Candidate Sourcing & Scoring with Clay

Hello.

In this video I will be showing you how you can use clay to source candidates and grade them for any specific job.

you may be absolutely new to Clay even if you have never used it before.

I think this would be a good starting point.

You can do it with either a free account or if you want to sign up just go to these links given in the Playbook.

and you can sign up to Clay.

So, so what would you do here?

What would be the example?

let's pick up since I've already made this for Birmingham.

I can show you a different role in Birmingham.

But here all you'll do is you will just set up the find people criteria for the candidates and everything else will happen automatically for you.

It will enrich the candidate.

We will find their headline summary, it will find their distance from the job location which here is set to Birmingham.

It will score them against the certain JD and then it will finally tell you if it is an overall fit, out of 10 or not.

And I've also added that it writes a small icebreaker to reach out on email making your emails look more customized and helping you get more responses from candidates.

it also finds their work email as well as personal email depending on which one you want to use now.

So first off you will just go here, click this button.

Create table using template.

I have multiple workspaces.

I'm using one of my clients just to build a sample for you.

So now it is creating the playbook for us.

It has no data, it has one filter that only show me candidates higher than 7.

What we will do now is you will click on the left top here in the find people this plus icon.

So let's hit the plus icon.

And now you need to define the criteria of what type of candidate you want.

So, so let's assume this was for Ops Manager.

let's do a similar one.

Let's do actually let's just do Birmingham and job title will be.

So let's do something tech.

Okay, job titles, data engineer.

I'm just putting one job title ideally I usually put exclusions, inclusions, a lot of different stuff but just this will do the job for you.

we want them to be working at companies that are at least mid sized or bigger.

I don't want someone who's from a startup but this is just me building like coming up with a criteria here.

I don't want specific experiences.

You will put the location here very important.

And let's say we're going to do the test for Birmingham because I have a filter in there already set up for Birmingham.

But if you want, you can do any other location as well.

This is where you will pick it up.

So Birmingham, could do London, you could do anything else and I'll show you later.

If you did London, then what would change?

So you're just putting up a simple criteria of what type of candidates you want, what companies should they come from, what job titles should they have, what location should they be.

If you want specific experiences, you can put it there.

You can also put what keyword should they have in their experiences.

Let's say I want SQL.

let's say I also want, I don't know, data warehousing.

That's fine.

Let's say I also want, for example.

Okay, a couple more candidates.

Okay, now this is a bare minimum criteria.

I have a few candidates and this is not costing you anything, this is costing you zero credits just to get their LinkedIn profiles.

Now I'm going to hit continue and this will bring all these candidates into this table.

And by default I think our table should be remove this filter which is set up from the last time I made it.

So people, first name, last name, the company domain and then the missing items are being enriched now.

So like I said, they automatically got enriched.

Okay, so distance from Birmingham, where are these people based?

Let's see.

So Birmingham.

Birmingham.

Okay, yeah, exactly.

So they are based in Birmingham.

So distance is zero if I did.

And this means they're a good fit.

So this is like 5km.

it easier for you.

Kilometers.

So distance here is 0.5kilometer.5km.

So all of these are a good fit.

Let's say it's an in office job in Birmingham.

that's the only reason I put this filter.

It's sometimes I don't even put it.

If it's a remote row, then we are scoring these people based on this criteria.

So right now this is based off of a job that I had previously.

This was for Operations manager.

So all you need to do is you need to go in here and you just need to replace the JD and ask it to write a different prompt and I'll tell you how to do that.

So you need to plug your JD here in the Generate section.

So I'll show you again.

You will go to the JD Match score, and this is the main part, even if you don't do anything else and just find the people like I showed you and then come here, click Edit column, go to Generate.

Because so how this works is in Generate, you can just ask AI.

This is what I want.

Write me a prompt.

And then you click this button and it generates a prompt for you with all the values and everything.

So you don't even have to write it.

The only thing you need to change is the JD here.

So you would say, here is the jd.

Yep.

And now we will.

Okay, right now it's too long, so I'll pick a smaller snippet of the jd.

So just try to summarize the JD as much as possible, but do not skip, all the specifics, like the scales of what they should be doing.

So I'm just picking up my key responsibilities.

apologize, I'm looking away because I'm doing it from another screen.

So here you go.

Okay, I've picked it up now.

Here is the jd.

And then I've given an inspiration prompt.

Okay, so jd.

And then I've given it a prompt that worked for me before and I've said write it like this.

If anybody wants to dive deeper there, then you will simply press the generate button.

So again, it easier, paste the JD here in the JD section.

Don't remove anything else, please.

This part, all of this stays the same.

then you will say, generate.

me again.

And now it will write a grading prompt for us.

Alrighty.

probably takes half a minute sometimes.

it's writing the prompt for us.

Here is the jd.

Here are the instructions.

So it will take the fields that were enriched, like the person's LinkedIn profile, their summary, their experiences, and then use that information.

Yep.

Headline summary.

Then in the end, it would give you something like this.

So these are input columns, like the name, the location, headline and everything.

And output would be the score and the reasoning for the score.

So let's say I say save and run on 10 rows.

the same thing that you would do.

It's saying, yes, it will cost credits.

That is okay.

And now as you see, it's running again and it is scoring our candidates.

And this could be done on like, I usually do it on way bigger data sets, like 400, 500 candidates.

But the idea is that you get the final 50, 70, 100, 10, 20.

However, many candidates you want but you get the top ones, the ones with the best score.

So you as a recruiter or a recruiting business owner who has a team of sourcers, they just go after the strongest fits.

Instead of just going after everyone and talking to people that are wrong fits, you can just go after these.

So what I do the last thing that I do once they're graded now they've been graded location fit.

Why is the location let's map it again.

So what I'm doing here is if you click here on a response you can say see all the details that AI gave you.

So in the reasoning it says this person is located in Birmingham.

So matching the perfect role core skills, this, this, this over 14 years of experience.

Overall fit, strong seniority or location alignment but lacks this specific things relevance, no explicit details on recent hands on.

Okay, so it's like a so so kind of fit and you can, this criteria can be made more strict or more lenient.

I've kept slightly lenient right now just to show you guys but let me show you for example this guy's a 9.

So here it says he has SQL, he has data, warehousing, cloud, everything.

Solid industry experience, strong fit with exactly 3 plus years experience like required like the JD I gave it, I gave it like a made up JD right now.

But this is how you will get it all graded.

And if you want overall fit is already mapped here.

Let's say we want to also map the experience part.

You will click here, add as a column experience fit I'll call it and then this comes here.

I think location fit isn't mapped, that's why it's not coming.

I can do it do that too.

So let's just map this fit.

Yep and finally we are also writing this icebreaker.

So would you be it?

It would say your extensive experience as a senior data engineer support data driven growth in Bournemouth for the technical manager role.

Would you be open to brief conversation about this opportunity?

So what I'm doing here is I'm just asking you to write an icebreaker, give me a short description of a candidate specific so by two line message tailored to the candidate using the job description.

So it would say hey John, I saw you're a data engineer, based out of Birmingham or Bournemouth.

I have a job here.

Let me know if you're interested.

So your email will look more customized and you don't have to write it every time then it's trying to find their emails also and you Already have their LinkedIns scroll to the left here.

So now once you have the gridded candidates, what can you do with this data?

You will click Actions here and you can the easiest way is just to download the CSV.

But then if you guys want to do more and you want to link it with your CRM, you want to create sequences and reach out to these people or do the same thing for sourcing clients and building a pipeline, just let me know.

You can book a call, there's a link at the end of the playbook.

Or just come join our school community where I do it with hundreds of recruitment Thank you so much.

See you on the other side.