
Marblism
AutomationHire a team of AI employees for email, calendar, support, lead generation, social content, blog posts, calls, and legal Q&A.
Overview
Marblism is an AI employee platform for small businesses that want several operational assistants in one workspace instead of building technical agent workflows themselves. Its agents can organize inboxes, draft email replies, manage calendars, create support tickets, write blog and social content, source leads, answer business calls, and use shared company knowledge to stay closer to your brand. It is most useful for routine admin, communication, and first-draft work, while creative output and edge cases still need human review.
Platforms
- Web
Video review
Prefer YouTube? Open this review on YouTube.
Video transcript
AI agents are transforming industries, but they are very technical and you have to learn how to use them. Wouldn't it be cool if we had a team of AI agents that look and act like normal humans and we just have to chat with them and give them instructions? Today, I'm going to show you a tool that gives you a whole team of AI employees that can take care of your emails, schedule meetings, manage support tickets, create social media content, and even give you legal advice and take phone calls for you. I will show you how to set this up, how to use it step by step, and then we will take an honest look how useful this actually is, and can this replace a human team. My name is Florian Walther and this is the AI tool corner where I review the latest AI software to find out which ones can actually improve our lives and businesses. To get started, go to marblism.com. I will also put the link into the video description below together with a 25% discount code and they have a 7-day money back guarantee. So, if you don't like it, you can get a refund. Go ahead and create an account and then we have to go through the onboarding flow. Fill this out with your information, your role in the business. Do you have a website? If you have one, it's really good to add it here because the AI agents can extract information from this website and then use it in everything they create like blog posts, communication. Now, I don't have a website for AI tool corner yet, so I use my YouTube channel instead, which should also work. Then fill out everything here. And don't worry, you can always change this information later in the settings. And then we can activate our AI team. We will take a look at the chat in a moment, but for now, let's go to the company settings here in the sidebar because there we can set some more values like our brand color or a logo. And again, the agents will use this information. For example, they can use our brand color or our logo in Instagram post they create. We can also change the tone of voice of our company. Let's keep it at simple. Let's also take a quick look at the media tab. Here we can upload additional images that the AI can then later use. So I'm going to upload my logo and YouTube banner. We will take a look at the knowledge base later. Also a very important feature. But for now, let's go to the AI chats. So each of these AI agents has a specific role. For example, Eva is our personal assistant. And we have to onboard her in order to use her. And here we can configure all the features we want to use. We want to allow her to draft email replies to organize our emails with label archive emails and so on. I enable all of this. Then we can configure the labels that she uses to organize our emails. I'm going to keep the defaults here. Again, we can change this later. And then we can connect her to our email inbox because she needs access in order to make all of these changes, right? We can connect her to Gmail, Outlook, or to pretty much every inbox via IMAP. I'm going to connect her to this AI tool corner Gmail account. And then we have to authorize her to give her access to our Gmail inbox. So make sure this is green and says authorized. And then in the next step, we want to connect her to our calendar as well. Of course, only if you want this. And then authorize her to manage and create meetings in our calendar. And this is also a cool feature. When there is a Google Meet link in the calendar event, she can automatically join it and take notes from your meetings. So let's enable this for our calendar as well. Now, Eva is onboarded and as she says here, she will now go ahead and organize the email inbox that we connected. And she also automatically reads our previous emails to come up with a tone of voice for future emails because she will draft replies and she will try to use the same voice that we used in our communication previously. And indeed, when we look into our email inbox, she automatically went ahead and organized our emails into these different labels. She created these different labels and labeled our existing emails. If you want to change the configuration of Eva, you can always click on edit and you see the same settings that we saw in the onboarding and you can change them here. Now let's see what happens when we send a new email to this inbox. So here I'm in a different Gmail account and I sent a collaboration request to AI tool corner. I asked for a meeting to talk about our YouTube channels at 3 p.m. GMT plus 2. I'm going to send this. Then Eva automatically labeled this email and she created a draft response that looks like this. So this was written by AI. We can fine-tune how she responds or at what times she agrees on meetings here in the guidelines. We can set rules to categorize emails. We can set a tone for the emails she writes. And down here we can set preferences for scheduling meetings. And here she basically accepted the meeting and we can now send this reply. And now when we look into our calendar, voila, we have an event with a title and a description that was automatically created for us. Now the time is wrong. My suspicion is that the AI can't yet handle time zones properly, but I already sent a message to the support to fix this. So I assume this will be fixed in the future. I would guess that this is US time zone. Anyway, since Eva is our personal assistant, we can also give her little task to do. For example, I can tell her to remind me to take my creatine every morning at 10:00 a.m. And now, Eva will automatically send us a reminder every day at 10:00 a.m. via the chat. Next, let's take a look at the blog writer, which we also have to onboard before we can use it. We can select a time, every day, where she will send us blog post ideas. Let's say 3:00 p.m. Then, she immediately goes ahead and generates a bunch of blog post ideas that we can start with. And as you can see, she takes a look at our website or YouTube channel in this case to get a feel for the style and the visuals that we use so that everything is on brand. And a minute later, she sent us our first five blog post ideas. And now we can ask her to actually write the full blog post. For example, I like idea one. Please go ahead with it. And there is our first blog post. We can ask her to make changes. We can also edit it ourselves, of course. And if we like it, we can go ahead and publish it either to a site like WordPress or Wix directly or we can also copy the markdown or HTML to post it on any website. So this AI writes blog post for you automatically. And in the guidelines again we can customize her writing style and the images it creates because it also creates images that it puts inside these blog posts using AI. Of course, similar to the blog post writer is the social media manager, which creates content for sites like Instagram or Twitter. So, let's onboard him as well. The concept here is the same as for the blog post writer. You get a bunch of ideas, you pick an idea, and it automatically generates a post which you can edit. You can post it to one of these different social media accounts that you can connect or you can schedule it for later. And there's also an AI generated image attached to it. Here we have this content calendar where we can see an overview of all the posts we are going to send in the future. Next, let's take a look at the customer support. And by the way, you can set up different workspaces here. And every workspace gets its own set of AI agents. So for every company or every project you have, you can set up a separate workspace and each workspace gets their own Eva, their own social media managers and so on. Now, let's onboard Cara, our customer support as well. We want to connect her to our email inbox, of course, the same one that we used before, and authorize all the access she needs. There we go. And now she works automatically behind the scenes, just like Eva. For example, I can send a request to make a video about Marblism. The AI detects this as a support request and automatically creates a ticket here in the tickets tab. It automatically creates a draft response that we can now send via email. We have this internal chat where we can chat with the AI about this ticket and tell it how it should handle it. Let's send the reply. The user receives the reply and let's say we are happy and we send our final response. The user reply gets added to the ticket thread and then it came up with another draft response. But we don't want to send this reply. We want to tell the AI to summarize the conversation for us and then we mark this ticket as resolved. So there we go, a fully fledged automated ticketing system. Let's take a look at the sales associate next. Let's onboard him again. We give him access to our Gmail account. We let him source new leads every day at 3:00 p.m. or whatever time you prefer. And we also let him automatically contact these leads. You can let him send follow-up messages automatically. I'm going to keep this disabled. And I connect my inbox. Now the job of this AI is to find other business people in our industry that we can reach out to. For this it analyzes our target customer based on our company information and our website. And then it generates a bunch of email templates that we can look at or modify. Then it asks us if it should start sourcing leads. Yes, please go ahead. And then in the leads tab, we can find a bunch of leads that it found on sites like LinkedIn or Facebook. Those are people in our industry. and we can start sending cold emails to them. And it asks us if it should go ahead and contact these leads. But instead, I tell it to please reach out to Florian at codinginflow.com first. That's me. He might be interested in our videos so that we see the whole contact flow in action. And look, it added my information into the leads tab. So, it found my name and the company name and even where I live on LinkedIn. And it added me to the list of leads and automatically sent an email to me. So let's take a look at this email. This is the cold email that we got. And now let's say I respond positively. Sounds good. Please send it to me. Now the status of this lead changed from contacted to replied. And we get a summary in the chat which then tells us that we get a positive reply from Florian Walther and it asks us how we want to proceed. So this is great for cold outreach and you can play around and customize this to your industry. Next, let's take a look at the legal associate, Linda. This one doesn't need onboarding. This one is pretty simple. It basically answers legal questions. For example, I asked it how I can register a trademark for AI tool corner and it gives me the necessary information. Now, I could follow up and ask it to check for the availability of my trademark. This is the idea behind this legal associate agent. The last agent is Rachel. And this one is really cool because she can take phone calls for us. Now, right now, I think this only works in the United States, but I know that they are adding more countries in the future. So, I select the US even though I live in Germany. You can customize the greeting message, and we will try this out in a moment. You can change the ambient sound. For example, there's a Zen garden or coffee shop in the background. And you can add rules how she should handle conversations. Let's keep the default. Under which conditions should I end a call when the caller says goodbye or when it seems like a spam? Call sounds good. Let's keep this. And we can let her automatically transfer important calls directly to us. The AI will decide if a call is important or not, but I don't have a US number, so I'm going to turn this off for now. And we will allow her to book appointments in our Google calendar. And now she has a real phone number that we can put on our website, for example. and that customers can call. Let's try this out.
Rachel: Hi, this is Rachel from AI Tool Corner. How can I help you today?
Paul: Hey, Paul Jackson here. I was wondering what is your business about?
Rachel: Hi Paul, thanks for calling. At AI Tool Corner, we review the latest AI software to see which ones can actually make life and business better. We share our findings through YouTube videos, blog posts, and tweets. Basically, we help people figure out what AI tools are worth their time. If you're curious about anything specific, just let me know.
Paul: Sounds good. Can you forward my call to your CEO directly?
Rachel: I can't forward calls directly to our CEO, Florian, but I'm happy to help you set up a meeting or pass along a message if you'd like to book a time with him. Just let me know what works for you.
Paul: Okay, let's schedule a meeting for tomorrow 2 p.m. if possible.
Rachel: Let me quickly check if available tomorrow at 2 p.m. for you. Florian looks free at 2 p.m. tomorrow.
Paul: Okay.
Rachel: Can I get a quick confirmation of your phone number?
Paul: My phone number is 1 2 3 4 5 6.
Rachel: Thanks, Paul. I've already got your number as plus then 602327912 from your call, but I'll make a note of the one you just gave as well. You're all set. Your meeting with Florian is scheduled for tomorrow at 2 p.m. If you need to change anything, just let me know. Have a great day.
Paul: Thank you very much. Goodbye.
Rachel: You're welcome, Paul. Have a great day. Goodbye.
Now, we have a new meeting with Paul Jackson in our calendar. And the AI gives us a summary about the phone call, the number he provided, the number where he called from, and a summary of what we talked about. This is really cool. Now, obviously, this is not perfect, and people notice that this is an AI and not a real human, but it's very low-hanging fruit because you might not want to take phone calls yourself, and you might not want to hire someone. So, it's easy to just plug in an AI and let her handle it. It's better than completely losing these customers. Right now, the knowledge base tab is also an interesting feature because here you can provide information for your agents to use. Now, I went ahead and added some examples like frequently asked questions, scheduling preferences, and some notes about our brand tone. And then we can decide which agent should have access to this information. By default, all agents can see it. But the scheduling preferences for example might not be relevant to our social media manager. And really only the personal assistant and the receptionist need access to this information. So let's change the visibility to just these two. And now whenever these AI agents do their work, they will automatically read and reference this information and let it guide their decisions. You can also click on how does it work to get to the knowledge base. Here you can find extensive documentation and instructions. And in the request features tab, we can see which features might be coming in the future. Now, let's talk about how useful this tool actually is and can it replace real employees. I really like the organization features like automatically labeling emails or creating drafts for responses or handling support tickets. I think this is really useful for small business owners who want to save money and don't hire personal assistants to do these simple tasks. I'm more skeptical when it comes to creating social media content because I think you can smell AI generated content from a mile away and I would not post this as it is on my website or social media accounts. But you can use this as a first draft or to generate content ideas and then either you or someone else can fine-tune and edit these posts to give them a human touch. This can work and also the AI learns from your previous posts. So this is a feedback loop where editing a post makes it better and then the AI learns and the process gets easier and easier. So I could see this working. I would just not post it as it is. Now while using the platform, I also encountered a few bugs here and there, but I'm sure they will fix them in the future. And of course, the LLMs themselves also get more advanced, so we will get better and better AI assistance in the future. If you want to try out Marblism, I will put a link into the video description together with a 25% discount code. Remember, they have a 7-day money back guarantee, so why not try it out yourself? Please subscribe to the channel for more AI tool reviews in the future, and then I wish you a nice rest of the day. Take care.
Standout features
What it's great for
- Automatically organize a business inbox and prepare email replies for approval
- Handle basic customer support requests with tickets, summaries, and draft responses
- Generate blog post and social media drafts from company context and previous content
- Source leads, send personalized cold outreach, and monitor positive replies
- Let an AI receptionist answer calls and book appointments when no human is available
- Run separate AI teams for multiple businesses or projects from one account
Pros & cons
Best for
Verdict
Marblism is strongest as a practical AI operations layer for small businesses that want help with inboxes, support, scheduling, calls, and lead follow-up without building agents from scratch. It should not be treated as a full human replacement, but it can reduce routine work when you keep humans in the loop for approvals, creative polish, and sensitive decisions.
FAQ
What does Marblism do?
Marblism gives you a team of role-based AI employees for business operations, including inbox management, calendar scheduling, support tickets, blog writing, social media content, lead generation, phone calls, and legal-style questions.
Can Marblism manage my email inbox?
Yes. You can connect Gmail, Outlook, or IMAP inboxes so the executive assistant can label emails, archive messages, draft replies, learn from previous communication, and help schedule meetings through your calendar.
Can Marblism answer phone calls?
Yes. The AI receptionist can receive calls through a real phone number, answer basic questions about your business, collect caller information, book appointments, and send you a summary afterward. Availability depends on the supported countries and call setup.
Is Marblism good for social media and blog content?
It can generate ideas and first drafts for blog posts and social media content, including images and scheduling workflows. The output is best reviewed and edited by a human before publishing, especially when brand voice and originality matter.
Can Marblism replace employees?
It can replace some routine admin steps, but it is better viewed as an assistant layer than a full human team replacement. Email drafts, tickets, reminders, lead summaries, and call handling are useful, while creative content, outreach, legal questions, and scheduling edge cases still need oversight.
