nvidia increases the number of viable AI builders.
More customers. Smaller individually. Way less leverage collectively.
Third, flood the ecosystem.
Open models lower the barrier for everyone else. Startups, smaller teams, companies that can’t afford to train from scratch.
Give them a strong base model, and suddenly they can build products.
Those products need infrastructure.
That infrastructure runs on Nvidia.
That’s the whole game.
They’re not trying to win the model race. They’re making sure there are as many racers as possible.
Because every new company that gets into the game becomes another buyer.
There’s a subtle power move in there too.
If one company “wins” AI, they start setting the rules. Pricing, access, distribution.
Nvidia benefits from the opposite.
A messy, competitive ecosystem where no single player dominates and everyone needs compute.
So the strategy looks weird on the surface.
Why compete with your customers?
In reality, they’re doing the opposite.
They’re making sure they never have to depend on just a few of them.
The perfect email is useless in a spam folder. This suite guarantees maximum deliverability
You’ve built the list and crafted the copy.
You’ve designed a stunning email in your ESP.
Then you hit send, and… crickets. Opens are low, clicks are embarrassing, and the only replies you get are bounce notifications.
Sending to a bad list is more than wasted effort. It’s a direct threat to your sender reputation that risks drowning future campaigns in spam and promotion folders.
And no readers = no ROI.
Stop gambling with your deliverability.
ZeroBounce is the all-in-one platform trusted by marketers at Amazon, HubSpot and Netflix for landing their messages in primary inboxes.
They provide 99.6% accurate email validation, inbox placement testing, blacklist monitoring, and more, so you can send with absolute certainty and get a bigger ROI from your emails.
Discover why over 500,000 businesses rely on ZeroBounce to ensure their message is heard.
Retention Isn’t a Feature — It’s a System
Getting users in the door is easy. Keeping them is where most products quietly fall apart.
Elena Verna from Lovable shared a few tactics that actually move the needle. None of them are flashy. All of them work.
The first one is almost embarrassingly simple.
When someone tries to cancel, don’t force a binary choice. Full price or goodbye is lazy.
Most churn comes down to “too expensive” or “not using it enough.” So they added a $5/month Lite plan directly in the cancellation flow.
About 10% took it.
That’s revenue you were about to lose entirely. All because you gave people a middle option instead of a door.
Next one fixes something most teams ignore.
Payment failures.
Everyone relies on email retries. Which is great, assuming people actually read their email. They don’t.
Lovable added in-product alerts when payments failed. Recovery rates jumped ~50% relative. From ~20% to 30%+.
Same users. Same issue. Just surfaced where it actually gets seen.
Then there’s a subtle psychological trick.
Even paid users get a small amount of free daily credits.
On paper, it looks like generosity. In practice, it’s a loop.
Credits refill tomorrow. That gives users a reason to come back. And once they’re back, a percentage of them just pay instead of waiting.
You’re not just retaining users. You’re creating momentum.
The rollover policy is another one most companies get backwards.
Typical approach: unused credits expire.
Their approach: credits roll over as long as you stay subscribed.
It sounds like you’re giving something up. You’re not.
Users build up a balance, which creates a reason not to cancel. Nobody wants to walk away from something they’ve already paid for.
Last one is about removing friction.
Top-ups.
Let users pay incrementally when they need more, instead of forcing them into bigger plan upgrades. That alone lifted retention ~7% across a million paid users.
Not because top-ups are magic.
Because frustration kills retention, and this removes a common one.
None of this is groundbreaking on its own.
That’s kind of the point.
Retention usually isn’t about one big feature. It’s about fixing the small, obvious leaks most teams ignore.
And once you fix enough of them, the system starts working.
Was this email forwarded to you?
That’s it for today and as always It would mean the world to us if you help us grow and share this newsletter with other operators.
Our mission is to help as many business operators as possible, and we would love for you to help us with that mission!
