{
  "id": 4796963,
  "title": "You might be using Claude wrong — 5 habits to stop right now",
  "url": "https://urgent.news/2026/09/01/you-might-be-using-claude-wrong-5-habits-to-stop-right-now",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-01T05:30:00.000Z",
  "source": {
    "name": "Tom's Guide",
    "slug": "tom-s-guide",
    "url": "https://www.tomsguide.com/ai/claude/you-might-be-using-claude-wrong-5-habits-to-stop-right-now"
  },
  "original_language": "en",
  "account": "1. Stop correcting Claude over and over\nWhen interacting with Claude, it's easy to fall into the habit of repeatedly correcting its responses. However, this approach can waste tokens and lead to a growing trail of incorrect answers. Instead, Hassid suggests editing the original prompt to provide clearer instructions. By doing so, Claude gets another opportunity to generate a more accurate response with improved context.\n\n2. Stop keeping every conversation alive forever\nAs you use Claude for various tasks, your chat history may accumulate with previous questions, rejected ideas, and unrelated tangents. While it may seem convenient to keep the conversation going, Hassid argues that long conversations can eventually hinder Claude's effectiveness. It's essential to start fresh more often, as context isn't always equal to useful context. When a conversation starts feeling cluttered, ask Claude to summarize the key points and decisions made, then continue the discussion in a new chat for better clarity.\n\n3. Stop obsessing over the perfect prompt\nThe internet has encouraged users to create elaborate prompts for optimal AI results. However, Hassid suggests a more relaxed approach known as \"brain dumping.\" Instead of spending a significant amount of time perfecting a prompt, simply share all your thoughts with Claude. This method is especially effective when using voice input, as natural explanations tend to provide more context. For example, rather than crafting a complex prompt like \"Create a weekly meal plan for a family of four,\" you could simply say, \"I need dinners for the next five days. I don't want anything that takes more than about 30 minutes, one person hates fish, I already have chicken and pasta in the house, and Wednesday is going to be really busy. Help me figure this out.\" In this case, the second prompt might not be perfect but contains useful information that Claude can work with.\n\n4. Stop telling Claude only what you want\nMany AI users tend to instruct Claude on what they want, such as \"Make this better\" or \"Give me something more interesting.\" While these instructions may yield results, Hassid recommends a more comprehensive approach. Instead of only specifying the desired outcome, provide Claude with all the relevant information, including what you've tried, what you dislike, and any constraints. This way, Claude can better understand your needs and generate a more accurate response. For instance, instead of saying, \"Make this easier,\" you could provide Claude with the current difficulty, reasons why it's challenging, and any suggestions you've already considered. This level of detail helps Claude generate a more effective solution.",
  "summary": "Using Claude every day doesn't mean you're getting the best results. These 5 common habits could be making the AI chatbot less useful.",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}