Peec.ai Review & Tutorial 2025 — Full Beginner’s Guide

In today's video, we're going to do a deep review dive of a AI visibility tool, Peak AI. This tool helps brands monitor how visible they are on platforms like Cad GBT, Gemini,

Proplexity, and so on. This tool

basically tracks which prompts or queries trigger the mentions or citations of the brand and which external sources are being cited by the AI models. This tool also has some

AI models. This tool also has some competitive insights allowing to see in comparison to the other competitor brands how visible you are against them.

Basically, this company is a startup based out of Berlin, Germany. They

raised €7 million to basically develop this tool and they're trying to become the next SEM rush but in answer engine optimization niche. I actually spoke

optimization niche. I actually spoke with the guys who actually made this tool and their mission is to help companies adapt to the world where more and more people are using AI to do the search and to help companies track

exactly what people are searching, how many times they're searching, and what they can do to better improve their positions on these LLM models. Let's go

ahead and jump in a tool so that you can get a visual understanding of how this tool works. After you set up the

tool works. After you set up the website, um you'll be able to see this overview dashboard here which gives you a lot of information. And I'm going to jump into that here in a second. But on

the top here, you'll be able to see the brand that you're tracking. Next to it, there is a delect dates that you can actually uh select between 7, 14 or 30 days, or you can actually do custom date

if you have uh more data usage that you need to be able to track. And then on the right hand corner here, you see the AI models. You can track chat GPT

AI models. You can track chat GPT mentions, perplexity mentions, AI overviews on Google. Um if you want to upgrade to a next uh level plan which

which they have more premium plan you'll be able to track uh GPT40 search you can do AI mode that Google has Gemini and Claude. Now if you take a look at this

Claude. Now if you take a look at this visibility dashboard here you'll be able to see that this is what lets you see how visible and how cited you are and

how that has been changed over time. For

example, you could track if your brand gained more mentions after you launched the product or whatever activity that you did online. It'll be able to see those data charts. And this is updated a

lot faster than how Google rankings used to take forever to update in platforms like SEM Rush. If you click export here, you'll be able to export this entire data in CSV file and you can feed that

to your custom GPT chat mode or whatever you're using to analyze your data with.

Now going back to the visibility here.

If you go over the chart here and you'll be able to see like you know different different dates here you can see chats or answers where your uh brand is mentioned compared to the other brands

and what are those percentages. Um as

you can see right here we're seeing that we're 76.9% uh mentioned when it comes to other other competitor brands. For you to understand why this matters. This is

your share of voice inside the AI search like SEO impressions but the LLM uh answers which is the version of those impressions. Now on the right hand side

impressions. Now on the right hand side here you'll be able to see all your competitors and this shows you which competing brands appear most often in the same exact AI chats as you. Now why

this matters? This actually matters because it helps you spot who the AI considers your peers or rivals are. If a

competitor is dominating and you know where to focus on, such as like content or link building or whatever sources that they're actually getting cited on, you'll be able to replicate that pretty easily. And this tool does a really good

easily. And this tool does a really good job showing you all that information.

The next section is source type, right?

And these are your top sources categorized by the source type, corporate, editorial, other, and UGC.

And what this does is this shows you uh the sources AI modes pull from when generating responses that it's involving your brand mansion. For example,

corporate, right? These are official brand websites, press releases or company blogs. And then you have the

company blogs. And then you have the editorial which is the online magazines, news outlets or reviews. You have the UGC which is user generated content

platforms like Reddit, forums, Kora etc. and other is like catching all the things like databases or smaller unclassified sources. This gives you a

unclassified sources. This gives you a map of where LLMs learn or site you from. If your competitors are cited more

from. If your competitors are cited more often from editorial or UGC, you might want to push PR or forum discussions. If

a corporate sites dominate, if corporate sites dominate, you know, authority websites matter more. Basically, this

tells you what fuels AI visibility. On

the right hand side, you'll notice the domain table, each domain AI models cited, how often it's used, average number of times it's cited per answer,

and its type. And then type is basically categorized into you like your domain, competitor, competitor's website, other uh which is you know uh neutral or

unrelated sources, um UGC like we mentioned, Reddit, Kora and etc. niche forums and then you have corporate which is the larger companies or official domains. Now let's take a look at

domains. Now let's take a look at skinnit.com as an example. Right? We're

seeing that it's used 72% of the time, average citation 1.2. As you can see, skinn.com is used 72% and then average

citation is 1.2. Now, this tells you how often AI trusts and sites your domain versus the other competitors. the higher

your percent of usage, the more you're considered as an authorative source uh and and competitors are lower. So if a competitors are ahead of you with the

percentages, it means the LLMs lean on them more than they lean on you. Now on

the bottom you'll see recent chats here, right? So I think this is pretty

right? So I think this is pretty important section and this is a log of recent AI conversations from chat GPT, Perplexity, Google overviews and etc. So

that mention your brand and this actually lets you see the contacts and the wording of mentions. For example,

did Chad GBD say Skinnet is the leading provider in custom skins or mighty skins is a cheaper alternative? You can use this information to adjust your

messaging, FAQ content, or even PR pushes. For example, we're seeing one of

pushes. For example, we're seeing one of the examples on Perplexity here. Where

can I design my own custom gaming skins?

So this is actually you know uh very useful information. We're also seeing

useful information. We're also seeing the same exact question being asked here on Google and then we're also popping up there. There is also best websites to

there. There is also best websites to design personalized MacBook covers. I

mean you can see what level of information that we can get access to and coolest part is like it has a timelines on there 18 hours ago one day ago you know we're able to see this which is really cool. So you can kind of

jump to speed. In this next section we're going to go over the prompts.

Right. So this is the section where basically we can track uh queries that people are typing in chat GPT perplexity Google AI overviews and etc that could

bring up your brand in a response. I

want you to think of prompts as keywords of AI search world. People are not typing keywords in chat GPT they're typing prompts. If you come around here

typing prompts. If you come around here in the right top corner you can actually click add prompt and you'll be given two choices right. So you can track singular

choices right. So you can track singular prompts or you can bulk upload multiple different prompts. For example, the

different prompts. For example, the prompt could be where can I design my own custom iPhone case. Peak

specifically asks its users to avoid mentioning your own brand name in the prompt. Why is that? Because the goal is

prompt. Why is that? Because the goal is to see if your brand actually shows up organically when someone is asking a general unbiased questions. Just like

SEO keywords like uh best roofing company near me versus your company name plus roofing. If somebody types the

plus roofing. If somebody types the prompt with the brand, it's obvious that you would come up for it, but most likely you want to come up for things that you're not yet recognized for.

That's how you tap into the new audience and you want to know that intel. Now in

here when you're adding the prompts you're basically reverse engineering which AI generated answers from chat GPT Google overviews or other LLMs include

your brand. So every prompt that you add

your brand. So every prompt that you add in here becomes a tracking query. Now if

you take a look at the IP address right you can change this to different countries and what this does is this defines the geographical region or a search location where this system

actually runs the prompts on. For

example, if you're targeting the US market, you want to see the data from United States, not from the whole world and not from a different country that you're not actually having presence in.

Just by clicking this, you can actually select whichever country that you're trying to search what AI knows about your brand. The reason this actually is

your brand. The reason this actually is important is because LLMs often tailor resultbased or regional relevance, especially when it comes to Google AI

overviews and perplexity. So tracking by country shows how your visibility changes across markets. And then on the bottom here, we're seeing the tags tab.

And what you can do here is you can actually uh categorize and filter the prompts within your project. Um you can do for example like gaming skins or

laptop skins or phone skins if you know these are the prompts that we're actually going after. So that way you can kind of see your prompts categorized in a data. Now in this section where you

can see tags, tags pretty much help you categorize and filter prompts within your project. So for example, if we're

your project. So for example, if we're tracking uh custom iPhone skins, right?

So I could uh tag this to be uh phone skins or you know laptop skins if if my query is related to laptop or gaming skins if my prompt is related to gaming.

This way I can keep the data organized and I know list of tags that I can pull up later. Now let's go back to the data

up later. Now let's go back to the data column here. So we're seeing as a first

column here. So we're seeing as a first column prompt, right? So the queries you're monitoring here. Then we're

seeing the visibility, right? So 100%

means you always show up. Higher this

percentage, better your visibility is in the AI models. Now if you take a look at this prompt here, we're seeing 23%. So,

what this means is um we're mentioned in only about one in every four answers.

The visibility tab here basically shows you where you're strong versus where you're being ignored. If we move over to the next tab, this is the sentiment tab.

The score here that you're seeing, this is usually measured from 0 to 100 based on how positive or negative the AI mentions of you is. Now, as you're

seeing 76, that's positive leaning. 66

is more neutral and 83 is strongly positive. Understanding this is useful

positive. Understanding this is useful brand perception tracking inside the AI search. This is my favorite part here,

search. This is my favorite part here, the position. So, this actually shows

the position. So, this actually shows the ranking positions inside the AI's generated list or the answers. And

basically, if we take a look at 1.4, this means that you appear in a first place most of the time. If we take a look at 3.4, this means that you're

typically like the third on the list.

This is similar to SEO rankings on Google, but for the AI answers. Next to

position, you'll see mentions. And in

the mentions, it basically shows you which AI models are showing you where.

This is useful for prioritizing which AI ecosystem that you want to focus on. So,

if you're seeing the increasing amount of leads come from Chad GPT and and nothing much in the other LLM models, but you're ranking in the top positions,

you're going to want to prioritize Chad GPT because that is the most impactful channel. Now, this one is pretty

channel. Now, this one is pretty important, right? Right next to

important, right? Right next to mentions, we have volume. This is in a beta right at the second. This might

change in the future but for you to simply understand right this is a bars that represent relative uh demand here.

So more bars that you see more valuable the query is. So we're seeing here you know we have like a one red bar, we got two bars, we got three bars, we got four

bars, and then we got five bars in here.

This helps you prioritize hightraffic prompts instead of wasting your time on the ones that have low demand. And the

next section is a location. Here,

obviously, we're seeing the US here, but this is useful information because, you know, you can just do like a local versus global monitoring. For example,

if your competitors are dominating in UK, but you're strong in the United States, you know exactly where to expand. And right now I don't have any

expand. And right now I don't have any tags in here because of the example, but these tag sections, you know, help you better organize and filter these prompts by product line or category. Next

section is sources. Here we kind of went over this data earlier in the overview section, but this is dedicated section tuning out all the noise and giving you more opportunity to see all the sources

down in here. But this tab actually reveals where AI models like Chad GBT and other LLMs um are pulling the data from when they actually mention your

brand or your competitors. It's like

seeing the content backbone behind AI's generated answers. What we're seeing is

generated answers. What we're seeing is that this brand has total of 112 different sources spread across corporate, editorial, UGC and other

types. Each domain is sourced for usage

types. Each domain is sourced for usage and citation debts. Now, what we're seeing here, right, this line here, uh, this graph allows you to see the visual

representation of how the top domains usage actually fluctuates over time. You

know, we're seeing Amazon mighty skins.com. Now, we can see the skinn.com

skins.com. Now, we can see the skinn.com is the gray line in here. We're seeing

that they have the strongest consistency in AI usage, picking nearly 100% and stabilizing around 70 to 80%. This means

that AI is reliably referencing this website as a trusted source. We're also

seeing the mighty skins.com here. Uh

this is the top competitor rising and falling alongside skin at brand. So this

indicates the AIS see both of these brands as a co-relevant. Now, Amazon

highlighted in a yellow, that's more moderate, consistent presence, which is a corporate source. So, it's a default uh fallback citation. And then we're

also seeing Reddit uh it's erratic, but you know, impactful spikes. We're seeing

that on the chart here. So, it's a you know, UGC content from Reddit, which influences the brand contacts when specific discussions actually start to trend. Now, here's the really

trend. Now, here's the really interesting thing about this, right? So,

if you're seeing that your lines are rising when Reddit or YouTube lines also rise, it likely means that AIs are mixing customer opinions and product

reviews into a brand mentions. It's

great for authenticity but volatile. You

can stabilize this by ensuring that your own website content stays authorative and updated. Again going over to the

and updated. Again going over to the source type in here we have 112 sources and it's divided into the four different types corporate official brand or a

company websites editorial your articles blogs magazines news outlets UGC Reddit uh YouTube Quora etc and other is more

miscellaneous or uncatategorized data things like you know smaller stores niche forums AI aggregators this matrix tells us exactly mix of where LLM AMS

are getting the context for your niche.

If corporate and UGC dominate, you're in a brand plus community trust environment, not a pressheavy one.

Meaning PR pushes or reviews could help you influence AI generated opinions.

Now, on the bottom here, we're seeing the source table, right? Um, and and this is source, we got type of source, we got used and average citations. Now,

we have our used percentages super high.

So this is dominance established among AI. So AI actually sees this website as

AI. So AI actually sees this website as a primary fact base. We can also export this data and let AI analyze this for us. Not happening through peak AI but it

us. Not happening through peak AI but it can happen through chat GPT or whatever LLM model that you're using to fetch your data. Let me quickly tell you how

your data. Let me quickly tell you how to use this data strategically. If you

see the Reddit, YouTube, and corporate sources dominate, post more high value UGC and long- form product pages. LLMs

prefer redundancy. Appearing in multiple source types increases your permanence in responses. You can also reverse

in responses. You can also reverse engineer the AI trust. This has been working really well, and I've been doing this a lot. So, you can use the used

percentage and average citations data to benchmark who AI believes most. Then

replicate their structure. Structured

content, FAQs, guides, schema markup, domain authority, internal linking. You

can also target the influencer domains.

If a small blogs or review platforms appear in a lower ranked pages, those are AI micro authorities. A backlink or a future from this will increase your

exposure in AI responses. Lastly, you

can monitor volatility in the source usage graph here. If you see a sudden spike or a drop, for example, Reddit surging, this means that conversation

trends changed, often triggered by viral content or new products. You can

correlate this with your prompt visibility chart to detect what caused it. Now, this next section is dedicated

it. Now, this next section is dedicated to competitors, right? And what you can do here is um pigai actually gathers all the brands that are being mentioned in

the same AI generated answers as you.

You can also manually add the competitors to see how they align and how they come up in a answers against you. To add a competitor, simply just

you. To add a competitor, simply just click add competitor, type their display name here and their domain and click create. This is going to automatically

create. This is going to automatically fetch the data. For example, we're seeing the software is suggesting these companies as competitors. In order for us to track it, what we have to do is

simply click track. Or if we don't want to track it, we can just click reject.

Now, for you to understand how this section actually works behind the scenes, right? So when you add a

scenes, right? So when you add a competitor or you accept one that PKI is actually suggesting you, you can start to monitor uh whether these competitors

actually appear in the same exact answers for the prompts that you're trying to optimize for. It literally

logs every time their domain or a brand name is cited or referenced. And you can actually pull that data into the dashboard and you'll be able to see that. And you were able to see that

that. And you were able to see that earlier in the overview chart where you were able to see in the source chart mighty.com, slickraps.com, and then some of the other competitors that we had because we

already added them there. By tracking

your competitors, if you're seeing that they're appearing in more searches like we mentioned earlier, you'll be able to reverse engineer their steps and kind of

hijack the authority the LLM models are using to suggest them. The next section here is the tags. Obviously we talked about tags earlier but this is where you

come to create the tags in a dedicated section. You can also uh create tags

section. You can also uh create tags just like I mentioned earlier but here is the dedicated section where you can just create the tags so that you can better aggregate the data that you want

to see. And in the people section here

to see. And in the people section here you just see like who has access to the tool that you've given access to. And

then in the workspace again in order for you to track a project you have to create a project in a workspace. We

mentioned that earlier. Um, and here you can kind of like, you know, pick and choose what you want to track. Exactly.

This video came out a lot longer than I intended to make. It was kind of like a mini course instead of a review. But

again, from what I'm seeing right now, a lot of the things they're still working the things out. These guys are like a startup. They just got started and, you

startup. They just got started and, you know, they've built this incredible tool. So far, I like it and we're going

tool. So far, I like it and we're going to be using this tool. Just based on what I saw on SCM Rush, it definitely gives way more data than SCM Rush does.

In my upcoming videos, I'm going to be making some AI visibility tool comparisons and be on a lookout for that. And I'm also going to review other

that. And I'm also going to review other tools and we're going to determine which one is the crown winner. But so far so good with this tool. If you guys had experience with PKI, please let me know

in a comment section below.