Dragonfly AI vs Attention Insight: Predicted Eye-Tracking Comparison Study

In the last decade, we started to hear an interesting and intriguing term – artificial intelligence very often. This term is everywhere, we hear it all the time in our surroundings and it’s not surprising to us anymore. Artificial intelligence became a natural part of life. A lot of tools, technologies, and software that we are using every day are based on AI. 

The progress wasn’t felt so strongly like the last couple of decades. While we tried to understand artificial intelligence’s pros and cons and discuss ethical aspects the science jumped to a new stage – machine learning and even furthermore – deep learning. In simple words, deep learning is a subset of machine learning, and machine learning is a subset of Artificial Intelligence.

dragonfly
Source: https://www.readspeaker.ai (Deep Learning Vs. Machine Learning Vs. AI)

The same situation happened with marketing. All of us know something about marketing in general, later we became familiar with digital marketing, today we are further talking about Consumer Neuroscience and Neuromarketing.

Why here are we talking about AI and neuroscience? The reason is simple – predicted eye-tracking tools which are based on AI and close related to neurosciences. More insights about neuroscience are in the article “What are the possibilities of neuromarketing now and in the future?”

If you are not familiar with predictive eye-tracking terms, the short explanation in simple words: it‘s an AI-based alternative to real eye-tracking studies. Heatmaps, generated by AI, just like eye-tracking heatmaps, show which areas of the visual materials people are likely to view the most and the least.

It’s hard to imagine where we will be after a few years or decades. Honestly, even today it’s hard to be on time with the newest technologies and tools, and choosing appropriately according to your needs becomes a strong headache because a choice is very wide. So, if we want to find the best options the most appropriate for our business model, tasks, or goals, it’s mandatory to do homework and to compare a lot of components like features, prices, accuracy, support, user experience, and many more.

We can find a lot of companies creating predictive eye-tracking tools with different features, but now our attention surrounds the company Attention Insight, using the newest and most advanced AI subset – deep learning, and a company with a very intriguing name Dragonfly, using artificial intelligence and data of neurosciences.

 Let‘s meet. On the scene – Attention Insight vs. Dragonfly AI

Naturally, the first question is, why dragonfly, and how it’s connected with the eye-tracking tool? Dragonflies see faster than we do, they see around 200 images per second, they can see in 360 degrees, and nearly 80 percent of the insect’s brain is dedicated to its sight. Now, when we got acquainted with dragonfly characteristics we can see the nice parallel and it’s clear why a company creating a real-time artificial intelligence tool that instantly predicts what an audience sees first chose the dragonfly name.

On the other hand, Attention Insight chose a very accurate name exactly characterizing their activity. 

If we started to talk about accuracy, it’s important not only choosing names but first of all talking about data accuracy.

Data Accuracy in Predictive Eye-tracking

Predictive eye – tracking is a very good tool helping win a greater share of attention in a crowded market, but first of all, it‘s important to find out about accuracy. Obviously, without data accuracy, all our research and efforts won‘t be useful and valuable. 

Attention Insight uses deep learning – a new generation subset of machine learning based on artificial neural networks which work like human brains and can solve complex tasks. The Attention Insight deep learning algorithm has been trained on 30,800 images from real eye-tracking studies, plus additional 24,000 image data sets, and evaluated by the MIT Saliency Benchmark proving the accuracy of 90-94,5 %.

AI Accuracy vs. Eye Tracking Attention Insight Dragonfly AI
Accuracy 90 - 94,5% 89%

Dragonfly AI is based on an algorithm that is not subject to the bias of machine learning data sets. It can analyze digital content across any channel, digital and physical, and it’s one of the first tools with real-time and full-motion capabilities. For added reassurance, Dragonfly AI has been validated by the industry-standard MIT Saliency Benchmark on its ability to predict the results of eye-tracking experiments to an accuracy of 89%.

The technology is based on biological brain function and employs a patented algorithm to determine what the human brain sees first. The innovative predictive visual analytics tool uses Artificial Intelligence to metrically analyze user attention and response during the first few, critical seconds they engage with your physical and digital content, based on over ten years of research at the Queen Mary University of London.  

Dragonfly AI analyzes anything it sees, so anything that can be imported into a camera roll or provided via a URL. Based on understanding the neural architectures in the visual cortex, Dragonfly AI replicates how the eye and brain process differences in light and shape and assigns each pixel with a numeric saliency value, turning the data into a virtual ‘heatmap’. 

Both companies’ heatmaps provide a powerful insight into what happens when someone looks at your website, content, video, shelves, or package. Using and understanding heatmaps can give you a clear picture of how your content is performing, allowing you to make data-driven decisions to increase interaction and return on investment. 

The best approach to compare and see how accurate our analyzing tools are is to compare genuine eye – tracking heatmaps with heatmaps from our analyzing tools but we have a minor issue here. Dragonfly AI makes it difficult to test their product and obtain a sample heatmap. There is no free trial available. Even if you directly request a sample heatmap, you will receive a negative response.

For this reason here will be compared just Attention Insight vs. Real eye – tracking heatmaps, and later the  Attention Insight vs. Dragonfly AI heatmaps which is possible to find on their social media.

First, is compared a poster – Real eye – tracking heatmap and Attention Insight predictive eye –tracking heatmap in the table below.

Attention Insight’s heatmap is very similar to real eye-tracking data. The emphasis is on the same areas, and we are shown the same places that are attracting our attention. The area receiving the most attention, based on color intensity, is the header. The woman’s face, slogan, and logo all draw attention too. Colors intensity is different, but it is visualization‘s nuances.

REAL
eye-tracking

dragonfly

Attention Insight

predictive eye-tracking

dragonfly

Real eye-tracking heatmap vs. Attention Insight predictive eye-tracking heatmap

Another comparing study – a package.

Real eye-tracking and Attention Insight produce startlingly similar results. The most attention is paid to the information in the package’s center. The information in the corner attracts attention as well, but not as much as the stuff in the middle.

REAL
eye-tracking

dragonfly

Attention Insight

predictive eye-tracking

dragonfly

Real eye-tracking heatmap vs. Attention Insight predictive eye-tracking heatmap

Compared these two samples – a package and a poster. We can see very similar results between Real eye-tracking heatmaps and Attention Insight predictive eye-tracking heatmaps, and it’s no surprise because according to  MIT Saliency Benchmark  Attention Insight the accuracy is 90-94,5 %.

The next step is to compare Dragonfly AI heatmaps with Attention Insight

In the first sample, we see a press conference visualization with a logo wall and bottles on the table. Interesting moment – which logo will win the attention game?  According to the Dragonfly AI heatmap „Just Eat“,  „Volkswagen“ and „Qatar Airways“ strongly capture attention with their logos. Also, Coca-Cola bottles get attention too, but the heatmap is not concentrated, so we notice different areas, just one brighter than another. It‘s a lot of places for interpretations.

dragonfly
A press conference visualization heatmap by Dragonfly AI

On the other hand, we have an Attention Insight heatmap and notice totally different results. The main attention is on a man’s face and then on the logos in the left corner and just after that a little bit on the right side. Coca-Cola doesn‘t get attention at all. 

This heatmaps comparison study confusing because the results are totally different.

dragonfly
          A press conference visualization heatmap by Attention Insight

Validate your concepts for performance during the design stage with AI-generated attention analytics

For more clarity just try another heatmap comparison study. Let’s try a banner.

dragonfly
Attention Insight heatmap

Attention Insight heatmap is very concentrated and clear. It‘s two the main areas which are the most noticeable. Extraneous triggers don’t confuse and clarity is maximum.

Dragonfly AI heatmap

Dragonfly AI heatmap shows us the same noticeable areas and plus more areas – the food plate on the bottom and some areas on the right side. The color visualization and blue color intensity distract the attention and it‘s harder to understand the information on the heatmap. 

When we put together Attention Insight and Dragonfly AI heatmaps we note differences and certainly clear which heatmap is concreted and more understandable. 

As we see, the results are different and it’s a pity we can‘t get the same heatmap that we could compare all three heatmaps: Attention Insight, Real eye–tracking, and Dragonfly AI.  It‘s hard to give objective final conclusions while no full data from the Dragonfly AI side. But according to these comparisons, we notice that Attention Insight heatmaps, are very similar to the Real eye–tracking heatmaps, but Attention Insight and Dragonfly AI heatmaps have differences. While Dragonfly AI doesn‘t agree to provide a heatmap sample that would be possible to compare with a real eye tracking heatmap, the accuracy scale unambiguously is on the Attention Insight side. Additionally, according to MIT Saliency Benchmark Attention Insight accuracy percentage is higher too.

Free trial, support, and user experience

Personally, I am testing and trying a lot of different tools, programs, and technologies very often and one of the most important things it’s to get this possibility quickly and easily. I like to try the free trial, check the features and prices and it’s clear, want to continue and use the program or tool or not. In my opinion, it’s a very important user experience part. No one of us doesn’t wants to waste time trying to look for information, registering, asking for demo calls, waiting, and communicating. Simplicity is the key to success. 

Trying to check the Dragonfly AI tool and compare it with another predictive eye–tracking tools I was confronted with problems. 

First of all, this company doesn’t provide the possibility to try the free trial. You must register for the demo call and wait for confirmation with an appointed date.

If you are not acceptable to have a demo call designated time you have to arrange other dates.  It’s really big time-consuming in the first phase when your goal is just to check the tool and get information about using convenience, features, and compare accuracy.

Even if you ask managers to prepare and send a sample of the heatmap for comparison with real eye tracking it’s not possible. 

As we talked about before, accuracy is one of the most important aspects of choosing predictive eye-tracking tools, data comparison it’s mandatory before making a decision. From that depend on your business success. 

Just in a demo call, a manager can show you what looks like a tool and how to use it. Any heatmap for studying and comparing is not possible to get. 

Besides, you don’t have any idea about prices. It’s not possible to imagine how much the cost tool is, while you don’t get a demo call. After talking and explanation about your needs selectable a plan and just after that count price.  

Actually, it’s a little bit strange, when in the world everybody wants speed and simplicity it’s so complicated to get prices and try tools. Obviously, my like user experience wasn’t satisfactory. It’s too much work and time to get information and prices. 

On the other hand, while I was trying to persuade and to get a sample heatmap for comparison with real eye–tracking studies, answers and communication by e-mails were kind and I got some links where I can read more information about the Dragonfly AI tool. Unfortunately, I like one time to try than to see or to read. 

Furthermore, if you are not an English speaker it will be very hard to get information about the tool and the price because after the demo call just you will be able to know it. The company’s focus is just on English speakers.

Luckily, Attention Insight has a free trial, and no problems trying this tool. You can go HERE and log in. A 7-day free trial with full access to all Team plan features is available. You have the choice of upgrading to a paying plan or letting your trial period expire when it has elapsed.

It’s feasible to compare reports that you can acquire while testing free trials. The report includes a focus map, clarity score, attention heatmap, and percentage of attention. Technology, accuracy, and explanations are presented. Everything has been offered on the basis of aesthetics. You can obtain a report in pdf format or a PowerPoint presentation that you can edit and add additional explanations or comments for colleagues or clients.

dragonfly
Attention Insight report sample

Every user’s registration and login to Attention Insight is straightforward and quick. The free trial procedure and navigation are simple. A chatbot and rapid responses are accessible for assistance. In the free trial, Attention Insight has developed a step-by-step tutorial so you don’t have to hunt for instructions. You will be guided through each stage. You can immediately view a demo study or seek advice from a chatbot. The firm created a comprehensive knowledge base as well as solutions to frequently asked topics. If you want more information here is the article “How to use Attention Insight App” for a clear and easy knowledge of how to use the Attention Insight platform, as well as explanations. If that isn’t enough, you can request a free demo with an expert, but first of all, you have the possibility to try everything yourself.

Both companies have blogs where is possible to find a lot of useful information and chat on the web page where is possible to ask questions directly.

Prices and plans

First and foremost, Attention Insight caters to a variety of clients and offers a variety of solutions based on your needs. You can choose a SOLO plan for a very low fee if you are a freelancer, one person, or an extremely tiny business. You can choose a Team plan if your team isn’t larger than five people. If you have special needs, the best approach to get them met is to contact the Attention Insight team, who will provide you with tailored solutions.

Solo and Team plan prices are in the table below.

dragonfly
Attention Insight plans

Dragonfly AI offers plans according to specific kinds of content. The company offers four plans.

  • Studio plan fits a wide range of digital assets like marketing material, ad creatives, or packing designs.
  • App plan simulates real-world consumer journeys and instantly evaluates how assets, such as product packaging and visual merchandising, as well as signage and point-of-sale incentives, are performing. It‘s possible to see live attention heatmaps and measure effectiveness in the real world, in real – time.
  • Extension plan maximizes performance in line with your commercial and customer experience objectives, simulates online customer journeys, discovers friction areas, and optimizes your digital touchpoints. It‘s possible to analyze websites in real time and test for multiple devices.  
  • Video plan it‘s possible to analyze a wide range of video assets for usage in social and broadcast media, from live user session records to short and long-form video clips.

Even if you know which plan is the best for your needs, it‘s unknown the price. It’s not possible to find any price ranges, while you haven‘t arranged a demo call with Dragonfly AI. Just after the demo call, you will get the price. So if you want to know the plan price you must be patient. If the price range is not fittable for you it will be wasted time. Most of you will agree with me, that one of the first steps is checking the price range before diving deeper into analyzing tools or technologies which you plan to use. If the price is not acceptable to you, no point to to go deep into analyzation.

The features of the plans are in the table below.  

dragonfly
Dragonfly AI plans

Features and integrations

Features Attention Insight Dragonfly AI
Attention Heatmaps Yes Yes
Areas of Interests Yes Yes
Auto AOIs Yes (CTA buttons) -
Percentage of Attention Yes Yes
Comparison Yes Yes
Clarity score Yes Yes
Live view - Yes
Image upload (drag and drop) Yes Yes
Focus map Yes Yes
URL Capture Yes Yes
Current View analysis Yes Yes
Full Page analysis Yes Yes
Plugin Access Yes Yes
Work in teams Yes Yes
Reports Yes Yes
App and mobile Yes Yes
Report Builder Yes Yes
Shareable Study URL Yes Yes
Attention heatmaps for video Yes (Video heatmaps feature is available upon request. The test is performed by Attention Insight team.) Yes
API Yes Yes
Gaze plot - Yes
Session recording - Yes

Both companies provide heatmaps and Areas of Interest with the percentage of attention for deeper heatmaps understanding. It helps to make better decisions and get higher performance. The visualization of the percentage of attention is pretty similar.

dragonfly

Dragonfly AI percentages of attention

Dragonfly

Attention Insight percentages of attention

Attention Insight and Dragonfly AI try to ensure users for useful features and convenient use. It’s possible to work with the team and easy to share, to build the reports, easy image uploads using drag and drop interface, and compare results using A/B testing, hotspot analysis, and integrations. 

Dragonfly AI tool is available on desktop, as an App for iOS, as an extension for Google Chrome, and as an API / SDK for enterprise integrations, but not possible in the Adobe package. 

Attention Insight offers Chrome Extension, Figma, Sketch and Adobe XD, Photoshop, API. No doubt it’s very convenient for design creators when they can use tools directly in the creating process. 

Dragonfly AI also has a gaze plot, which allows you to observe the sequences. 

dragonfly
Dragonfly AI gaze plot

In the picture is shown what looks like a gaze sequence, what will be seen in the first place, what in the second, the third, and so on.

Attention Insight misses the gaze plot consciously, they doubt that there is a correlation between the main attention-getting places and the gaze sequence.  But if we compare the same picture using the Attention Insight heatmap and the percentages of AOI, we notice that the attention places coincide. 

The first reading place – 54,5 %

The second – 12,2 %

The third – 3,4 %

And the last- 3,1 %

Dragonfly
Attention Insight heatmap with percentages of AOI

Dragonfly AI offers a live view (analyzing the environment live via the camera feed), motion capability, and session recordings.

No doubt, that both companies provide all features for users’ convenience, just Dragonfly AI works more in live view and motion sphere. Attention Insight tries to offer more integrations and focus on heatmaps accuracy. Upset that it’s not possible to try Dragonfly AI features yourself and compare everything step by step.

Feedbacks and ratings

Testimonials are really helpful since we need additional approvals, we want to know what to expect,  to hear opinions from real people, and we want to compare the advantages and disadvantages. It’s a great power and makes a huge impact on the decision-making process.

On many search pages, Attention Insight has a lot of positive feedback and ratings. More than a hundred testimonies can be found on the independent web pages in the fast view, and the score is 4.5 – 5 stars.  Good testimonials, achieved outcomes, case studies, and partnerships are all available on the website Attention insight, Trustpilot, Capterra, g2, and other pages. Good testimonials validate the usefulness of the predicted eye-tracking tool, especially after checking the free trial and trying all functions by yourself.

On the web page of Dragonfly AI, we can find positive testimonials, a list of clients, well-known brands, partners, and memberships but to find reviews on the other pages on the search are not possible. No comments about the tool in the review pages like Trustpilot, Capterra, g2, or others. When is not possible to find any testimonials on the independent pages and when is not possible to try the tool by yourself, the confidence decreases. Testimonials could be a good motivation to try the predictive eye–tracking tool, especially when free trial and plan prices are not provided.

Final Thoughts 

If your content hasn’t grabbed the user’s attention within a few seconds, that’s it, they’re gone. You’ve lost the chance to ensure a good outcome by converting a user into a customer. The market for attention has never been more competitive and consumers are becoming increasingly time-poor.

The purpose of this article was to compare two useful predicted eye–tracking tools based on AI and neuroscience insights that could help be more competetive in attention games. 

These tools allow you to identify actionable insights before your content is published and are designed specifically to match the needs of marketers, and creators, to create both brand differentiation and performance acceleration. On the other hand, it’s possible to eradicate any costly design issues before distributing them to the public using predictive eye-tracking methods. For a wider understanding of how Artificial Intelligence works in the creative industries, it’s possible to read the articleWhat Artificial Intelligence is and how does it work for creative industries?“ 

Both companies provide heatmaps, it‘s a powerful insight into how consumers interact with your content. They can help you to make informed, data-driven decisions to improve attention and increase conversion rates, whether you’re designing your website, looking to improve your content, package, shelves, or optimizing. 

Based on the analyzed data, it is important to note that both tools – Attention Insight and Dragonfly AI have similar offerings in terms of features, design analyses, areas of interest, and percentage of attention.

However, Attention Insight uses the deep learning method and for this reason, the heatmaps are more concentrated, clear, and understandable.  It’s easier to get insights and make corrections.  

The comparison study revealed the importance of the free trial. It‘s the best way not only to check features and find the most fittable option for business but to compare the accuracy and the clarity of heatmaps. According to samples, we can see that Attention Insight heatmaps visualizations are more accurate and it was confirmed not only by heatmap analysis but the percentage of MIT silence benchmark too.

It’s disappointing that the Dragonfly tool could not be tested and the results weren’t compared to real-world eye-tracking benchmarks. On the other hand, the unknown price of the plans reduced the intention to try this tool. It‘s not clear whether is it an acceptable tool according to financial possibilities. To delve deeper into analysis, substantial motivation is required, such as a large number of extraordinarily good testimonials or very high ratings, but this organization can only brag about testimonials on its own website.

One time to try is the best option to test any technology. Noticeable, Attention Insight pays attention to user experience and provides a simple way to test its predicted eye – tracking tool, focusing on ease of use, a wide range of integrations, and accuracy. 

 

About Author

Exclusive Insights On your Users Attention

Complete Guide to DATA-DRIVEN DESIGN

Get ebook for free. Just leave your email below.

You will be subscribed to our mailing list.