When are Bounce Commerce recommendations shown?

That depends on the solution in use. Bounce Intent detects the moment a visitor is about to leave and shows matching product and content recommendations right then - directly on your page, in your shop's design. The visitor does not have to leave the page and is not redirected.

404 Rescue kicks in as soon as someone lands on an error page. On-Page Recommendations display recommendations permanently on product, category and content pages. You can find more information in our overview of the different solutions.

Is the visitor not annoyed? After all, they want to leave the page?

A good question, one we hear often. The answer is almost always no, for three reasons:

1. We only become active when an actual intention to leave is detected - not on every page and not at every opportunity. Anyone browsing your shop happily will never see us.

2. The visitor does not see a generic discount code or a newsletter popup, but products that match the content of the page they are currently on. As long as they find what they were looking for, they are satisfied. The recommendations do have to fit, of course.

3. All triggers are strictly limited per session and only fire after a configurable delay. Nobody is addressed repeatedly, and browser navigation remains untouched - the back button works without restriction at all times.

Can I control where and for whom Bounce Commerce is shown?

Yes. Using wildcard URL matches, we define on which pages and in which areas of your shop the recommendations appear - and where they explicitly do not. This way you can target individual categories or exclude pages such as terms and conditions, legal notice and privacy policy entirely. If a traffic channel is reflected in the URL, for example through utm parameters, you can include or exclude it the same way.

You can also control by device type: desktop, tablet and mobile can each be enabled or excluded individually - for example if another solution is already in use on desktop.

The disadvantage of restricting things heavily is that potential is lost this way, which means there is a risk that users will leave the online shop elsewhere and then possibly buy from a competitor.

Is this not just an exit intent popup?

No. Exit intent solutions fire at almost every opportunity, often aggressively and usually on every page - and they show every visitor the same content, typically a discount code or a newsletter popup. Bounce Intent only reacts when an actual intention to leave is detected, and plays out AI-curated recommendations that match the current page and your product feed.

There is another very important difference: most exit intent providers only work on desktop, because they rely exclusively on the mouseleave event - an event that does not exist on touchscreens. That means more than half of all possible engagements are lost. We detect the intention to leave on desktop and mobile alike, through a combination of several disengagement signals.

Then there is billing: other exit intent tools are sold as SaaS with a monthly flat fee - regardless of whether they generate revenue. We bill purely on a performance basis through your existing affiliate network.

Is Bounce Commerce GDPR compliant?

Our AI works entirely without personal or traceable data. Our solution therefore stands in clear contrast to conventional product recommendation systems and AIs, which base their recommendations on individual preferences. These systems try to identify personal preferences and then suggest matching products.

We take a different approach. Our product recommendations are based on the relationships between different products and are optimized by analyzing completely anonymous data such as click rates. We do not track who clicks, only what is clicked. This allows us to deliver results comparable to those of other systems that use personal data - entirely without personal data.

How does the AI at Bounce Commerce work?

To determine similar products using artificial intelligence, we vectorize all relevant content with the help of a large language model. This assigns content to several thousand topics simultaneously and relates them to one another. An evening dress and a summer dress are thematically closer to each other than a dress and a gas grill.

Our system is extremely flexible. We support several established large language models (LLMs) such as Llama2, Mistral, Gemma, Notux and others. These models have different strengths, for example in certain topics or languages, so we can select the optimal model for each customer.

Building on this, we can use any field from a customer's product data feed. Results based on product names can differ significantly from those based on product descriptions.

This way we find the best results quickly and easily, without lengthy training.

Finally, all AI suggestions can be modified further. Additional rules can exclude or include products in the result. The AI can also learn from the performance achieved and automatically optimize the result with regard to various KPIs.

How much does Bounce Marketing achieve? How many users can I address again? How successful is the approach?

According to a DPD study, an average of 62% of customers abandon the purchase process. On average, we can win back 30% of bouncing visitors for the shop, and 10% of customers also buy again. The success is of course very individual. We are happy to offer a trial month so you can measure and test the effectiveness yourself. In addition, we have received information from individual customers that in some cases more than 50% of referred customers are new customers, which is a high value in affiliate marketing. When using voucher codes in the recommendations, our experience is that a discount code can increase conversions by another 30%.

How is Bounce Commerce set up? How is this maintained?

After activating tracking via the master tag/container tag of the affiliate network or directly integrating the JavaScript code, no further change is required on your site.

We handle the ongoing configuration for you: which pages are included, which are excluded, which products are recommended and which rules apply. We implement change requests for you - you do not have to touch anything in your shop.

Do Bounce Commerce cookies/scripts negatively affect shop loading times?

No. The script is loaded "async" by default, meaning asynchronously. This means it has NO negative impact on page load time. Accordingly, the rendering (display) of the website is just as fast. In addition: the script is very small. Only 946 bytes are transferred.

Execution is also very fast. Usually, execution takes around 400 ms. The AI variant has exactly the same technical specifications. There is therefore no difference in loading time.

How often does a visitor see the recommendations?

All triggers are strictly limited per session and only fire after a configurable delay. Once a visitor has seen the recommendations, they will not be addressed again within the same session - because nobody wants to be pestered.

You can configure the exact limits and delays to suit your needs.

Where are the recommendations displayed? Do I need an external page for this?

No. The recommendations appear directly on your page and blend pixel-perfectly into your shop's existing design. There is no external domain, no external hosting and no change of page - the visitor stays exactly where they are.

One exception is 404 Rescue: there you can additionally choose to forward the visitor from a dead URL straight to the matching page in your shop, instead of showing recommendations on the error page. That destination is in your shop as well.

Are the sales not merely cannibalization effects?

At a large fashion online shop, our solution was shown to only 50% of users in an A/B test. The result was that revenue and sales dropped by exactly 50%. This suggests that users who were not shown the recommendations did not return either. The test did not only analyze affiliate figures, but cross-channel figures as well. It also showed that the sales did not shift to other channels, but could be directly attributed to Bounce Commerce.

Do you have any further questions? Just send us a message

By email to: kontakt@bounce-commerce.de