Lately I’ve been posting bug reports and feature requests here, so I thought it was time to also share some honest feedback about what it’s like to use Cursor — not just what’s missing.
I’ve been using Cursor for a few months now, and I can say without exaggeration:
it saves an enormous amount of time.
When you compare the value it delivers to its cost — especially for professional, paid development work — the time savings are massive.
1. Constant Upgrades
Cursor evolves incredibly fast.
Every time I imagine a new UI tweak or feature improvement, I usually find it already released within one or two versions. This kind of development velocity is rare.
2. Cursor vs. Alternatives
After the recent pricing changes across the industry, I tried almost every other alternative — yes, I even paid full price for many of them.
But in the end, I always had to come back to Cursor.
I’m not sure what Cursor’s “secret sauce” is — especially when competitors get a lot of positive marketing buzz — but the truth is, the same prompt, with the same model, in Cursor, actually works.
No bugs. No weird behavior. Just results.
With those same prompts, in other apps, it just… doesn’t.
I won’t name names, but I had to abandon many of the credits I purchased elsewhere because the tools didn’t deliver working results.
I do believe others will catch up someday — but for now, this is the reality.
(For the record, I don’t count tools that are just a thin terminal UI locked to a single provider — in my experience, no single model is good for everything.)
3. Customer Support
To be honest, the email support (hi@cursor) has been a really poor experience — the bot there almost never understands what I’m asking.
But when I post here in the forum, the responses are usually quick, clear, and helpful. So far, most of my issues have been resolved in a reasonable way.
4. Pricing and Auto Mode
True, we no longer pay “per request” — it’s all token-based now. But the Auto mode is still extremely useful.
For a large portion of my tasks — summarizing, navigating large codebases, understanding what other models are trying to do — Auto is often more than enough. And it feels unlimited.
I reserve the more expensive models for very specific or complex tasks.
Interestingly, I’ve also found some tricks to identify which model was actually used in Auto mode — and it’s helped me understand why Cursor doesn’t always reveal the model names. Their approach makes more sense to me now.
5. What’s Still Missing
Most frustrating: lack of credit/balance transparency.
There’s currently no reliable way to know how much I’ve spent this month, or how much I have left.
The dashboard shows all usage — including things from Auto mode that technically don’t count against your paid balance — which makes the numbers meaningless in practice.
I can only guess how much credit I have left, which is extremely frustrating. I hope this changes soon.
Second issue: editing & file cleanup.
Since billing is token-based, it would make a lot of sense to allow users to edit messages and remove file attachments before resending a prompt — to avoid unnecessary costs.
That option is currently missing and would make a big difference.
Final Thoughts
I spend over $200/month on Cursor — but I’m completing in a single hour tasks that used to take me a week.
I finish a month of work in a few days.
So yes — the benefits far outweigh the downsides. Especially when there’s no real competition right now.
That’s my honest review — maybe it helps someone else.
And thank you to the Cursor team for building such a powerful and focused tool.
