CRMChat 101 | Lesson 6: Finding New Prospects and Preparing the Database for Outreach

[0:01]Hey, it's Serge. So in these couple of videos, we'll show you the exact step-by-step guide to Telegram outreach using CRM chat. So the first step of the process is finding our prospects for for the campaign. And to do that, we have to first find a relevant Telegram group or chat, which we can then parse to extract numerous database. Uh for example, here on the screen, you can find the example the example of the table uh that we that we parsed. So here we have a Telegram username, a user ID, and a profile description. And now let's explore how we can find uh relevant groups for to to parse later. So, uh for example, for this grip, uh what I did is I just very straightforward, did a research on Telegram. Uh let's say I'm doing consulting services for Web3 companies. And I know there is a conference called Epic Web 3, uh which I just uh typed in in the search bar. And I found here the the chat, uh the open chat of of the conference. I can see that members here are opened, uh which means that we can parse this uh we can parse this chat. So once I have that, I go to Serum chat, uh to the outreach tab and find group parser. So here you type um you paste the the link to the group, and in 24 hours we'll send you back uh the the table like this. Uh so this is what you will receive. And also, if you don't have any groups in mind, there is another way to find Telegram groups. is by going to Serum chat website to Serum Chat AI Telegram Group Finder, and here you can submit your industry keywords. So you can submit your keywords like crypto, Web3, investments, marketing, uh user acquisition, whatever it can be. Then you submit your Telegram username, and in 24 hours we'll send you the list of groups that might be relevant for your search that you can then upload to Serum chat uh to parse. So here I have my table, uh with with my prospects. And the only thing I have to do is extract the first name from this um name column. So this is a list of Telegram names for for every user, but obviously, not everyone has their real first name in this um in this section. So we have to find uh people who do have a real name and we have to extract it, uh in order for us to be able to use it in our campaigns and to use it as a snippet. So we could um include the first name in our messages so it didn't look like a spam, but like a message that we wrote manually. So there are two ways to do that. One of them is free, uh but it requires a little bit of manual work. So it probably works if you have like under 500 contacts in in the table. If you have more, then it makes sense to use AI to just type a prompt and uh use it for all the cells and it will um make these changes um accordingly, but of course it would cost you like 10 or 20, uh but if you have a large set of data, it probably a good investment to make. So let's explore the first way, uh the first way which is free. And um to do that, we pick a column, we go to data and we click on text to column. By the way, I'm not saying that this is the best way to do that. If you know any better way, please let us know. Um, so you pick a tab, a semicolon, and space and we can see that already we have some first names extracted here in this table and then we have to click on finish. Yeah, there is already data, it's okay to overwrite it because we don't need it. Okay, so we can see that we already have some first names here, but some of them are obviously not real, so we just have to leave these um leave these cells empty. So M is probably not a name, um J is probably not a name here as well. Oh, here I have Mathias, um which could be yeah, it's probably surname, so. Um so yeah, you get a point, you have to still review manually the the this column and see if this is actually a real first name. So and now let's see how it works with AI. So I've used uh two tools. One of them is numerous AI and one of them is GPT for Excel. Uh you can obviously find find another ones here in the add-ins section. There are a lot of them. If you just type like GPT, you'll find at least 10. Uh yeah, so you can you can just use any of them. Um but anyway, I think the prompt that you will use is uh is the same for any AI because everywhere uh the same the same models are used. So um with numerics, you just put the the prompt uh here in the cell. And with GPT for Excel, um you type a prompt here in the in this um separate section. So I will attach the prompt uh in the video description so you can so you could use it uh for yourself. Um yeah, but as I said, I would recommend GPT for Excel because I love the uh the data that that it returned, uh and it looked more um more natural to me sometimes. Okay, so now I have um checked the whole uh list. Um by the way, if you're not sure that if the name is real or not, it's better to erase it. So you better not use a first name, then using uh either a not not real name or the wrong name, which will completely um ruin the uh the perception of of you in in the face of the recipient. Um so the last thing that you have to do is to save the file as CSV and it's ready to uh to be imported to.