.png){kind=spacer}
A step-by-step guide for how to use ChatGPT for recruitment from creating job descriptions, reaching out to candidates, interviewing, and winning new clients.
Recruiting top talent can be a time-consuming and challenging process, but with the help of ChatGPT, you can speed it up by hours and even days.
As previously explained, the faster you blend your recruitment efforts with AI, the better you will be as a recruiter. Embrace it.
In this guide, we’ll show you how to use ChatGPT for recruitment from start to finish, automating almost every step of recruitment.
Some choose to use AI recruitment agents like Uwi instead of prompting ChatGPT, but if you're curous to how you can use ChatGPT as a recruitment tool this blog is for you.
You’ve just onboarded a new ‘client’ in Amsterdam who is in need of a Frontend Developer to help build out their new browser extension. It's time to start the search, but before that, you need to make sure you have the essentials to create a job description.
Before you begin crafting your job description prompt, you’ll first need to identify the 3 main requirements for the position (or any position):
If you, like most recruiters, don’t have a background working in your specific recruitment niche, you can use ChatGPT to help you learn and break down technical jargon by prompting something along the lines of: Describe the difference between a software developer and a software engineer in a way that’s easy to understand for someone with no technical knowledge.
ChatGPT can help you create detailed job descriptions and generate key responsibilities, required skills and qualifications, and other necessary information. You just need to focus on your prompts. The more you put in, the better the result.
For example: Instead of prompting a generic “Please write a job description for a Frontend Developer” and then doing a bunch of edits, we took the time to write as much information as possible that we needed from it. See below
Using this formula, our result was informative, formatted perfectly, and captivating to candidates.
So when prompting for job descriptions, keep these key things in mind:
Be specific: Specify the information that you want to receive, not just the information that you put into your prompt.
Give it Feedback: Don’t be afraid to tell the AI exactly what you were expecting and how it did not meet expectations. It will learn your style and what your goals are in its process
Break it down into steps: Often, the AI can be overwhelmed with all the information you put into it and like a child, learns better when you break it down into small steps. If you’re prompting a large paragraph of say, 600 words, break it down and prompt it in 4-6 steps.
Prompt Example: I will tell you to create a job description in steps. Step 1: Generate the job title, Step 2: Write about the job responsibilities, Step 3: Highlight the compensation.
The next step in the recruitment process is to start screening candidates for the role.
Although filtering candidates is the bread and butter of being a recruiter, you could still automate a large chunk of the responsibility using advanced algorithms.
You’ll need to narrow down your candidate shortlist and when ready, parse the candidates’ CVs into ChatGPT and prompt the following into the thread:
Here are 3 CVs of candidates that I shortlisted for the job. Using the job description we created, please ascertain which candidate matches the job best and provide a list of why they are a good or not-so-good fit.
Note: It's important to do this 2-3 CVs at a time, otherwise the system can be overwhelmed
The algorithms will not only scour the keywords of the job description and CVs but also read and understand key information.
Before you start interviewing candidates, you need to create a list of screening criteria–like a tick-box. This list should include the skills, qualifications, and experience required for the job. You can ask ChatGPT to create a list of pros and cons to consider when screening candidates. Ask it to provide it to you in tabular form so you can paste it into Microsoft Excel or the equivalent.
Prompt: Provide a list of how to screen candidates for the role. Please provide this in tabular form.
The result:
Once you have a list of screening criteria, it’s time to filter the candidates that you shortlisted.
To ensure that ChatGPT is able to screen candidates as accurately as possible, you need to train it on data. This means providing ChatGPT with examples of qualified and unqualified candidates.
To do this, create a desktop folder of great candidates and candidates that were not a good match for roles you’ve recruited for in the past. For example, copy the job description and the CV of a candidate that you successfully recruited.
Prompt: Here is a job description. Here is the CV of the Frontend Developer candidate that I recruited for the job. He had great skills and knew Typescript very well. I'm looking for the same type of candidate for the role in Amsterdam but someone who has better experience with Redux.
Training ChatGPT on data is an iterative process. You’ll need to provide ChatGPT with feedback on its performance and continue to refine the dataset. Over time, ChatGPT will remember your inputs and become more accurate at screening candidates.
Once you’ve identified potential candidates for the job, it's time to engage with them and start building a relationship. ChatGPT can help you generate personalized outreach messages that are somewhat tailored to each candidate's profile.
When crafting your initial outreach message, keep in mind that candidates receive tons of similar messages from recruiters every day. Therefore, you need to make sure your message is as personal as possible to stand out and grab their attention. Use the following as a base but be sure to add as much personalization as possible.
Upload their resume into ChatGPT and Prompt: Craft a personalized recruitment outreach message to a frontend developer based in Amsterdam who previously worked at Mozilla Firefox. The message should be tailored to our job opening at XYZ, a company that also operates as a web browser. Please highlight specific aspects of the job and company that align with the developer's expertise and experience while expressing genuine interest in their potential contribution to the team.
Result:
Once you have sent your initial outreach message, it's important to follow up with candidates and hit all those 8 touch points. ChatGPT can help you create a basic follow-up template that you can copy and edit, but better, it can also create a list of FAQs that candidates will likely have about the job or company.
During your outreach, copy as much information as possible from the company’s LinkedIn bio, job description, and website and prompt:
Here are 3 pieces of information about the company I’m recruiting for. The job description of the role in question, their LinkedIn bio, and information from the company’s website. Please answer further questions I have.
Paste the information all together in the same prompt but separate it. The AI will scour the information and you can ask it any question for a quick answer about the company. This will come in handy.
One of the best use cases of ChatGPT is that it has great ability to create excellent sales and recruitment templates for you to copy and paste into an outreach email with minimal edits.
You can create multiple templates for business development emails and segment them by:
And any other relevant point to segment companies by.
Note: It’s important to segment as much as possible to hit peak personalization at scale.
Prompt: Create four distinct variations of a business development email intended to attract new clients to utilize our tech recruitment services. Each variation should be tailored to different client types or industries, highlighting the specific benefits and value propositions of our services for each target audience. The emails should be engaging, persuasive, and personalized to resonate with potential clients.
Result:
Once you have your variations, copy each one into a Google sheet document.
After winning new clients, you can be nearly fully autonomous by copying and pasting the contents of an email into ChatGPT and asking for a response.
Let the AI know how you want your response to be. For example:
Sure. Here are a few steps for screening applicants using AI:
ChatGPT does in fact store data. So you should probably redact names, full addresses and any other sensitive information or you could be breaching GDPR rules.
Also, the AI will sometimes make unconscious bias errors but you can minimize its biases and prompt: “Revise this job description to ensure that it is free from unconscious bias. Please consider the language used to describe the responsibilities, qualifications, and characteristics required for the role. Ensure that the revised job description promotes diversity and inclusivity.” to make sure it's not inadvertently introducing bias into the hiring process.
You can use ChatGPT to automate almost all of your workflow but remember, it does have the following downsides:
Limited Understanding: When performing tasks like emailing, understand that ChatGPT has limited understanding of human emotions and may not always provide the best answers. It’s built on a database of information, rather than having an actual brain.
Accuracy: ChatGPT does tend to make a lot of mistakes, so be sure to iterate and double-check everything.
AI shortlist from 1 billion profiles, reach out on autopilot
