Best practices and sample prompts for AI-enhanced search - LinkedIn Recruiter Tutorial

- [Harley] Sometimes facing a blank search bar can be just a little intimidating. Let's look at some best practices when you're writing prompts so you can uncover hard to find talent and skills and maximize the relevance of your results. First of all, be as specific as possible. Let me show you what I mean. I'm going to look for a program manager. I'm going to put, "Find a senior program manager," and I have a lot of results here. Oh, oh dear. Okay, you know what? Check it out, I've got 240,000 people so let's hone in on what I really want. I'm going to go back to my search bar and type in something much more specific. I'd like to find a senior program manager who uses Asana, has managed people for five or more years and is based in France. Now, take a look. Recruiter has a much better idea of what type of talent I'm trying to find, and my results are much more specific and much more relevant. It's going to be a lot easier to refine these results now that there aren't hundreds or even thousands of them. Another best practice is we're going to go back out to the homepage. This ties into what we were just talking about. If there is a chance for misinterpretation, you must be clear. For example, I'd like to find an accountant based in London. Seems pretty straightforward. We get my results, and you know what I forgot to say? It is somebody in London, Ontario, Canada, not London, England. Whoops. Well, okay, what I can do is use this search bar right here to let Recruiter know that's what I want. By the way, I am telling it not London, England specifically because if I leave just London, Ontario, Canada, I will get results for both London, Canada and London, England. Wonderful, now my results are all in London, Ontario, Canada, and I can continue to filter them. All right, so from this page I want to show you one more cool thing. That is that you are going to be getting suggestions, and I really encourage you to take advantage of them to either broaden or narrow your results. Use them to start thinking creatively. For example, we could take a look at show me suggested titles and now I've got the top titles for this talent and recommended titles based on search activity. I'm so grateful to have these because now I understand more what the titles are and let's say I'm learning about this talent pool, I could even ask Recruiter to tell me what types of skills these people have. Great, now I can look at the top skills for this talent pool, add them to my search, and also learn about these people. This helps me, I'm grateful to have these suggestions because now I understand more about this role. Another best practice is this, and it totally relates to this example of our accountant in London, Canada. Don't give up on a search and start over if you don't need to. There may be times when you want to restart your search because you feel you weren't specific enough, but it's often better to refine your search rather than starting over from scratch. If you get it wrong, you can type, "Give me additional search suggestions." I now have suggestions which can help me improve my search. And remember, AI is learning from you and giving you better results the more you use it. It's looking at who you are qualifying for your projects and it's also looking at the people you are choosing to not qualify. So the more you refine your results, the better your recommended matches will be. All right, we're going to go back out to my homepage for one final tip. If you know somebody that is amazing, consider using an ideal candidate to start your search. For example, if you know someone who has the qualifications you need or even someone who was in the role before, they may be a perfect example to start with. You can have AI look them up to find other great candidates with similar experience. Let me show you. I'm going to ask Recruiter to find me a candidate similar to this amazing person I know named Bethania. Notice I'm putting an @ symbol in here, this is an @ mention, and it's telling Recruiter, "Hey, I'm looking for an actual person. As I type, you'll see that there are suggestions and the closer I get to her name, there it is, the first one. I have Bethania Barboza, and now I click back in, choose AI-assisted Search and Recruiter is working to find me candidates. And now I have my results, 619 people that Recruiter found who were chosen based on Bethania's profile using information such as job title, years of experience, industry, and so on. I know these are going to be great candidates. And as I scroll through, I notice I also have a suggestion here. I can add some filters to increase the percentage of male candidates in this case. Whenever you see these recommendations, feel free to use them. Again, I really encourage you to play with these features to continue to refine your results. My final request as you're using these tools is to please give us in-product feedback by clicking the thumbs up or the thumbs down icon to rate the results. This way we learn from you and the results are going to continue to improve your overall experience.