For years I preached about salary data being incorrect when it comes to salary data in Mexico and Latin America, in general. Websites like Salary.com, SalaryExplorer and Indeed typically have information based on reporting from Spanish speakers and very little data on English-speakers living in Latin America. We’ve done countless YouTube videos on this topic in hopes of helping international companies understand the full scope of salaries in Mexico and Latin America.
After years of struggling with this topic, I can finally say there could be a solution to this puzzle, which is OpenAI, better known as ChatGPT. Over the last several months, I’ve had the opportunity to test several scenarios where we had a job description and criteria for a role in a bilingual vacancy and was happy to see the output from ChatGPT. The results from ChatGPT have been pretty accurate. I have run into a few occasions where the salary ranges are a little off or I can see that ChatGPT is using the traditional sources like Salary.com and SalaryExplorer. In those cases, I would shy away from those results.
What I’ve found is it’s all about the context and prompts you provide to this AI tool. So, how do you achieve the best results? Well, let me show you. What I’d like to do now is give you 3 simple steps, outlining exactly what to type in the ChatGPT prompt. Please keep in mind, these are prompts I have used in the past and results will very based on the context and wording you’ve provided.
Let’s assume you have a job description for a Project Manager in Mexico City. Here’s what you want to do.
Step 1: Context
“I need to post a job for a Project Manager in Mexico City. This person needs to be English speaking. If I provide you with a full job description, can you give me a good salary range for this role?”
Step 2: Prompt
“Here is the job description” upload or copy JD
ChatGPT will likely respond with a few different results based on industry and experience level:
Step 3: Fine Tune results
Here, you’ll want to point out a factor that could have been missed or include an additional criteria you left out.
Examples:
“If this Project Manager needed to have 7 years of experience or more, how much would the salary change?”
“If this Project Manager is working with an international company, would the salary be higher?”
There are a few ways you could fine tune ChatGPT’s output, but you get the drift. As the person seeking this salary information, you’ll have a better knowledge of what to ask and how to fine tune the criteria. If you still find that ChatGPT is using sources like Salary.com or SalaryExplorer for its data, I’d shay away from the information if you are seeking an English-speaking professional. Those numbers will likely be lower than the market is offering. At this point, it may be a good idea to pick up the phone and contact a local authority on the subject matter.
If you found this blog helpful, let us know in the comments. Also, we’ll be adding a YouTube video on this topic in the next few weeks, so be sure to subscribe to our page for tips on hiring in Mexico
