What I like most about Claude is its strong AI capabilities, clear explanations, and ability to understand complex questions. The interface is clean and easy to navigate, making conversations comfortable. Its coding, writing, document analysis, and problem-solving capabilities are particularly useful. Available integrations can also improve productivity by connecting workflows. Performance is generally good, and the overall experience offers useful value for everyday tasks. Getting started is relatively straightforward, although there is room for improvement in integrations, response consistency, and the overall value of paid plans. Review collected by and hosted on G2.com.

Although Claude is powerful, there are some areas that could be improved. It can sometimes provide incorrect or outdated information, so important answers need verification. Usage limits can interrupt longer conversations, especially when working on complex coding or research tasks. Some integrations and external tool capabilities could be more extensive. Response speed can occasionally vary, and paid plans may feel expensive for users who do not use them regularly. Customer support and onboarding could also be more helpful for new users. Overall, improving accuracy, consistency, integrations, and usage limits would make Claude a better experience. Review collected by and hosted on G2.com.

Ease of use. Quick and accurate results when given complex tasks or questions. Simple, intuitive interface. Integrates easily with third party tools and applications. Great at breaking down complex tasks into easy to follow steps by step instructions. Review collected by and hosted on G2.com.

Likes to run many third party integrations via Powershell. Its results and methods are not always correct or accurate. It can hallucinate sometimes and go off track from the requested task. Review collected by and hosted on G2.com.

What stands out most about Claude is how well it handles long, complex context and code logic without losing track of instructions. When I feed it entire code files, error stack traces, or technical documentation, it gives clean, well-reasoned explanations instead of generic boilerplate. Review collected by and hosted on G2.com.

The usage limits can be frustrating during long, intensive coding sessions. When debugging complex multi-file projects, you can hit the message cap relatively fast and have to wait a couple of hours for the limit to reset. Occasionally, it can also become a bit overly apologetic or verbose when you ask for a minor fix instead of just outputting the updated lines directly. Having slightly more generous token limits on active sessions would make the workflow much smoother. Review collected by and hosted on G2.com.

What I like most about Claude is the way it works through problems: it really thinks things through, lays out a structured plan, and then delivers a solution. I also appreciate how it writes—its language feels much better than other AI. Review collected by and hosted on G2.com.

What I dislike about Claude is, actually, it is its strength only but sometimes it goes way far with that. Sometimes it just thinks a lot, so it consumes a lot of tokens and the limit gets used up. It automatically tries to use visualisations so I have to explicitly tell it to not use visualisations to save tokens. Review collected by and hosted on G2.com.

I love its ability to understand complex commands/prompts and provide an impressively detailed and structured answers. For analyzing lengthy documents, Claude is my go to tool. Also, I like how efficiently it summarizes long texts and also improves my written content and that too with contextual understanding. Review collected by and hosted on G2.com.

Like every other AI, Claude too sometimes provides incorrect information or make citation mistakes. Even after repeated prompts, sometimes it does not work perfectly. I have to cross check important details sometimes. Also, the usage limits keep interrupting longer tasks, which at times is so much frustrating. Review collected by and hosted on G2.com.

I consider it the best AI tool. I’ve been using the Pro version, and it’s the best AI experience I’ve ever had. It feels like there’s a human being on the other end, though, of course, that depends on how you customize it, but my first impression was extremely positive, and it doesn’t fill its responses with a lot of fluff. Review collected by and hosted on G2.com.

It is an excellent tool for what it does, but it could be more comprehensive for instance, by including features for image and video creation. I previously had a complaint regarding credits, but the latest update has improved that in every respect. Review collected by and hosted on G2.com.

The quality of its reasoning is higher than other LLMs, which makes it useful for strategy as well, not just execution. I find it more useful for deep work and complex projects. Tied for first place in what I like best about Claude is Claude-in-Chrome, which has saved me a massive amount of time when having to do almost anything in browser, troubleshooting, etc. Review collected by and hosted on G2.com.

Sometimes Claude gets locked into patterns from earlier in our conversation (i.e. template lock-in) and needs explicit resetting to properly address new requirements - I've had to develop specific prompts to force fresh perspective when this happens and that still only helps some of the time. The tool occasionally oversimplifies complex strategic recommendations or adds unwanted polish when I need precise, unvarnished language for client work. Compression bias is also a huge problem (though as context windows grow, this is getting a bit better). I'll often need it to review everything we've discussed in a thread because it's lost the bigger picture, and it will tell me it has reviewed it and seems to truly think it has -- but in actuality, it's only reviewed the summary it created which didn't capture the important nuance I needed it to understand, and it doesn't realize it's just reviewing the summary. I also feel like it's not particularly consistent. A prompt that works wonders one day can bring terrible results the next day even when the prompt is exactly the same. And it needs a lot of critical thinking to manage it or big misses slip in. That said, all of these same issues exist for all LLMs that I've used. They are not unique to Claude. So I also often use ChatGPT, Gemini, and other tools giving them the same prompts, and who provides the best result just depends on the day (and type of request). Review collected by and hosted on G2.com.

What I like most about Claude is how naturally it understands context and handles complex tasks. It’s especially useful for working with long documents, brainstorming ideas, and coding-related work. The responses feel clear and well-structured, and it does a good job of maintaining context throughout a conversation, which makes working on projects much easier. Review collected by and hosted on G2.com.

Sometimes Claude can be a bit slow with longer tasks, and there are times when it gives more explanation than I actually need. I also feel the usage limits can be frustrating when I'm working on a project and need to continue for a longer time. Review collected by and hosted on G2.com.

Among the most useful aspects of Claude, I find its ability to integrate with other applications and work environments highly valuable. It facilitates my workflows by leveraging the context these integrations provide, enabling the rapid, effective, and integrated development of various tasks. It performs well with moderately complex problems. The availability of multiple AI models allows for the appropriate selection of the intelligence level required for the task at hand. Finally, the UI is easy to understand, with a fluidity and versatility that make the user experience very pleasant. Review collected by and hosted on G2.com.

I find it challenging to keep track of token consumption on various occasions, depending on the tasks and the model. This translates into a real challenge when trying to fully understand the cost/benefit relationship of the product, considering its price. Likewise, it is complex to decide in the end the most convenient type of plan or membership to carry out the intended activities with the product. Review collected by and hosted on G2.com.

What I like best about Claude is how much time it saves on recruiting work. I use it to screen resumes against job descriptions, benchmark JDs against other companies, and build sourcing searches. It doesn’t just summarize. It gives clear reasoning and points to evidence from the resume, so I can trust the output and explain my decisions to hiring managers. It feels like having a sharp, well-organized teammate. Review collected by and hosted on G2.com.

My main frustration is hitting usage limits on heavy days, especially when I’m processing a large batch of resumes or long documents. It would also help if it remembered context across conversations more consistently, so I didn’t have to re-explain things. These are minor compared to how much time it saves overall. Review collected by and hosted on G2.com.