8 Powerful Grok Bot Routines to Automate Your Business

Moritz | AI Systems· 16 min· 3,227 words· 15 min read· English ·Watch on YouTube

This is the full transcript of 8 Powerful Grok Bot Routines to Automate Your Business, published on YouTube by Moritz | AI Systems. Every paragraph carries the moment it was spoken, so you can click any line to jump straight to that point in the video, search the whole thing for a word, or copy it out.

0:00I've been running Rockbot for the last 2 weeks, and the best thing about it is how easy it is to set up routines, which are essentially automations. In Claude or ChatGPT, setting up a routine always meant you have to first decide whether it's going to be a local or a cloud routine, then you had to configure all of your connections, and if it's a cloud routine, then you would have no computer in the cloud. And don't even get me started on Integromat or Make, which are just totally outdated tools for setting up automations. But in Rockbot, it's super easy. You just basically tell it what routine you want, it will set it up

0:29for you end to end. They all run in the cloud by design, which I think is the most natural way to do this, and they all have access to your connected tools and a computer in the cloud. So, this in my opinion really makes Rockbot the best tool to set up all of your automations. In this video, let me show you eight powerful Rockbot routines that you can

0:47set up today to automate your business. All right, I've put the eight Rockbot routines that I'll be covering into these three categories. First one is marketing/content, the second one is sales, and then the third one is general product- ivity. So, let's just go ahead and get started with the first routine, which is the YouTube competitor analysis. All right, so I've got my Rockbot app opened up here, and you'll see kind of how I organize my bot. I have a chief of staff bot, who is like my obviously main bot, the one I that I talk to, and the one that coordinates my other bots. And then I

1:19have this section here for the uh content-related bots, and these are my YouTube bot, my X bot, my newsletter bot, and my short-form bot. And then one group here which has all of my content bots in one group. And then I have my other bots in this unassigned section here, so these are my agency bot, uh my daily planner bot, life bot,

1:44school bot, and a coding agent as well. So, that's just to give you a quick overview of the bots I have, but for this routine, we'll be looking at the YouTube bot. And you'll see here on the right side, I have this YouTube inspiration routine set up, and if you look at the instruction of that routine, uh it basically says produce more of this daily YouTube content inspiration brief. The outcome is the is newest long-form videos from track channels with thumbnails, plus how each one is performing against its own channel baseline. And it should use the Notion MCP. I'll get into why in a bit. And

2:20also to get the YouTube data, use public YouTube data basically just to keep it simple. And then to add some additional rules, so it should be long-form only. I'm really only interested in the long-form videos, not in the short-form videos. Um it should always include thumbnail, title, and a link to every video. Uh thumbnail is very important because in the first run it actually didn't include that, so I had to correct it. And then uh it should always be recent videos. So to make sure that the report is new, it looks at the last 7 days only. So the last 7 days of published videos. And then it goes on

2:56with a couple more specifications, but let me just show you the output of this. So what it does for me is it creates this Notion page, and you can see I've had it running for a while, so it has produced quite a lot of pages already. And if I click into the latest one, you'll see that uh this is essentially, you know, scanning a pre-determined list of YouTube channels that I've given it. And then out of those channels is just identify find the newest videos and kind of putting the thumbnail of that

3:27video here. It's putting the title here. And then it's doing a little like comparison and analysis on the performance of that video. So it says here it's 7 17K views on this particular video by Nate. And um it kind of gives this baseline here, which is like 4.3x this channel's usual. However, YouTube did recently change the way that the views are measured, so it's kind of indicating that here it's like mixed errors uh mixed error, and so this uh the views actually might be higher than uh than normal. And then it's also giving me a link to the YouTube video so

4:05that I can check it out if I want to. And yeah, it does this for several other channels as well, and yeah, it just makes sure that I get a really good overview of everything that has been happening in this uh space that I am am in. And obviously, if you set this up for yourself, you will be specifying your own YouTube channels, so the the competitors in inside of your niche, and this way you get a really good scan of what's happening out there in terms of the the content that's being produced in your niche. And as someone creating content, you'll know how important it is

4:36to get inspiration from what other people are doing, because that's usually the starting point for you to think about what next uh video or type of content that uh you're going to make. Now, why did I decide to have all of this created inside of a Notion page? It just keeps things nice and organized. I could obviously have uh the simple way would have just been to have the uh output being generated here in the chat so I can just, you know, whenever this routine runs, Grock Bot will just tell me inside of this chat like what the output of of the run is. Um but then I I

5:10think quite a lot of information would uh get lost over time. So, this is just a really good way to organize it. Obviously, you can do the same thing inside of a Google Drive, uh and uh wherever you prefer to have your information. But I really like to use Notion, and so I just decided to have it all logged in here. All right, the next routine is the audience tracker. I've set up this routine inside of my content team group, so basically, this is the group where all of my content bots are inside of. So, you can see it's set up here as a routine. I don't think it

5:42actually matters where exactly the routine is set up, so you can set it up inside of the group, or inside of the bot itself. Um it's just like where the routine is set up, it is it is where the routine will basically run. So, you can see here, today it ran inside of the group, so it will basically post this message here. And the routine essentially just says update the audience tracker Google Sheet with today's number. So, what this routine does is it goes across all of my social media channels and pulls the latest um subscriber numbers or follower numbers

6:17and puts it into this Google Sheet here. So, you can see here uh every day there's an entry, and it pulls my YouTube subs, my YouTube views, Instagram, TikTok, Twitter, newsletter, and school. And then it adds everything together and gives me a total audience. And this is just a really good way for me to kind of keep track of my metrics and see how things are moving. Now, how does it actually get this data? It varies by channel, but uh for most of them it actually just goes and gets the public data. For example, Instagram, it will just basically like fetch the public data using the

6:53uh tools that it has available and kind of just like goes and fetches that. Uh for TikTok as well, I think. For Twitter, I've given it access to the Twitter API, so it can go directly through the API to get that data. For my newsletter, I've also hooked it up to the uh Beehiiv MCP, and so it can also go and get that data directly. But yeah, basically you can just run it once and have it try to figure out by itself how to get that data, and then if it doesn't work for some channels, you just steer it a bit, or you look for alternatives

7:23on how it can how you can help it get that data. And then the way it writes to the Google Sheet is just very simple. I have in my plugins here, I have Composio uh hooked up, where I try to do everything uh through Composio now because, you know, you can basically centralize all of your connections in Composio, and that makes it very portable. So, I have my Google Drive and Google Sheets hooked up in Composio. So, you can see right here, Google Sheets and Google Drive is connected in Composio, and this way it's able to just use my Composio connection and then my Google Sheets connection to create that

7:59Google Sheet. All right, the next one is my tweet ideas routine. So, this is a very simple routine that just essentially gives me tweet ideas every day at a specified time. So, I specified it to give me those ideas at 11:00 45 a.m. and 3:15 p.m. And you can see the output here of that. It's basically just like giving me tweet ideas. And the cool thing is that it's based on what I've been doing that day. And how does it

8:27know what I've been doing on that day? It Well, I've previously set up a time tracking skill, and it's very easy to use. It essentially just lets me call this skill here called time tracker, and then I can say things like start, and then just like tell it my activity, so YouTube video filming, which I just did like a couple minutes ago. Start YouTube video filming, and it will then start tracking the time. And when I'm done with that, I can, you know, just type here stop. I can log times retroactively as well. And so, this way it kind of just keeps track of what I've been doing throughout the

9:02day. And this, you know, gives me a bunch of other benefits down the line because I have all of this data and then can analyze this. And so, this is kind of the source for my tweet ideas routine as well. So, my tweet ideas routine essentially just takes whatever I've logged that day, kind of looks at what I've been doing, and then gives me ideas

9:22what to tweet about based on that input. So, yeah, I think it's pretty handy because it just, you know, reminds you to tweet. And my tweet st- strategy right now is essentially just I want to try to have it as like a uh stream of, you know, consciousness type of thing where I just tweet whatever I'm doing right now. And uh this is really helpful in that respect. Okay, let's go into the next routine, which is the sales call analysis routine. And what this routine does is it will essentially pull my sales calls through that were recorded through Fathom, which is my call recording tool. And every morning when

9:56it runs, it will look at the previous day of, you know, which sales calls have happened and then analyze those sales calls based on the transcript and give me a kind of recap of what I've been doing well and what I can improve. And that is based on a framework that I've passed it uh by Hormozi, the Hormozi and the Closer framework. I've kind of given it that context to do that analysis for me. And again, I like to use Notion to keep track of everything here. And if I go into one of these sales reviews documents, uh you can actually see how it is broken up. Now, this one is a demo

10:32one, it's not a real person, just for demo purposes. But essentially, this is how it would look like. You have like this call info section, you have an executive summary, and then a closer scorecard. So, uh what I've been doing well on these closer criterias here. And then, you know, gives me like a Hormozi lens, what went well, missed opportunities, uh key moments, so quotes from the actual call, and then kind of talk ratio, and also action items. So, if you do sales calls, this is just a really good way to kind of get that constant feedback loop. And after every sales call, kind of know, okay, how did

11:08it How did I do there? Or how did my team do there? And what can be improved? Now, this next routine is kind of similar, but it sort of happens before the sales call, and it's the prospect research routine. And what this does is it will actually go and look at my calendar for the day and check out whether there are any sales calls planned on that day. And if there are, it will pull the information from the person from the prospect and and do some research about that person and then give me that research. And you can see here the output of that research kind of

11:42looks like this. It gives me a brief on the person like what or on the call when the call is happening, uh who that person exactly is, um why this call is happening, whether they've, you know, given any information before the call in the form, for example. It will do some research and kind of check out the website of the person and check out their business and just give me some information about

12:05their business. And then, you know, add some other um information here like whether it's a likely fit, what to ask, and whether there's anything to watch out for. And then also give me some useful links that I can click on to get some further information. So, this is just a really useful way to get some more information about your prospect in a very time-efficient way, and this should then increase your likelihood of, you know, building rapport with the person and also increase your likelihood of closing. Okay, moving on into the productivity routines, the first one here is the bot team stand-up. So, I've set up this group here called the bot

12:42team, which essentially just has one group where all of my bots are inside of, and I kind of want them to act like a real human team where every morning they just talk about what their job is for the day, what they're working on, and to keep everyone updated, uh you know, like unblock any blockers that they might have, and uh also to keep me updated. And yeah, what this routine will do is basically just to have that stand-up. It's triggered from my chief of staff, and then every morning my bots will have this conversation with each other about what they're doing this day and what their

13:16priorities are. And then now I've actually also set up a log of these stand-ups, so it will uh just put into a Notion database here what they've been talking about on that day, and I can see kind of what each of my bots is up to. All right, the next routine is the nightly improvement routine. So, this routine is also set up with my chief of staff, and what it does is essentially just runs like a nightly kind of check of its own memory and what has what has been happening throughout the day, and kind of suggests some some improvements on like the files that that have been

13:51created or some of the tasks that have been done and things that are left over or like small mistakes that have happened throughout the day and just improve them. So, here it says, "Do a scan read only first this bot's memory profile and log." And if you didn't know, in Grok Bot actually everything is being stored inside of the file system here, so you can actually go in here and kind of check out all of the things that

14:15is it has stored over time about you. And these are also the files that it will look at in the during this nightly improvement routine to just go into those files kind of have a look everything is, you know, organized in the right way. And so, you can see here some of the outputs of this nightly improvement routine, and this was actually kind of running while I was on holiday last week where it was just running in the background and doing like small improvements. So, here it kind of dropped a leftover Claudia drive note that still said topics were being

14:45processed. Here it just did like a minor fixed a typo in my own rules, so it can be really tiny stuff as well. And it just kind of like um improves its own setup and every mistakes every night, which is a really good thing to have. Okay, now let's have a look at the last routine, which is the weekly time tracking analysis, and you can probably already tell what this is. Again, I have this set up with my chief of staff here, and yeah, as the name says, it goes through my time tracking data and gives me a weekly review of that. And so, I've showed you earlier

15:19how I track my time on a daily basis, and it essentially just goes through all of those files that I've tracked over the week and gives me this report, which allows me to have a look at, you know, how much time I've spent working. So, this is just small data. I don't want to give my actual data out here, but um you know, I did like a mock analysis here so that I can see, for example, how much time I'm spending on YouTube, how much time I'm I'm spending on shorts, on my community, and things like that. And it just gives me a really high-level overview of where I'm spending my time

15:51and what I can optimize for the following week. And that's it for this video. If you want to implement these routines for yourself, um I'll be putting all of them into my school community, so you can basically just take them from there, copy-paste them, and implement them in your own business. And uh if you want to work with me directly, go and check out

16:08joinagentos.com. I'll be putting a link below as well as the other links. All right. Thank you. See you in the next video. Bye-bye.

Where these words come from. This is the caption track YouTube holds for this video, written automatically by YouTube rather than by the creator. We read it, tidied the line breaks and laid it out so it can be read. The plain text version is at https://viewrankai.com/tools/youtube-transcript/0fyVaDintWQ.txt.

All rights in this video belong to Moritz | AI Systems. Watch it on YouTube. If this is your video and you would rather this page did not exist, tell us and we will remove it.