Like a lot of people, I have tried plenty of different ways to track my fitness and wellness goals over the years. I have used apps, spreadsheets, notes on my phone, and even just tried to keep track of everything in my head. The problem was never that those systems were bad. The problem was that they did not really fit how I think or how I actually move through a day. I do not want to spend more time managing a tracking system than I do focusing on the habits themselves.

That is what led me to create what I call my Performance Dashboard project. Instead of using AI as a one-time tool, I built a project where it acts more like an ongoing thought partner. Throughout the day, I simply tell it what I have eaten, what workouts I have done, my weight, whether I lifted, walked, played golf, or even hit balls in the simulator. Some days the information is very detailed. Other days it is as simple as, “I had a protein shake, a burger, and we lifted upper body today.” The point is not to create perfect entries. The point is to create an easy system that I will actually use consistently.

What makes the dashboard work is that I gave AI very specific instructions from the beginning. I started by setting clear goals for the project. For me, that meant focusing on building strength and muscle while also losing weight. I gave it a calorie target range, a protein goal, and context about my routine. Since I lift regularly, play a lot of golf, and often have days that include both workouts and more indulgent meals, I wanted the dashboard to understand that the goal is not perfection. The goal is long-term consistency.

Inside the project, I instructed AI to always estimate calories and protein when I do not provide exact numbers, to track trends across multiple days rather than overreacting to one meal, and to be honest about where I am doing well and where I am drifting off track. I also told it to keep the feedback practical. I do not need a long report or generic encouragement. I want a quick, useful summary that tells me where I stand.

At the end of each day, I can ask the dashboard to pull everything together. It gives me an estimate of my total calories and protein, compares those numbers to my goals, factors in my activity, and gives me a quick evaluation of the day. Sometimes it points out that I stayed right on target. Other times it reminds me that the late-night snacks, extra drinks, or “cheat meal” added up more than I realized. The value is that it shows me what actually happened instead of what I think happened.

Over time, that has become the most powerful part of the project. The dashboard is not just tracking food. It is helping me notice patterns. I can see which days I tend to overeat, how my lifting days impact my nutrition, and how weekends are often very different than weekdays. It has also helped me recognize that I am often more consistent than I give myself credit for, but that there are a few recurring habits that can throw me off if I am not paying attention.

One of the things I find most interesting is that my wife, Rose, has started creating her own version of this project as well. Her dashboard looks different because her goals, routines, and priorities are different than mine. But the overall structure is the same. She has set goals, given AI instructions, and created a system that fits how she wants to track her progress. That is really the key. The value is not in copying somebody else’s perfect dashboard. The value is in building one that fits your own life.

For me, that is why this approach has worked better than anything else I have tried. I am not forcing myself into someone else’s app or system. I built something that adapts to me. There is very little friction, I can update it naturally throughout the day, and I get useful feedback without spending extra time. In a lot of ways, that is exactly what I think AI does best. It does not replace the work. It helps make sense of it.

How to Build Your Own Performance Dashboard

If you want to create a system like this for yourself, the easiest way is to build it as either a Project in ChatGPT or as a Gem in Gemini. Both can work well. The biggest differences are cost and organization.

A Project in ChatGPT is available with a paid ChatGPT subscription and has the advantage of keeping everything together in one place. Your instructions, your ongoing conversations, and any files or notes you want to upload all live inside that same project. Over time, that makes it feel more like an evolving dashboard instead of a collection of random chats.

A Gem in Gemini can accomplish much of the same thing and is included with some Google Workspace and Gemini plans. If you are already working heavily in Google tools, it may fit naturally into your workflow. The tradeoff is that it can feel a little more scattered because your instructions live in the Gem, but your day-to-day conversations may not be organized quite as cleanly over time.

Building It as a Project in ChatGPT

In ChatGPT, create a new Project and give it a name like “Performance Dashboard” or “Wellness Tracker.” Once the project is created, add instructions that explain your goals, your routine, and how you want ChatGPT to respond.

Your instructions might look something like this:

• My goal is to build muscle and strength while losing weight.
• My daily calorie target is 2,400-2,700 calories.
• My protein goal is at least 220 grams per day.
• I lift 4-5 days per week and often play golf or walk.
• Estimate calories and protein if I do not provide exact numbers.
• Track patterns across days and weeks.
• At the end of the day, provide:
  • Total calories
  • Total protein
  • Activity summary
  • Whether I met my goals
  • Two or three observations or suggestions

You can also add context about yourself, like your height, weight, favorite foods, common routines, or even the fact that weekends tend to look different than weekdays. The more context you provide up front, the more useful the responses become.

Then, throughout the day, simply add updates inside that same project:

• Breakfast: protein shake with milk, breakfast bowl, banana
• Lunch: chicken wrap, chips, cottage cheese
• Lifted upper body and walked 1 mile
• Had pizza and a few beers during the game

At the end of the day, ask something like:

“What does my dashboard look like today?”

or

“Give me my daily summary and tell me where I am trending for the week.”

Because it is all happening inside the same project, ChatGPT can use the earlier information from that project to recognize trends and keep building on previous days.

Building It as a Gem in Gemini

In Gemini, create a new Gem and give it a name like “Performance Dashboard Coach.” In the instructions section, add the same type of information: your goals, your calorie and protein targets, your workout routine, and how you want Gemini to respond.

A sample prompt for the Gem might be:

“Act as my performance dashboard. My goal is to build muscle and lose weight. Track my calories, protein, workouts, and weight over time. Estimate values when I do not provide exact information. At the end of each day, summarize my total calories, protein, activity, and whether I met my goals. Also point out any patterns or habits I should notice.”

Once the Gem is built, you use it much the same way. Throughout the day, open the Gem and type in what you ate or what activity you completed. Then ask for a summary at the end of the day or week.

The main thing to remember is that Gems work best when you regularly remind them of your recent history. Since Gemini does not organize everything into a dedicated project space in quite the same way, you may need to occasionally paste in recent summaries or ask it to reference what you have tracked over the last few days.

Both approaches can work really well. If you want the cleanest experience with everything stored in one place, the ChatGPT Project is probably the better option. If you are already heavily invested in Google tools and want something tied into that ecosystem, a Gemini Gem can absolutely do the job too.