transcribe

How to stop HubSpot duplicates with Clay.com and AI fuzzy match search

Jacob Tuwiner · 5m · transcribed May 2026
More from Jacob Tuwiner Business
𝕏 Share ▶ YouTube 📥 PDF 🤖 .md

Transcript

0:00 In all these cases, you want a way to look in the CRM and figure out does this person already exist or not. This is specifically going to return the HubSpot properties that we want from the record that we find. So, the rest of the enrichment workflow is actually going to now append all the enrichments to this record. Hey, how's it going guys? Jacob Tony here from Sculpted. Today, I want to show you a really cool clay workflow that we've built for all of our clients.

0:19 It essentially does an AI fuzzy match in our client's HubSpot instance to look for an existing property using things like first name, last name, company name, job title, but not necessarily a record ID or even an email. This way, we can avoid creating duplicate records in HubSpot when we're doing a list import or importing an inbound lead, whatever it might be. even if we don't have a proper unique identifier like a LinkedIn URL or an email or a record ID from an existing record. This is really really cool and you can grab this and use it in your workflows to essentially run this on all contacts before they go through the core workflow. We have at sculpted done this a bunch of times for our clients cleaning up HubSpot and therefore we have pre-built templates that we've incorporated and we have a system that we run on all records before they go through our clients pre-determined enrichments so that we can clean up the record and avoid duplicates and a bunch of other issues, mistakes we've already made so that you don't have to. In this video, I'm going to show you exactly how it works. If you're interested in me auditing your clay table to find areas of opportunity to save money, save clay credits, and do a better job with your clay spend, comment down below, and I will personally review your table, send you a loom, showing you all the areas we can save a ton of money. And if you're okay with it, even post an anonymous video going through the clay table like this.

1:25 You can be featured on my channel, on my LinkedIn. It's totally free. I usually can save you 80 to 90% of your clay credits. It's usually a massive saving, especially if you're enriching a large data set. That could be thousands or even tens of thousands of dollars in savings for free. comment down below and within 10 minutes I'll be able to save you a ton of money. Cheers. As you can see here, I've created a contact for myself in HubSpot. My name, my title, my company name, my email. I'm going to show you how Click can use AI to look up this record, find it, and pull in the object ID so that when we run subsequent enrichments, we're not going to create a new contact and create duplicates.

1:57 Rather, we're just going to enrich this record, but we're going to do it without having the email or a LinkedIn URL as a unique identifier. Okay, folks. Now you're getting an exclusive look inside our proprietary Clay enrichment engine that we've built specifically for HubSpot. This is a multi-step process. I'm not going to go through the whole thing, but in this part of the process, this is called part of the pre-enrichment process where we import the records. We determine if we're enriching existing records into the HubSpot CRM or if we're creating a new record. In this case, we're creating a new record. Therefore, we've imported a list with existing information. So, your imports might have an email, job title, company name, etc. It might just have a first name and last name and company name from an event list that you got that's only partially filled. In all these cases, you want a way to look in the CRM and figure out does this person already exist or not. And if so, we're going to update that record or we're going to create a new one. In this case here, you can see we have only the following first name, last name, job title, and company name. First, we're going to pull out those properties right here, and we're going to try to find existing contacts because we don't have a record ID already. We're going to use our existing input property values like first name, last name, company name, job title to create a fuzzy search query for the HubSpot API using chatbt.

3:04 Essentially, just like you can search for records in HubSpot using the search bar, they also have an API key that allows you to search for records using fuzzy match search using a variety of different properties, we have to format our raw data structure into an organized API call. The way we're going to do that is by connecting or passing in these input properties into this prompt that we built here. And this is essentially a prompt that is made for the HubSpot search query API to output a search query that we can use. So in this case you can see here we have company last name first name job title, right? And then we have these filter groups. We have the company token contains sculpted. We have the uh last name token contains TUW for my last name. Over here we have also other parameters as well.

3:47 Uh, we have the first name contains JAC, last name contains TUW, and job title contains founder. So, we're looking for all of these different things. We're now going to check out this API call that we made to the HubSpot database to essentially look and see if we find any matches. So, you can see here we're passing in the filter groups from this last property. We have all of them right here that I just showed you. And this is specifically going to return the HubSpot properties that we want from the record that we find. So we have a separate list of all the HubSpot properties that we want to return when we call on this API.

4:21 So if I open this up, you can see here we found the Jacob 200 record. I can open this up and you can see Jacob at school. All my information is right here. And essentially now we can go ahead and pull out the record ID from this record. And we're actually going to go ahead and use this as the object ID. So the rest of the enrichment workflow is actually going to now append all the enrichments to this record. And as you can see, we started without having a clear record ID or email or any other unique identifier, only using a fuzzy match search using the HubSpot API. Did we find this record? This prevents us from creating duplicate records in the CRM, which happens all the time in Clay.

4:57 And I can guarantee 99% of folks using Clay for HubSpot don't do this step ahead of time. This is a key step and it's a really easy way to prevent duplicates from entering your CRM. If you would like me to audit your Clay table, it will take me less than 10 minutes to find mistakes like this. cuz I'm sure I can find plenty of them and really turn your clay workflows around, make them much more efficient. So, if you're interested, if you're on YouTube, click the link down below in the description. And if you're on LinkedIn, you can comment on this post and or go to my LinkedIn profile where I have a link to submit your clay table in a type form. Really simple. And then I can audit it for you and even make a video like this if you wish. Hope you enjoyed the video and I'll catch you in the next one. Peace.

Summary

Jacob Tony from Sculpted discusses a clay workflow designed to prevent duplicate records in HubSpot by using AI fuzzy matching to identify existing contacts without relying on unique identifiers like emails or record IDs. This process enhances data integrity during imports and enrichments, ultimately saving clients time and money.

- The workflow uses fuzzy matching based on first name, last name, company name, and job title.
- It avoids creating duplicate records during list imports or lead entries.
- The process involves searching the HubSpot API for existing records using a structured query.
- Clients can benefit from pre-built templates that streamline data cleaning and enrichment.
- Jacob offers free audits of clients' clay tables to identify potential savings and efficiency improvements.
- The workflow is particularly useful for clients with large datasets, potentially saving significant costs on clay credits.
- The video provides a step-by-step demonstration of the fuzzy match search and enrichment process.
© transcribe · For agents Built with care and craft by Gokul Rajaram