Predicted Eye-tracking Tool Comparison: 3M VAS (Visual Attention Software) vs. Attention Insight

Anyone in the marketing world knows that competition for attention dramatically increases every day and how hard to get noticed by the right audience at the right time. Today it‘s not enough to have a good product or service for success. From the business side, it’s extremely hard to get attention, but on the other hand, it pushes to develop innovations and technologies.

Logically, if people are paying attention to your competitor, they can’t pay attention to you at the same time. For this reason, marketers look for ideas on how to be noticed all the time and it‘s not hard to find a lot of different bits of advice, popular at different times, like using intriguing infographics, eye-catching creatives, or headlines, blinking, moving content and others.

It‘s glad, today we have many modern and effective solutions based on artificial intelligence, machine learning, and even deep learning which is a type of machine learning and typically needs less ongoing human intervention, but characterizes high accuracy. These methods do not require a long time and big expenses of experimenting but can help to win the battle and keep the leadership in the “attention game”.

One of the most powerful tools in the competition of attention is predicted eye-tracking. It is a very good sample of how science and visual art intersect to deliver the business superior results. In the market, we can find at least a few companies creating and improving such kinds of tools based on AI. Here will be analyzed and compared two predicted eye-tracking tools – Attention Insight vs. 3M VAS (Visual Attention Software).

This comparing study will be covered according to:

  • accuracy
  • features
  • integrations
  • price and free trial possibilities
  • support, users experience, and reviews

Can we trust predicted eye-tracking accuracy?

It‘s the main question when we are talking about the predicted tools based on AI.

Logically, if we don’t know Artificial Intelligence accuracy vs. Real eye-tracking, we can’t be sure whether data is valuable for our business or not.

When we are talking about accuracy it‘s important to notice that 3M VAS started research earlier and used machine learning. Attention Insight uses newer technology evolution of machine learning – deep learning. A deep learning model is designed to continually analyze data with a logical structure similar to how a human would draw conclusions. With a deep learning model, an algorithm can determine whether or not a prediction is accurate through its own neural network – no human help is required.

Below will be analyzed heatmaps real eye-tracking, Attention Insight, and 3M VAS for a deeper understanding of the accuracy.

First of all, need to compare percentage accuracy and information on how was collected data.

AI Accuracy vs. Eye Tracking Attention Insight 3M VAS
Accuracy 90 - 94,5% 87 - 92%
Model evaluated by MIT Saliency Benchmark Yes Yes
Algorithm has been trained on ~30,800 images from real eye tracking studies. Plus Additional 24,000 images data sets The model powering VAS was developed by 3M scientists using hundreds of thousands of data points from academic eye-tracking studies.

Like we see in the table, the Attention Insight deep learning algorithm has been trained on a very concrete number of images and evaluated by MIT Saliency Benchmark (It is data set with held-out human eye movements and is used as a benchmark test set), proving the accuracy of 90-94,5 %.

3M Company declares that VAS stems from more than 30 years of research on the human response to visual stimuli. This legacy has been combined with tech expertise in machine learning and analytics. The model is 92% accurate versus an eye-tracking study capturing the critical first glance.

They provide very abstract numbers information that getting results was used hundreds of thousands of data points from academic eye-tracking studies. Data used for this validation consisted of four different data sets, two of which were collected by academic research labs (MIT & York University) and the third and fourth sets were collected by 3M. A variety of images were used, which included indoor scenes, outdoor scenes, and advertising content. The validation study contains more information about 3M VAS researches.

For more clarity below will be compared visualizations of real eye-tracking studies with Attention Insight and 3M VAS. Real eye-tracking data uses as a benchmark.

At first – webpage analyses:

Real
eye-tracking

Attention Insight
Predicted eye-tracking

3M VAS
Predicted eye-tracking

The real eye-tracking study shows us the main attention on the bigger font in the middle of the picture and a little bit on the middle of the face. Attention Insight demonstrates the main focus on the face and after that on the information field together with the e-mail form. A little bit of information gets the logo. 3M VAS shows pretty similar results – the main attention is on the bigger font, face, e-mail form, and plus logo on the top of the page.

In the 3M VAS heatmaps, we see intensive blue color for more clarity needs to explain that in the VAS models the color means the range of probability.

 

Red means a higher probability;

Yellow mid – range probability;

Blue – lower probability, but still attracting attention;

According to this explanation, the attention diapason is wider and more possibilities for interpretations.

Let’s do another comparison – package:

Real
eye-tracking

Attention Insight
Predicted eye-tracking

3M VAS
Predicted eye-tracking

Real eye-tracking and Attention Insight demonstrate very similar views. The main attention focus on the information in the middle of the package. Another attention area is the information on the corner, but not so intensive as the information in the middle.

In this comparison, 3M VAS gives us surprise and shows that the strongest attention area is the design decoration on the right-hand side, and after that, the main information is in the middle of the package. According to the VAS model, almost all areas of the package attract attention, so need to evaluate the probability of where attention is strongest. Like in the first sample heatmap is more expansive.

The third of our comparison field is a poster:

Real
eye-tracking

Attention Insight
Predicted eye-tracking

3M VAS
Predicted eye-tracking

Like two samples before, Attention Insight predicted eye-tracking heatmap is very similar to real eye-tracking data. The focus is on the same areas and is shown to us the same places attracting attention. According to the intensity of colors, the area getting the most attention is the header. The face of the woman, slogan, and logo get attention as well.

3M VAS heatmap shows that the main attention is focused on the woman’s face and gets down on the chest. The heading, information field, and slogan demonstrate the lower probability of attention. It‘s a different result than real eye-tracking data and Attention Insight heatmap. Additionally, almost all areas of the poster show us attention probability. It leaves us more interpretation than Attention Insight, where areas are very concrete, clear, and very similar to real eye-tracking.

Comparing Attention Insight and 3M VAS heatmaps on webpage, poster, and package sample we notice that higher accuracy compared with real eye-tracking studies is in Attention Insight side. While both using methods- deep learning and machine learning fall under the broad category of artificial intelligence, deep learning is what powers the most human-like AI. According to analyzed samples obviously, Attention Insight accuracy and eye-tracking concentration are very similar to real eye-tracking benchmarks.

Luckily, the heatmaps are not only one metric for comparing results. Attention Insight and 3M VAS provide us with more additional and very useful features.

3M VAS provides us with information about AOI ( Areas of Interest). You get a percentage of AOI on all marked areas.

3M VAS AOI (Areas of Interest)

Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

3M VAS Hotspots

Each region shown has a numeric score which is the probability that a person will look somewhere in that region during first-glance vision. The content that is most likely to be seen during first-glance vision (3-5 seconds) and during that period has a higher probability of grabbing your audience’s attention.

Besides, VAS tests pre-attentive vision and identifies the five raw elements that people are naturally drawn to when they first gaze at something (3-5 seconds) while they are not affected by gender, age, culture, or other influence. The five raw elements are:

  • Edges
  • Intensity
  • Red-Green contrast
  • Blue-Yellow contrast
  • Faces


The visual elements results break down each area of interest score based on the five attributes that the human brain is naturally noticed – edges, color contrast (red-green and blue-yellow), and faces. The higher the number, the more influence the visual attribute contributes to attracting the viewer‘s attention.

Below is a sample of how visual elements are divided and what gets the highest attention.

Attention Insight counts the percentage of attention too and can do it in a more convenient way – AUTOMATICALLY. Another advantage in the Attention Insight camp is – the clarity score which allows you to see how clear a design is and compare it with benchmarks

Attention Insight clarity score
The focus map shows the parts of the design noticed in the 3-5 seconds. According to this information, it‘s possible to identify the most important elements in the analyzing design. This Attention Insight functionality is similar to 3M VAS hotspots functionality and helps to reach the same goals.
Attention Insight Focus map

Features of predicted eye-tracking

Every business has different purposes and needs according to that are making decisions what type of tools and technologies to use. Both companies offer a wide range of feature sets.

The similarities – attention heatmaps, areas of interest, reports, comparison, plugin access, app, and mobile use.

3M VAS lacks some things like AUTO AOIs (areas of interests), clarity score, URL capture, API integration, but on the other hand, gives very detailed information about five visual elements, hotspots, and gaze sequence. The gaze sequence is the order, what four areas a person will notice the first. It can be useful if the predicted viewer attention aligns with your intended visual hierarchy or goals.

3M VAS gaze sequence

Attention Insight misses gaze sequence and the main reason is the doubt that is a correlation between the main attention-getting places and gaze sequence.

On the other hand, Attention Insight offers other features for businesses that could be more beneficial and provide useful insights. One of the most useful features nowadays is VIDEO heatmaps and the company intensively is improving this sphere. If it’s necessary to predict eye-tracking information not only for statics view but for video, the Attention Insight team will prepare the heatmap for video.

The easiest way to choose the predicted eye-tracking tool is to compare all features in the table and take the most fittable for the business profile.

Features Attention Insight 3M VAS
Attention Heatmaps Yes Yes
Areas of Interests Yes Yes
Auto AOIs Yes -
Percentage of Attention Yes Yes
Intensity Factor - Yes
Comparison Yes Yes
Clarity Score Yes -
Clarity range recommendation (Alexa Score) Yes -
Focus Map Yes Yes
URL Capture Yes -
Current View analysis Yes Yes
Full Page analysis Yes -
Work in teams Yes Yes (but just after requirement and not less than 20 people)
Reports Yes Yes
Recommendation Yes Yes
Downloadable Study Report Yes Yes
Shareable study URL Yes -
Allowed filetypes jpeg, png, webp jpeg, png, webp, mp4
Attention heatmaps for video Yes (Video heatmaps feature is available upon request. The test is performed by Attention Insight team) Yes
Hotspot - Yes
API Yes -

Possibilities of integrations

Both companies work intensively creating convenient integrations for clients’ needs.

Through the web or plug-in for Photoshop, Illustrator, or XD, VAS analyzes your designs and provides feedback as to what is attracting attention, and what might distract from them.

Attention Insight offers a Google Chrome extension that will allow you to generate heatmaps on your website by clicking on the extension in your browser. The API can enable software creators to add heatmap functionality to their tools. The wide range of plugins like Adobe Photoshop, Adobe XD, Figma, Sketch as well.

Intergrations and API Attention Insight 3M VAS
Google Chrome extension Yes Yes
Adobe XD plugin Yes Yes
Adobe Photoshop Yes Yes
Adobe Illustrator - Yes
Sketch plugin Yes -
Figma plugin Yes -
API Yes -

Price plans and free trials

Very often PRICE and the plan options are the main points having an influence on the decision.

Attention Insight has three options to fulfill different needs. The solo plan is dedicated to one person and it’s a very good choice with a friendly price for freelancers working in design, web, UI/UX, marketing spheres. Team plan perfect for small or middle size companies or agencies. For companies having special needs, creators provide unique solutions. Trifle, but very convenient for users from different countries – one push and the price you can see in euros or dollars.

Differently than 3M VAS, Attention Insight offers just monthly plans and doesn‘t ‘t offer annual plans with discounts.

3M VAS has a totally different pricing strategy. They offer four packages.

One plan is just for Web app access, another just for plugin access, the third is a full plan containing both access – web, and plugin. All of these plans are just for one-person use.

Team subscriptions are available just after requirement and not less than 20 people. Determined to use 3M VAS predicted eye-tracking tool not less than a year get the option to buy an annual plan with discount.

While PRICE is one of the most important criteria before the decision, FREE TRIALS are a crucial point. Naturally, to try the tool before a decision is mandatory.

Luckily both companies provide the possibility to try tools for free. Using free trials is a fantastic way to test features and compare accuracy.

Attention Insight and 3M VAS let you use a free trial without payment information, so you can decide to subscribe to a paid plan or let your trial expire and don’t continue. Both companies have a simple login and account-creating processes.

Attention Insight offer is a free 7-day trial and enjoys access to all Team plan features.

3M VAS free trial will allow you to analyze three images before purchasing a subscription.

Always the best option is to try and make sure which features and user convenience are the best fit for business purposes.

Support, user experience and ratings

One of the most important things starting to use new technology is to get support on time and find the information easily.

Attention Insight provides users with a step-by-step guide for maximum using clarity. It’s possible to look at a demo study or get help on a chatbot. If you need more information they have a very wide knowledge base, webinars, video guides, tutorials, and even live chats with experts.

If you want to know more about 3M VAS, you can try to find answers on the FAQ page or to fill out a contact form and wait for an answer. If you want to dive deeper you can try to find videos yourself on YouTube. Unfortunately, 3M VAS doesn’t make your life easier and doesn’t provide useful links or chatbot conversation.

Free trials of these predicted eye-tracking tools were a pleasant user experience, it was possible to find everything intuitively, explanations and reports were detailed.

Always is helpful before decision-making to check reviews or ratings. Attention Insight has a lot of high ratings on different pages. According to 2022 January searching data, google was found 154 reviews and ratings. The score was not less than 4,5 from 5 stars.

About 3M VAS was possible to find just one review in google search, but the rating score also was the best – 5 stars.

In conclusion

All kinds of businesses moving to the digital world and taking up space. Competition is fierce. The market is saturated with an enormous quantity of products, brands, services, and everybody requires attention.

Luckily, today we have attention prediction, and would be silly don‘t use advanced AI technology in the attention race. The main challenge is how to choose the appropriate and the most valuable tool for business.

All predicted eye-tracking tools use AI technology, but different concept. 3M VAS uses machine learning, Attention Insight a newer subclass of machine learning – deep learning. Deep learning came onto the scene a few years ago with the growing amount of data in today’s digital world. Deep learning is modeled after the human brain and represents a higher level of artificial intelligence.

The poster, webpage, and package analysis using these two predicted eye-tracking tools and comparing them with real eye-tracking data show us that Attention Insight conformity of heatmaps is bigger. The results are more clear and concentrated. The accuracy is validated by the concrete numbers of images on which the deep learning algorithm was trained.

3M VAS uses the machine learning concept and declares a very similar percentage (92%) of accuracy but gives very abstract numbers for validation. Analyzing heatmaps and comparing them with real eye-tracking the results are different in two of three analyses.

Otherwise, both companies try to provide the best package of features – heatmaps, areas of interest, percentage of attention, possibilities to compare different designs, storage, reports, and plugins integrations for the maximum user convenience.

Like everywhere are some differences too.

Attention Insight has Auto counting percentage of attention, clarity score, and try to provide the possibility to work in teams easy and comfortable – URL Capture, shareable study URL, API integrations, plan for the team. Video creators can get the video heatmaps.

3M VAS, on the other hand, provides very detailed information about five visual elements and the gaze sequence.

Companies have different pricing strategies and different types of offers but are the same orientated for customers’ convenience and offer a wide range of integrating plugins.

Both predicted eye-tracking tools offer easy and for free to try them and discover the best option for business needs by yourself. Just important to remember that comparing prices and features do forget to compare accuracy too. Mostly, the accuracy becomes sacrificed for a better price or more features but in the predicted eye-tracking technology more than others, the accuracy is a crucial point determining success in reaching your business goals.

 

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.