Google Bard Statistics By Traffic, Visits, User And Demographics

Introduction

Google Bard Statistics: The Google Bard Statistics gives important insights into how people search online today. Google Bard, a smart search platform using AI and natural language technology, has been getting a lot of attention lately. Recent data shows that its usage has grown by 35% in the past year. This increase shows that more people are relying on its advanced features. These numbers suggest a big opportunity in the market, with expected revenue growth of $500 million by the end of this year.

These statistics highlight how Google Bard influences the digital search world, giving businesses key information to adjust their strategies and stay competitive in today’s fast-changing market. If you haven’t added the Chrome extension yet, check it out!

Editor’s Choice

• Google Gemini receives around 275 million visits each month on average. 31.10% of its users are between the ages of 25 and 34.
• The United States has the highest share of traffic, making up 14.6% of total visits.
• People from over 230 countries and regions can use the Gemini chatbot.
• Google Bard officially launched on February 8, 2023, and opened to the public on March 21, 2023.
• The name “Bard” is inspired by Celtic storytellers known for their skill in telling stories.
• Bard is based on Google’s LaMDA model, which contains language models with up to 137 billion parameters.
• The estimated cost to run each Bard search is between $0.006 and $0.031.
• Google Bard Statistics stated that more than 76% of the traffic to Gemini comes directly, while 16.77% comes from organic search.
• In 2023, Google Bard had an average of 142.4 million users each month. When it first launched, 62.6% of the users were from the United States.
• The platform’s users are 40% females and 60% males. About 40% of users search for topics that interest them on Google Bard.
• Each search on Google Bard costs between $0.006 and $0.031.
• The Infinite dataset used by Bard contains 1.56 trillion words. Half of the data used to train Bard comes from conversations on public forums.
• The United States brings in the most traffic, making up 21.66% of global visits to Google Bard.
• Bard is available in over 180 countries and regions. In May 2023, 380,567 devices were responsible for all the traffic to Google Bard.

What is Google Bard?

Google Gemini, previously known as Bard, is an AI chatbot tool created by Google to mimic human conversations using natural language processing (NLP) and machine learning. It not only enhances Google Search but can also be added to websites, messaging platforms, or apps to give users natural, human-like answers to their questions. Gemini is a set of advanced AI models that can understand language, audio, code, and video.

Gemini 1.0 was launched on December 6, 2023, and was developed by Google’s DeepMind team, which focuses on cutting-edge AI research. Google co-founder Sergey Brin played a key role in developing Gemini, along with other members of the Google team.

At its launch, Gemini was the most advanced AI model at Google, powering Bard before it was renamed and replacing the company’s previous AI model, Pathways Language Model (Palm 2). Like Palm 2, Gemini was incorporated into several Google technologies to offer AI-powered features.

(Source: alternatives.co)

Jan 2023	Development Phase and Initial Research	Bard is regularly tested and enhanced by Google’s AI research team.
Mar 2023	Limited Beta Launch	Bard is initially opened to a small group of invited users for testing and feedback.
May 2023	Closed Beta Expansion	Following positive feedback, Bard extends its beta program to a wider range of users from various groups.
July 2023	Public Beta Launch	Bard is launched in open beta, allowing anyone interested to try it out.
Sep 2023	Initial Feature Updates	Bard receives its first big updates, such as improved search options, data visualization tools, and more natural conversations.
Nov 2023	Integration with Google Products	Bard begins connecting with other Google products like Google Sheets and Docs, making it simple to transfer and work with data.
Oct 2023	Wider Language Support	Bard adds more languages and dialects to support a wider range of users.
Dec 2023	Focus on Personalization and Accuracy	Bard’s algorithms are updated to personalize the user experience better and make its answers and information search more accurate.
Jan 2024	Export Functionality	Bard adds a new feature that allows users to export their query results directly to Google Sheets for better analysis and organization.

On December 11, 2024, Google released Gemini 2.0, an updated version of the AI model, and introduced an experimental version of Flash 2.0 in the Google AI Studio and the Vertex AI Gemini API. Gemini’s NLP abilities allow it to understand and process language, including interpreting user queries and data. It can also recognize images, including charts and figures, without needing external tools like optical character recognition (OCR). Gemini can translate across many languages, making it useful for tasks in different languages.

What sets Gemini apart from older Google AI models is that it is multimodal. This means it is trained to handle various types of data, such as text, images, audio, and video, all at once. Because of this, Gemini can reason across different data types. For example, it can analyze handwritten notes, graphs, and diagrams to solve problems. The Gemini system can directly process text, images, audio, and video in a mixed sequence.

General Google Bard Statistics

• Google Bard was officially launched at the Google I/O event in 2023.
• Google says Bard regularly beats its competitors in several AI performance tests.
• Bard has reached an estimated 1 billion users, which means around 1 out of every eight people worldwide might have used it.
• Bard is not only used for basic search tasks but also for things like educational tutorials, personal stories, idea brainstorming, and even music creation.
• Users have said that Bard is great at having smooth and interesting conversations.
• The training data for Bard is 750 GB, and it includes 137 billion parameters and 1.56 trillion words.
• Bard gets about 340 million visits per month, and it had 30 million users in its first month.
• On average, people spend about 30 minutes per day using Bard.
• Google Bard Statistics stated that Bard is available in 46 languages and over 238 countries.
• Most users are from the United States, making up 22.9% of total traffic. Each search on Bard costs between $0.003 and $0.028.
• In 2023, Google Bard had an average of 142.4 million monthly users. When Bard first launched, 62.6% of those users were from the United States.
• The platform has a mix of users, with 40% female and 60% male. About 40% of Bard users actively search for topics that interest them.
• Each search on Google Bard costs between $0.006 and $0.031. The dataset called Infiniset, which powers Bard, includes 1.56 trillion words.

(Source: meetanshi.com)

• Half of the data used to train Bard comes from conversations on public forums.
• The United States has the highest traffic to Google Bard, accounting for 21.66% of all visits. Bard is available in over 180 countries and territories.
• In May 2023, 380,567 devices were responsible for all traffic to Bard.
• Google Bard officially launched on February 8, 2023, and opened to the public on March 21, 2023.
• The name “Bard” is inspired by Celtic storytellers known for their skill in telling stories.
• Bard is based on Google’s LaMDA model, which contains language models with up to 137 billion parameters.
• LaMDA has been in development since 2017, which means Bard is based on a project that has been worked on for 6 years.
• The data used to train Bard is 750 GB in size and includes content from websites, books, and user-generated content, making up 1.56 trillion words.
• Google Bard Statistics stated that Google Bard is estimated to reach 1 billion users, or about 1 out of 8 people worldwide.
• It’s available in more than 230 countries and territories and supports 43 languages.
• The gender breakdown of users is 60% male and 40% female. The United States makes up 37.24% of all traffic, followed by India with 9.56%.

Google Bard Traffic Statistics

• The United States generates the most traffic for Google Bard, accounting for 21.6% of the total.
• India is the second-largest source of traffic, with 8.34%, followed by Brazil with 3.98% and Mexico with 3.47%.
• Here is a table showing the top countries by Google Bard traffic as of 2023:

Top Countries	Share of Google Bard Traffic
Colombia	3.31%
Mexico	3.47%
Brazil	3.98%
India	8.34%
The United States	21.6%
Others	59.24%

• In the United States, 90.74% of Google Bard traffic comes from mobile devices, and 9.26% comes from desktops, according to data from Semrush.
• In the United Kingdom, all of Google Bard’s traffic (100%) comes from desktop devices.
• Here’s a table showing Google Bard traffic by devices in different countries:

(Reference: answeriq.com)

Country	Mobile Traffic	Desktop Traffic
Australia	0%	100%
Canada	0%	100%
Ireland	100%	0%
The United Kingdom	0%	100%
The United States	90.74%	9.24%

• In the United States, 365.9K devices were used to access Google Bard. Since the US is the biggest market for Google Bard, it has the highest number of devices sending traffic to the platform.
• Next, the United Kingdom used 3.8K devices to visit Google Bard, with Ireland not far behind at 3.1K devices.
• Here’s a table showing how many devices from different countries accessed Google Bard:

Top Countries	Devices used to send traffic to Google Bard
Australia	1.5K
Canada	1.9K
Ireland	3.1K
The United Kingdom	3.8K
The United States	365.9K

• In May 2023, a total of 380,567 devices were used to visit Google Bard. Of these, 336,322 were mobile devices, and 44,245 were desktops, based on Semrush data.
• Here’s a table showing the monthly breakdown of the devices that sent traffic to Google Bard:

Month	Total Devices	Mobile Devices	Desktop Devices
May 2023	380,567	336,322	44,245
April 2023	154,077	95,049	59,028
March 2023	95,564	37,156	58,408

• Visitors to Google Bard view an average of 5.54 pages per visit. According to Similar Web, users typically look at 2.65 pages and spend about 4 minutes and 32 seconds on the site.
• The study also shows that Google Bard has a bounce rate of 29.97%.
• Furthermore, 88.4% of the traffic to Google Bard comes from mobile devices, with the rest (11.6%) coming from desktop devices, according to Semrush data.

Goggle Bard Estimated Costs

• The estimated cost for a query on Google Bard is around $0.003 per inquiry. On top of that, there’s an additional charge of $0.003 to $0.028, depending on the query. So, the total cost for each Bard search is between $0.006 and $0.031.
• These are just estimated costs, and the actual price for each Bard search can vary. This depends on factors like how complex the question is, how long the response is, and the computing power Bard needs to process the query.
• However, the estimated costs show that the launch of Bard has increased Google’s expenses significantly.
• According to Morgan Stanley, Google will need to spend $1.2 billion for every 10% of queries that get AI-generated responses of 50 words or more.

(Source: 9to5google.com)

• Here’s a table that shows the costs for generating responses from Google Bard with different word counts.

% of Google queries handled by AI	25 words created each query	50 words created each query	75 words created each query	100 words created each query
50%	$3 billion	$6 billion	$9 billion	$12 billion
40%	$2.4 billion	$ 4.8 billion	$7.2 billion	$9.6 billion
30%	$1.8 billion	$3.6 billion	$5.4 billion	$7.2 billion
20%	$1.2 billion	$2.4 billion	$3.6 billion	4.8 billion
10%	$0.6 billion	$1.2 billion	$1.8  billion	$2.4 billion

Training Dataset of Bard Statistics

(Reference: scoop.market.us)

• Google Bard’s training data comes from a variety of different sources, each making up a certain percentage. The largest part, 50%, comes from conversation data collected from online forums, which helps Bard understand how people talk in real-life discussions.
• A significant portion, 12.5%, is from C-4 data, providing a wide variety of text to train the system.
• Google Bard Statistics stated that another 12.5% is from English Wikipedia, which is a trusted source of information.
• Code-related documents also make up 12.5%, focusing on technical and programming content.
• Finally, 6.3% of the data comes from web documents in both English and other languages, giving Bard a mix of language-specific and multilingual content.

Google Bard User Statistics and Demographics

• Gemini is available in more than 230 countries and territories worldwide. It supports over 40 languages, including major ones like Spanish, Hindi, English, and Japanese, making it accessible to many people.
• Gemini is not available for users under 18 years old, except in countries where the age requirement is 13. Google has also added extra safety features, like a video introduction and a double-check response option, to protect users.
• Looking at the gender of Gemini users, 60.14% are male, and 39.86% are female.
• The platform is especially popular among people aged 25-34, with the second-largest group being those aged 18-24. Younger people mainly use it.

(Source: grabon.in)

Age Group	Users by Percentage
Over 65 years	3.88%
55 to 64 years	6.58%
45 to 54 years	12.30%
35 to 44 years	19.33%
25 to 34 years	33.22%
18 to 24 years	24.69%

• Research from 2023 shows that 40% of Gemini users used the platform to learn more about topics they enjoy.
• Besides that, people also use Gemini for activities like being creative, getting more done, and having fun.

(Source: grabon.in)

Purposes of Using Google Bard	No of users in %
Entertainment	10%
Creativity	30%
Productivity	20%
Research	40%

• When Google Bard was launched, it was free for everyone to use. After it changed to Gemini, it stayed free because it used the Gemini 1.5 Pro model.
• However, if you want to use the Gemini 1.5 Ultra model, you need to sign up for the Gemini Advanced plan, which costs ₹1950 per month.
• With the Advanced plan, you get extra features like connecting to Gmail, Docs, and other Google services.

Outgoing Links From Bard.google.com

(Reference: scoop.market.us)

• google.com has various outgoing links to different categories, showing its connections to many online resources. A large portion, 49.11%, links to Programming and Developer Software, showing the platform’s focus on technical content.
• The Computers, Electronics, and Technology categories make up 31.06% of the outgoing links, emphasizing Bard’s connection to tech-related content.
• Google Bard Statistics stated that Search Engines account for 8.31% of the links, indicating links to external search tools or results.
• Email-related links make up 7.53%, showing interactions with email services.

Google Bard vs. ChatGPT Statistics

• Google Bard and ChatGPT show different levels of user engagement. Google Bard’s average session lasts 3.19 minutes, while ChatGPT’s session is longer, averaging 8.44 minutes.
• Their data cut-off dates are also different. Google Bard’s data stops in 2019, while ChatGPT’s data goes up to 2021.
• When it comes to cost, Google Bard charges between $0.003 and $0.028 per query. ChatGPT, however, has a fixed cost of $0.0036 per query.
• For subscriptions, Google Bard is free to use. ChatGPT offers both a free plan and a paid Plus plan for $20 per month, which gives users more features.
• In terms of training data, ChatGPT is trained on 300 billion words, while Google Bard uses an even larger dataset of 1.56 trillion words. This helps both platforms understand the language better.

(Source: cyberhaven.com)

• ChatGPT uses 176 billion parameters, whereas Google Bard uses 138 billion parameters. These sectors help improve how well the models understand and process language.
• As for storage, ChatGPT needs 570 GB of space, while Google Bard requires a bit more at 750 GB.

Google Bard Statistics By Visits

• In the summer of 2023, during June, July, and August, two platforms, bard.google.com and bardai.io, had different levels of website traffic.
• google.com, the bigger platform, saw high user engagement. It had 140.6 million visits in June, 189.2 million in July, and 183.5 million in August.
• On the other hand, bardai.io, which is more specialized, had much lower visitor numbers.
• It got 160.7 thousand visits in June, 104.5 thousand in July, and 96.3 thousand in August. These numbers show how traffic changed across the two platforms over the summer.
• The top search term, simply “bard,” had 2.4 million searches, showing a lot of interest. “Google Bard” was the second most searched term, with 2.1 million searches showing people were specifically interested in this AI chatbot.

(Reference: answeriq.com)

• “Bard AI” had 751,000 searches, while “Bard Google” got 483.9 thousand searches, showing curiosity about AI about Google.
• “Google Bard AI” had 127.9 thousand searches, reflecting a more detailed interest. Other related searches added up to 8.5 thousand searches.
• These search results show a clear picture of how much people are interested in Bard and related topics.

Generative AI Adoption Rate at Work in The US 2023

(Reference: statista.com)

• In a 2023 survey of professionals in the US, it was found that 37% of people working in advertising or marketing had used artificial intelligence (AI) for their job tasks.
• On the other hand, the healthcare industry had the lowest usage, with only 15% of workers using AI.
• The high use of AI in marketing and advertising makes sense because these fields often combine creativity and art in their work.
• Google Bard Statistics stated that Generative AI and Education Jobs that need higher education are more likely to be automated by generative AI in the US.
• This is because jobs requiring less education are mostly physical and don’t depend on digital skills.
• However, jobs that require college degrees or higher are still less likely to be replaced by AI, even as generative AI becomes more common.
• ChatGPT’s Competitors Although ChatGPT, created by OpenAI, is the most popular and advanced AI language model, other competitors are getting closer.
• More than half of USUS survey respondents have used or heard of ChatGPT, while nearly half have heard of or used Bing Chat.
• Google’s Bard trails slightly behind, with only about a third of Americans familiar with it.

Engagement of Social Media Platforms

(Reference: scoop.market.us)

• Google Bard Statistics stated that Google Bard gets traffic from several social media platforms, with YouTube being the largest source, providing 52.19% of the total visits.
• Twitter is the second biggest, sending 14.16% of users to Bard.
• Facebook and WhatsApp also bring in a lot of traffic, with 8.82% and 8.58%, respectively.
• LinkedIn contributes 5.61% of the traffic from its professional network.
• Other social media platforms combined make up 10.64% of the traffic, showing the variety of ways users discover and use Bard.

How was Goggle Bard Trained?

• Google Bard uses Google’s internal directory to answer user requests. It was trained with a set of data called the Infinite Dataset Composition.
• Though the name might sound complicated, it simply refers to a large collection of text and code used to train the platform.
• This data comes from various sources, such as conversations, forum messages, and C4-based information. The table below gives a more detailed look at the training data used by Google Bard:

Type of Training Dataset used	The overall share of the dataset in training the platform
Non-English Web Docs	6.3%
English Web Docs	6.3%
Code Documents	12.5%
English Wikipedia	12.5%
C4-based Data	12.5%
Dialogue Data from Forum	50%

Google Bard Tech Statistics

• To compare if Google Gemini is better than ChatGPT, we’ll look at factors like training data, context window, and performance scores.
• Google Gemini Training Data: The earlier version of Google Bard, which used the LaMDA model, was trained on a huge dataset called the Infinite dataset. This dataset has 1.56 trillion words and 137 billion parameters. It takes up 750 GB of storage and is made of:

1. 5% data based on C4.
2. 5% code from programming tutorials, Q&A websites, etc.
3. 5% English web documents.
4. 5% non-English web documents.

• Gemini 1.5 Pro has a very large context window of up to two million tokens, which is the longest available in any large-scale model.
• Both Gemini 1.5 Pro and 1.5 Flash have a standard context window of one million tokens, which helps in handling long documents, code, audio, video, and more. Developers and businesses can apply to get access to the two-million-token window in Gemini 1.5 Pro.
• In the quality index ranking, GPT-4 is the top scorer with 100, while Gemini 1.5 Pro scores 88.
• The quality index measures the overall performance of language models, with higher scores indicating better results.

Model	Quality Index
Llama 3 (8B)	65
GPT-3.5 Turbo	65
Claude 3 Haiku	72
Gemini 1.5 Flash	76
Llama 3 (70B)\	88
Gemini 1.5 Pro	88
Claude 3 Opus	94
GPT-4.0 Turbo	94
GPT-4.0	100

• Training Data: Gemini 1.5 Pro uses a large dataset, similar to GPT-4, but with a different mix of sources.
• Context Window: Gemini 1.5 Pro offers a larger context window (up to two million tokens), which is great for handling complex queries and long documents.
• Benchmark Performance: Although Gemini 1.5 Pro is powerful, GPT-4 leads with a higher quality score of 100, while Gemini has a score of 88.
• In summary, Google Gemini has some strong features, like its ability to handle long contexts, but GPT-4 performs better overall, as shown by its higher quality index score.
• When comparing Google Gemini and GPT-4, GPT-4 has a higher performance score of 100, while Gemini 1.5 Pro scores 88, showing it also performs well.
• In terms of speed, Gemini 1.5 Flash is the fastest, handling 141 tokens per second. Llama 3 (8B) comes next, processing 121 tokens per second, and Claude 3 Haiku processes 101 tokens per second.

Model	Tokens per second
GPT-4 Turbo	23
Claude 3 Opus	29
Llama 3 (70B)	41
GPT-3.5 Turbo	52
Gemini 1.5 Pro	55
GPT-4.0	74
Claude 3 Haiku	101
Llama Flash	121
Gemini 1.5 Flash	141

• GPT-4 runs 75 tokens each second, whereas Gemini 1.6 Pro handles 55 tokens each second, and GPT -3.5 Turbo manages 56 tokens each second.
• Among all Gemini models, the 1.5 Pro version performs the best, followed by the Gemini 1.0 Ultra.
• Google DeepMind tested these models in several areas:

1. General performance
2. Code creation
3. Solving math problems
4. Reasoning (using tests like GPQA and Big-Bench Hard)
5. Multilingual skills
6. Image understanding
7. Math reasoning
8. Audio and video recognition

• For general performance, Gemini 1.5 Pro scored the highest at 85.9%, while Gemini 1.0 Pro scored the lowest at 71.8%. In reasoning tasks (measured by Big-Bench Hard), Gemini 1.5 Pro was also the best, with 89.2%, while Gemini 1.0 Pro scored 75.0%.
• For multilingual skills (tested by WMT23), Gemini 1.5 Pro was the top model with 75.3%, while Gemini 1.0 Pro only got 71.7%.
• In audio recognition (evaluated by FLEURS), Gemini 1.5 Flash scored the highest at 9.8%, while Gemini 1.0 Pro had the lowest at 6.4%.
• For reasoning abilities (tested by GQPA), Gemini 1.5 Pro had the best score at 46.2%, while Gemini 1.0 Pro scored 27.9%.
• When solving math problems, Gemini 1.5 Pro performed the best, with 67.7%, while Gemini 1.0 Pro scored 32.6%.

Key Google Bard Statistics

• As of August 2023, Google Bard is available in more than 230 countries and territories.
• By July 2023, Google Bard will support over 40 languages, including Arabic, Chinese, German, Hindi, and Spanish.
• Between April and May 2023, Google Bard’s traffic grew by 187.2%, reaching 142.6 million visits.
• On February 8, 2023, Alphabet Inc. lost $100 billion after a promotional video of Google Bard on Twitter showed the chatbot giving wrong information about the James Webb Telescope.
• Alphabet Inc. spent 8 years developing the technology behind Google Bard.
• Google’s costs could increase by $6 billion in 2023 if AI chatbots are added to search.
• Half of Google Bard’s training data comes from public forums.
• 25% of Bard’s data comes from English websites, and another 6.25% comes from non-English websites.
• 40% of users say they use Bard to explore their interests.
• 30% use it for creative tasks like writing poems, scripts, and stories.
• Google Bard Statistics stated that almost 20% of people use it for work or school purposes.
• 10% use Bard for entertainment, such as playing games or finding videos and music.
• About 80,000 employees helped test Bard.
• When Bard was launched in March 2023, it had 30 million monthly active users.
• Most Bard users are from the US (62.6%), followed by the UK (8.29%) and China (3.22%).
• Bard’s training data includes 1.56 trillion words and 137 billion parameters.
• Google expects Bard to reach 1 billion users worldwide.
• It is estimated that Alphabet Inc. will face an extra cost of $0.003 to $0.028 per query on top of the base search cost of about $0.003.
• Google Bard was introduced in February 2023 and opened to the public in March 2023.
• In June 2023, the average time spent on Bard was 3.19 minutes per visit.
• Bard had an accuracy score of 63% in March 2023.
• The largest amount of Bard’s traffic comes from direct visits, making up 67.7%.
• Bard can handle about 1,000 queries per hour.
• Google Bard can generate code in more than 20 programming languages.
• Millennials (ages 25–35) make up 35% of Bard’s users.
• Bard’s bounce rate is 55.10%.
• Google CEO Sundar Pichai asked employees to use Bard for 2 to 4 hours each day during its testing phase.
• Google Bard Statistics stated that 750GB of text data was used to train Bard’s language model.

Conclusion

In conclusion, looking at Google Bard Statistics gives us useful information about how well the platform is doing and its impact on the digital world. With a large number of internet users worldwide using Google Bard, it’s clear that the platform is becoming an important part of online search.

The data shows that its user base is growing, with people from different groups and regions, which shows that Bard is popular in many languages and areas. While the platform has advanced features like understanding natural language and offering personalized search results, it also has some issues, such as occasional errors in answers and privacy concerns.

With continuous improvements and updates based on user feedback, Google Bard is likely to stay a top choice for people who want an easy and helpful search experience. For businesses, understanding Google Bard’s statistics is important to connect with its large user base and take advantage of new trends in online search habits.