Hey everyone, Jason here, and I'm excited to present my latest Make.com scenario designed to streamline meal planning for clients, an essential tool for nutritionists and fitness coaches alike. In this workflow, I've integrated Airtable and OpenAI GPT to create a personalized meal plan that not only meets dietary preferences and goals, but also generates a corresponding shopping list for easy grocery shopping. This scenario automates the entire process Allowing you to focus more on your clients while ensuring they received a well-structured, nutritious meal plan.
Let's dive in and see how it all works. This is my submission for week 21, the meal plan. We'll start off with air table. I've got three tables, client, meal plan, meals.
The client has client ID, which is a formula of the concatenate function. Using left first name, the first character, last name and ID. which gives us Elon Musk one for Elon Musk. I've got a status field here, active, inactive. You don't want to be sending meal plans to people who are not active or not paying. Capture an email, mobile number, if later you want to send anything to their mobile number.
Got the diet preference. We've got just a whole heap here of different diet preferences, vegan, low-carb, vegetarian, dairy-free, Mediterranean. The goals.
I'm not being a personal trader, so I've just gone with calories, 1200 to 1500, protein, carbs, fats, and then the Mark Cross ratios. I don't know what that is, but ChatGPT gave it to me. Then also capturing the type, whether it's seven days, five days, and just breakfast, lunch, dinner, and two snacks. Also capturing their allergies and any notes which may be required for the meal plan.
and then just the ID which is the auto number, then record ID, then we have a link to the meal plan table. In the meal plan we've got the again the meal plan ID which is a concatenate of the word meal plan and then the ID field here. The week is the Monday starting date and then I'm also providing a shopping list so they easily know what they need to buy and then a just an inspirational type quote. that go in the top of the email this is then linked to meals we grab in the day so whether it's you know monday tuesday wednesday the meal and whether breakfast lunch dinner or snack and then description portions ingredients instructions and then plans which are all in this meals table move on to the scenario so i've got um an air table search records um And then it's got a form of status equal to active, picking up the active people.
Forgot to say that this is running every Saturday at 8 o'clock in the morning. This is to allow, with a shopping list, people to buy their groceries and have everything ready to go for Monday morning. So we're passing through to the chat GPT now. I've just said you're a nutrition expert. And then given a...
So I'm providing the diet preference, goals, numb days, the allergies and notes. And then it's just going through some instructions, identifying any allergies that must be avoided, creating a meal plan, that sort of stuff. And then compiling a shopping list.
And then a quote, an uplifting one I've said. And we've given it a JSON output. And then yeah, and then down here we've just got the JSON object as well. So the output is as I said a JSON object We're then creating a record in the meal plan.
We're adding because we're sending this on the on a Saturday We're adding two days for the get to the Monday and we're formatting it and then we're just Providing the shopping list as you can see the output is a meal plan but a shopping list and then an uplifting quote The shopping list is broken up in dairy, other fruits, grains, proteins, vegetables. Once that's done, we move on to the iterator. We're iterating the meal plan first.
And then we're also then iterating the meals, which comes out of the meal plan. We're then creating another record in the meals. Or there are going to be many records in the meals, as you'll see shortly. We're just converting the day from a number to a Monday. or for instance for one, Tuesday, Tuesday and so on.
Caption the meal name. Here is whether it's a breakfast, lunch, dinner, snack. Give a short description, the portions, ingredients, and then instructions.
And then we're linking the meals to the meal plan. We're then doing a text aggregator. This is to grab all the details of the meals. And we're formatting into a HTML format. And then finally, we're sending an email.
The subjects, the meal plan for week starting. And then we're just formatting the date. And then we've got the HTML content and some...
CSS code and then just providing the uplifting quote and then just going through each Each day and then each meal for that day then right at the end We have the shopping list and we're doing some check boxes as you'll see and then just some please call Jason You know any issues and then just a sign up. Okay, so let's run it. Okay, so let's grab two records So the two active people We'll see here where Elon and Steve Wozniak. It takes a while to chat GBT but once done it, it goes through fairly quickly. Pause it here while we wait.
Alrighty, so it's gone through the first one. As you can see, um, read one record there. It's seven days. There's five meals, so there's 35 records. We go into the meal plan, you'll see that there's one there.
And then there's a whole heap of meals. And then, yeah, just organ-based day, what type of the time of the meal, for breakfast, lunch, dinner. And then description, portion size, ingredients, instructions.
Lucky enough we kept adding it there and then finally I'll show you the email this is the first one so yeah D'Elon's got the quote there then the meal plan for each of the days with all what to do and the portion sizes then right at the end we've got the shopping list And I'm signed off, so there you go. Thank you for watching. Have a great day. So that's how I automated the meal planning process using Airtable and OpenAI. This setup not only helps nutritionists provide tailored meal plans, but also empowers clients with the tools they need to maintain a healthy lifestyle.
If you're interested in trying this scenario or need assistance setting up a similar workflow, I've included a link to the make.com blueprints in the description below. As always, if you found this video helpful, please give it a thumbs up and subscribe for some more automation tips and tricks. Don't hesitate to reach out if you have any questions, I'm here to help. Thanks for watching and remember to keep innovating.