Crazy Egg AI writing tools newsletter

We ran an AI writing tool test. Same prompt, same brand guidelines, same audience doc. We gave it to three general purpose models (ChatGPT, Claude, Gemini) and eight popular AI writing platforms, then scored every output out of 50. The results: General purpose models scored 32 to 34 Tools that wrap other models (orchestration layers) scored 18 to 23 Tools built on custom models scored 16 to 28 So the general-purpose models won and it wasn’t close. Here’s what I’d take from it: Skip the wrapper. Give a top model a document with your style guidelines and audience research. That does most of what the wrappers charge for. Use AI to edit, not to form opinions. Every tool was good at logic and coherence. Every tool was bad at taking a real stance. They default to consensus. Your point of view has to come from you. Build a reusable context file. Put your brand voice, examples, and audience notes in one place. It works with any provider, so you can switch models without starting over. Check leaderboards, then trust your gut. Rankings shift constantly. But run your own test too. The “best” model on paper wasn’t the one the tester preferred for his own writing. If you want the full scorecard, how each tool did, and what the writing quality trend looks like over time, it’s all here. Worth 10 minutes before you renew that subscription. Hiten P.S. The Reddit thread is here. Crazy Egg, 13337 South St, #269, Cerritos, CA 90703, United States Unsubscribe Manage preferences