TopazLabs Moves to Subscription-Only - Your Thoughts as Creators?
Hey folks π, recently, TopazLabs (well-known in the AI video/image enhancement space) announced they're shifting from lifetime licenses to subscriptions only. That's sparked a lot of chatter - some say subscriptions feel cheaper upfront, but many users feel it becomes more expensive long term, especially since canceling can be tricky and refunds are limited.
We're curious to hear what the Product Hunt community thinks about this shift:
As a user, do you prefer subscriptions or lifetime licenses? Do you see a difference between desktop apps vs. online/cloud apps? Why?
As an AI developer or maker, which pricing model makes you feel more confident investing in for the long term?
For context:
We build a competing product in this space and offer both 1-year subscriptions and lifetime licenses. Our vision is pretty simple - to help everyday users solve real video and image problems without locking them into a pricing model that doesn't fit. Seeing Topaz's move made us want to open up the conversation here.
Would love to hear your takes! πβ€οΈ



Replies
Cal ID
As I'm a creator myself, I'd say flexibility is key.
Subscriptions help companies stay sustainable and ship frequent updates, which is great if the product evolves fast. But lifetime licenses build trust and let users βownβ their tools.
In my opinion, the sweet spot is to offer both and let people pick what fits their workflow. Forcing everyone onto subscription-only often turns loyal users off
Aiarty Video Enhancer
@sanskarixΒ Really appreciate your comment, glad you shared this! π
Promomix
with subscriptions the company can generate more predictable recurring income, which can justify investing more in R&D, building new AI models, pushing updates frequently, and supporting infrastructure (especially cloud).
Aiarty Video Enhancer
@margret_rhymeΒ That's a great point - predictable recurring income is definitely a big advantage for companies. :)
Raycast
I mean, this feels like touching a third rail, especially if you read the comments on similar pricing changes on Reddit. π
Ultimately the strategy should depend on the software you're building and how much you intend to innovate and continue developing it.
For desktop apps, I think Sketch landed on a reasonable approach β where you pay for a year of updates. If you stop paying, then you can keep using the version you have, but you won't get the updates. It was controversial when they adopted this approach, but they were tboughtful and communicative about it.
They've continued to roll out regular updates and as a result, you really get a sense for the value of what you're paying for.
Some apps go the subscription route, but then rarely or infrequently make updates or improvements, which works against them.
So if you're able to commit to a regular schedule of updates and improvements and can communicate about them, then a monthly subscription can be aligned, especially if you make it easy to cancel. This approach is also suitable for web apps where the expectation is that constant small updates are being made frequently β if not daily.
Aiarty Video Enhancer
@chrismessinaΒ Thanks for the thoughtful take π
The Sketch model you mentioned is a great example because it strikes a middle ground - users still feel like they "own" something, but there's a clear path to ongoing updates if they want them. That balance of control + continued value is exactly what a lot of people seem to want.
I also think the key isn't just the pricing model itself, but what comes with it - updates, service, and overall fit. Like you said, if a company ships steady improvements and communicates them well, subscriptions can feel fair. But when prices quietly creep up or cancel/refund becomes painful, that's when frustration really sets in.
Another angle: a balanced pricing model can also give developers the confidence to keep investing in the product long term. The tricky part is finding that balance. And sometimes, companies don't just change their model for revenue - it's because their target users have shifted. What fits pros might not fit hobbyists, and vice versa.
Really appreciate your perspective!