CUX (cux.io) is the world’s first UX Automation tool. It helps companies not only save the time that so far had to spend on independent and manual data analysis, but also shows how to maintain sales growth by predicting which customer behavior may inhibit conversion.

The tool identifies areas of user frustration that may cause conversion drops, filters heatmaps and recordings, and identifies behavior patterns. CUX eliminates the problem of independent search for problems on websites – it does it automatically, additionally selecting only those visits that are relevant to the business goal. UX Automation enables the analysis of user behavior on any type of device – from phones to smart TVs. In addition, CUX independently tracks all events on the website, thus not involving the customer’s IT department.

CUX free version
As part of the cux.io free trial period, you will have access to 1000 recordings of visits, which you can collect at any time convenient for you. Trial includes all the functionalities of the tool available in a paid subscription.

Paid version CUX
The CUX price starts at 49 euros per month. The final cost is influenced by: the amount of traffic on the website that we want to analyze and the data storage time (retention). When purchasing an annual subscription, 2 months are free.

CUX – how to analyze user behavior?

To start using the tool, it is enough to attach the CUX tracking code to the page you want to follow. This can be done using the integration with Google Tag Manager or Google Analytics. After starting the analytics, CUX automatically catches all the events on the pages, thus not involving you in setting technical parameters.

The most important of the CUX functionalities:

– Experience Metrics – proprietary metrics showing the frustration of users on the website. They are divided into: rage clicks, rage key presses, chaotic mouse movements, zooming and constant content refreshing.
– Recordings of visits – CUX analyzes entire visits, instead of individual sessions. This means that it tracks the user’s visit even when he opens many tabs or is inactive for a long time.
– Heatmaps – CUX generates heat maps for single pages and SPA (single page apps). The tool is also compatible with various types of devices – from phones and computers, to tablets and consoles, to smartTV and the so-called wearables.
– Conversion Waterfalls – CUX helps to track visitor behavior in various scenarios. In other words, Waterfall is a graphical representation of any sequence of steps visitors take on your site.
– Analysis in the context of a business goal – regardless of what your business goal is (sales, training registration, filling out the contact form, downloading an e-book, etc.), CUX will select for you only the data about user behavior that they have on it real impact. The number of goals for a project is unlimited.
– Automatic event catching – CUX is able to automatically capture events on the website, allowing code-free tracking of each stage of the user’s journey without the need to involve the IT department.
– Retroactive analysis – CUX allows you to create and modify conversion funnels even using historical data and without having to wait for new recordings to be collected.

Three levels of optimization

Working with qualitative data boils down to optimizing the real user path – not the one we dreamed of when creating the product! To make it as effective as possible, we divide it into 3 levels: marketing, product and conversion optimization.

In order to properly optimize marketing activities (directing traffic from external sources), we need to know exactly what is happening to our users along their entire path. Where they land, what they see, how they react. So we optimize based on observation.

How do we go about optimizing the customer journey in the context of the product? We shorten it! Why? Customers – those online in particular – do not always have a strong purchasing incentive. It is not uncommon for them to be driven to the side of your product or service by their impulse. The longer the path to conversion – the more time for the user to change their mind!

Conversion optimization does not only consist of the last steps of the sales funnel (cart, checkout). For our improvements to make sense, we need to take a few steps back and look at the entire customer journey – from marketing communication to product interactions.

Analysis in the context of the goal

The advantage of CUX for us with other behavioral analysis tools is that it filters all results in the context of your business goal. Regardless of what will be your main goal (Sales? Filling the form? Registration for training?), The data read through the prism of its implementation are much more valuable.

Imagine your business goal is for customers to fill out a contact form – this is the main place where you sell. Setting the goal “Send contact form” will allow you to automatically filter out those visits that show the achieved goal. Thanks to them, you will be able to improve the contact process, spot problems that your users are struggling with or change the page layout so that they convert more and faster.

Event analysis

To make the data you analyze meaningful, accurate and relevant to your business goals, create extensive conversion funnels based on events. What? Completely any! These can be clicks on specific elements on your website, filling out forms, entries to specific websites or coming from specific links, e.g. campaign links.

How do you go about an event-based analysis?

– Start with the whole – check the “global” paths of your users, compare, see which are doing better.
– When you notice a big drop, take a look at this place in detail. Watch the visits, create a funnel for this particular fragment on the page.
– Remember that the sequence of steps you set in Waterfall is important, and when counting conversions, we take into account the execution of individual steps in the order set by you.

Heatmap analysis

As you probably know, the heatmap graphically shows the most important activities on your pages. Every action – like clicking or mouse movement, scrolling, chaotic movement – adds color to the area. To draw reliable conclusions from the heatmap analysis, try to take into account those for which min. 250 events. On pages that contain various dynamic elements (such as pop-ups, drop-down menus, hoover), remember that they may not be captured on the heatmap. However, do not forget to consider them from an analytical point of view – perhaps the rage clicks that “fly” on the heatmap relate to these elements.

Analysis of recordings of visits

A truth that entrepreneurs often forget: analytics is not just watching random recordings! When you start working on recordings, keep a business goal in mind and filter only those visits that meet it. There is no point in wasting precious time on meaningless recreations.

Always start your analysis from general to specific and look for patterns of behavior. Watch the visits until the picture of your users’ habits begins to form a complete picture in your head. So how long? Usually, a dozen / several dozen visits are enough for the pattern to start clearing up. You can also use Experience Metrics in CUX to start with the behavioral patterns we have identified (rage clicks, zooming, refreshing the page, etc.).

FAQ

How is CUX different from Hotjar?
CUX not only combines all the functionalities of analytical tools (such as Hotjar), but also automates the data analysis process, which makes it an innovation in the user behavior analytics market. Unlike Hotjar, CUX is able to present a complete picture of data – both quantitative and qualitative – and interpret it in the context of a business goal.

Will CUX be useful for industries other than e-commerce?
Of course! The analysis of customer behavior will benefit not only e-commerce, but also businesses in the SaaS and B2B model, companies providing marketing services, software houses – all industries in which the customer comes into contact with a digital product. In addition, CUX is able to facilitate the work of various departments and teams within the company – from marketing, through sales, IT, content creation, to product development and management.

When is it worth analyzing user behavior?
We often come across the erroneous assumption that behavior analysis comes in handy when we see that the website is not working properly. Yes, this is a good time. But not the only one. Studying the behavior and introducing minor implementations on their basis at every stage of the digital product’s life helps to almost immediately increase the level of conversion or prevent its decline.

What’s the CUX price?
The cost of a paid subscription starts at 49 EUR / month and depends on the amount of traffic on the site and the customer’s required data retention time. In addition, CUX provides additionally paid technical and analytical onboarding, personalized workshops, webinars and bootcamps as well as ready-made website audits.

Why is product feed optimization so important?

At the beginning, it is worth asking yourself the basic question why the care of the product feed is so high in the hierarchy of activities that we undertake when creating a PLA campaign. It is such an important factor for several reasons. First of all, in this type of campaign, we have no influence on what key phrases our ad will be displayed on. It is the Google Ads algorithm that decides when to show our product, based on the information contained in the product feed. Therefore, optimizing this file, for example by adding key phrases of interest to us in the title or description of the item, gives us at least an indirect influence on to whom and in what context our products will be published.

Secondly, a number of attributes can be added to each product, i.e. product features, such as, among others, size, color, material, gender (e.g. in the case of clothes) or pattern. In the Google Shopping tab, the user can filter the results based on exactly the same attributes. So the more fields describing the product are filled in, the greater the chance that it will also appear in filtered views. The third reason to optimize the feed is that Google Ads algorithms constantly determine its quality and reward products from those files that they consider to be of high quality in views.

High feed quality, in addition to influencing your impression position, can also help reduce CPC, the rate per click on your ad. If the feed is well structured, our product is more suited to the customer’s inquiry, thanks to which we burn less budget. This means that the campaign can achieve much better results, as it gives more clicks within the set budget and reaches a better selected audience.

What must be included in the product feed?

In the case of the product feed, some information is mandatory for the feed to be accepted by the system, and some is optional, and in addition, some options are required only in campaigns for selected countries or only in the case of specific product categories. Therefore, when creating a product campaign, it is best to check the feed requirements in our industry and in the country where we want to publish the advertisements. In simplified terms, however, it can be said that the information that must be included in the product feed includes:

– ID – unique identifier of the product. Ideally, it should be the same as in our warehouse identification.
– Title – the name of the product and the most important information about it. We will write a little further about what exactly it should look like.
– Description – in this field, describe the product in detail, using the keywords that are most relevant to you.
– Link – this should lead to the product page where you can make the purchase.
– Image link – address of the main product image, which will be available for indexing by Google.
– Availability – product status. It can have the value “in stock”, “unavailable”, “temporarily unavailable” or “presale” (in the case of the last two options, it is also obligatory to complete the attribute indicating the date of availability).
– Price – it must be consistent with the price on the store’s website and on the payment page.
– GTIN – Global Trade Item Number – GTIN is required only for products that have this manufacturer assigned number, but Google strongly recommends it be used for all products.

If the product feed includes the above-mentioned elements, it will be accepted by the advertising system, but it is not tantamount to high quality of the feed. These are only the basics and to get a high rating for the algorithm, you should also complete some of the attributes from the optional list. Before we get down to describing them, however, let’s figure out how to optimize this basic product data.

How to optimize your product feed titles?

The title is a very important element not only because it is one of the first pieces of information the user sees. It is also important because it is one of the main parameters that the algorithm takes into account when determining which keywords to display a given product on. Therefore, it is worth including not only information about the name of the product, but also, for example, its brand, color or size. We will repeat this information in other attributes, but if we also include it in the title, we increase the chance that the products will be displayed with the best queries in their case. The order in which the information is provided is also very important, which may be slightly different for different product categories. Google states that the best practices for title optimization for the most popular industries are:

– Clothes: Brand name + Gender + Product type + Product features (color, size, material, etc.),
– Food: Brand name + Product type + Product features (taste, number of pieces in a package, weight),
– Home & Garden: Brand Name + Product Type + Product Features + Model,
– Electronics: Brand Name + Product Features + Product Type + Model,
– Seasonal products: Opportunity + Product Type + Product Features.

Of course, these are not universal models that will work in all cases. For example, a brand name in a title will only attract an audience if it is recognizable and has a good reputation. So it’s best to experiment with names and test different options to choose the combination that works best for you.

It is also worth remembering that product titles may be displayed differently on different devices and thus the system may cut them off in some cases. We have 150 characters to use in the title field and it is good to use them, but make sure you have the most important information at the beginning of this section. In the Google Shopping tab, the titles are truncated after 70 characters, so it’s best to fit the most important data in this pool.

How to optimize the descriptions in the product feed?

In the case of the product description in the feed, we have 5,000 characters at our disposal. Of course, we don’t have to use all of them, but this field is a place for us to place many keywords. Admittedly, this will not guarantee that the product will show up for these words, but it does significantly increase the chances of it happening.

The description should not only be saturated with keywords, but also valuable for the user. Therefore, it is worth ensuring that it describes the product in detail and is of high quality. The description for each product should be unique, which increases the chance that the system will correctly select the one that best suits a given search query from the list of our products. For this field it is best to use a minimum of 500 characters, although of course the more useful information for the user, the better.

However, the question of where to get the information from which keywords will be most valuable for a given product remains. This is where the free Google Trends tool comes in handy. In it, we can enter terms that are associated with our product and check which of them are most frequently searched by users of the search engine. Thanks to this, we can include in the description those key phrases that have a chance to bring us the most impressions and clicks on the ad. For example, if Google Trends shows that “women’s black high heels” than “women’s black pumps” are more frequently searched for, it is worth using the first option in both the description and the title.

How to optimize product photos in a product feed?

An obligatory element of the feed is a link to a product photo, which the system can then use in advertisements. However, in order to obtain the best possible file quality, it is worth not to stop at one photo, but add a few to show the product more precisely. This can be done by using the Link to additional photo attribute. This attribute can be repeated 10 times for one product, which means we can add up to 10 additional photos. However, the standard is that for one product, we put 3-5 photos that show the product from different angles.

All photos should be of good quality, but their weight must not exceed 16 MB. Although in the official specification of the feed on the Google website we can find the information that the photos should have a minimum of 250 × 250 pixels for clothing and 100 × 100 pixels for other products, it is better to focus on photos with a higher resolution. The optimal size for product photos is 800 × 800 pixels. It’s a good idea to stick to these guidelines, because high-quality images improve your click-through rate and, as a result, can also increase your conversions.

Optional attributes that should be completed to optimize the product feed

As already mentioned, in addition to the basic data, it is also worth supplementing at least some of the attributes that are considered optional by the system. It is true that the file without them will be accepted and the ads will be displayed, but their effectiveness will not be as good as if we added additional information. Here are some attributes that should definitely be supplemented.

Brand
The Brand Attribute is most helpful when we sell products of well-known and frequently searched brands, but it is worth using in any case. However, this is an attribute that is often misspelled and can make your ads less effective, so pay close attention to it. A common mistake is to enter the name of your online store here, while the correct value is the name of the manufacturer of the product. So if, for example, we sell Nike sneakers in our store called XYZ, we should enter “Nike” in the brand attribute, not “XYZ”. Enter the name of your online store here only when our company produces the item from the offer. If the product does not have a brand or the brand is unknown, do not complete this attribute. It is a mistake to enter a value of the type “not applicable” or “missing” here.

Product group ID
This attribute is used to group products that come in different versions, such as sizes and colors. Thanks to this, they will not be displayed to users as separate products, but as variants of the same item, which will make it easier to get to know the offer and make a purchase decision. For all products to be grouped, we should add the same product group ID.

Size
This attribute is mandatory for clothes and shoes, but recommended for all products with different size variants. It is best to use the most popular sizes in the country where the campaign is to be used to determine this parameter. If a product is available in multiple sizes, it is important that you add the Product Group ID to all versions of the product. If the garment we sell has one size, we can enter the value “universal size” in the attribute.

Material
Material is one of the most important attributes for many products, including clothing, accessories, footwear and furniture. If the product has different material variants, each of them is added to the file separately and combined into one offer using the same product group identifier. It happens, however, that an item is made of several materials, e.g. a mixture of cotton and polyester. Then all the materials are added in one version in the Material attribute, separating them with the “/” sign without spaces, eg “cotton / polyester”.

Mobile link
If we have different URLs for traffic from computers and mobile devices, we can also complete the Mobile link attribute. Thanks to this, we will make sure that users using smartphones and tablets, after clicking the ad, will go to the subpages of our store optimized for these devices. This, in turn, will give them a better shopping experience for our brand. If the page URL is the same for both the desktop and mobile versions, we don’t need to complete this attribute.

What else can be done to optimize the product feed?

The high level of the product feed is also influenced by including information about product types in it and filling in the field with Google categories. Thanks to the proper classification of the product, the system is better at positioning the offer and adjusting it to the needs of the recipients.

It is also worth ensuring that only up-to-date information about our product range appears in the product feed. For this purpose, it is best to set up automatic product feed updates in Google Merchant Center once a day. You can also fix more frequent updates (up to 4 times a day), but such frequent refreshing of the product feed is only necessary in individual cases. In addition, if we make changes to the website, e.g. we add new products to the offer or introduce a promotion, we can also force the feed update manually, without waiting for the automatic download of data. When setting automatic updates, it is a good idea to choose night hours so that refreshing the file does not interfere with the display of ads to recipients.