The most useful ChatGPT tips and tricks focus on prompt structure, not clever one-liners you copy from social media. Telling ChatGPT to "act as an expert" helps a little. But combining role context with a specific task, output format, and explicit constraints helps much more dramatically. On aidowith.me, the Practical Prompts route covers 15 tips and techniques organized as hands-on steps you complete on your own work. You'll work through role prompting, chain-of-thought instructions, output formatting with Markdown tables and bullet points, iterative refinement loops, and methods for fixing weak or generic responses. The route also covers lesser-known tricks: using examples in your prompt to show desired output style, setting word count constraints, and asking ChatGPT to evaluate its own response before you accept it as final. Each tip is applied to a real work task so you see the impact immediately. Total time: about 75 minutes.
Last updated: April 2026
The Problem and the Fix
Without a route
- You've read dozens of "ChatGPT hack" threads but none of the tricks stick because there's no system
- AI output varies wildly between sessions and you can't figure out why
- You know ChatGPT can do more than basic Q&A but haven't found the techniques that unlock it
With aidowith.me
- 15 prompting techniques applied to real tasks so each one sticks
- A system for getting consistent output instead of hoping for a good result
- Lesser-known tricks like self-evaluation prompts and example-based formatting
Who Builds This With AI
Marketers
Content, campaigns, and briefs done in hours instead of days.
Sales & BizDev
Prep calls, draft outreach, research prospects in minutes.
Managers & Leads
Reports, presentations, and team comms handled faster.
How It Works
Set up role and context prompts
Give ChatGPT a role, background information, and task context before asking for anything. This single change improves output quality on every task.
Apply formatting and constraint techniques
Control output with Markdown tables, word count limits, bullet structures, and explicit inclusion/exclusion rules. Get the format right on the first try.
Fix weak responses with refinement patterns
When output misses the mark, use iterative refinement, self-evaluation prompts, and example injection. Turn a bad first response into a strong final version.
Level up your ChatGPT results
15 steps. About 75 minutes. Techniques that make every prompt produce better output.
Start This Route →What You Walk Away With
Set up role and context prompts
Apply formatting and constraint techniques
Fix weak responses with refinement patterns
Lesser-known tricks like self-evaluation prompts and example-based formatting
"Adding role context and format constraints to my prompts was a game-changer. Same tool, completely different results."- Data analyst, retail company
Questions
Start every prompt with role context ("You are a [role] with expertise in [area]"). Add output format instructions ("respond in a markdown table with 3 columns"). Set constraints ("under 200 words, no jargon"). These three changes alone make the biggest difference. The route on aidowith.me covers 15 techniques total, but these three will improve your results immediately.
AI models have built-in randomness (called temperature). Same prompt, different results. Reduce this by being more specific: add examples of desired output, set word count constraints, and define the format precisely. The more detailed your prompt, the more consistent the output. The route shows you how to lock in quality across sessions.
All 15 techniques work with both GPT-3.5 (free) and GPT-4 (Plus). GPT-4 follows complex instructions more reliably, so you'll see slightly better results. But the techniques themselves are model-independent. Start with the free tier, apply what you build in the route, and upgrade only if you need longer or more detailed outputs.