Good bot, bad bot, chatbot: what does it mean?

It’s the hot topic of the moment. For the past few months, chatbots have been all the rage. Closing the gap between customer relationship and conversational marketing, they are turning into a tremendous asset for innovative brands. Yet, they are not the first form of artifical intelligence used by businesses to grow and evolve – and some of the other forms are built with far less benevolent purposes in mind.

The word “bot” covers a wide variety of automated programs – some designed to help brands, others trying to grab proprietary content to feed databases or impair marketing campaigns. Those bots still have a lot in common – primarily their ability to imitate human behaviour. While chatbots mimic our conversations, bots that are programmed by search engines or designed for fraudulent purposes copy the way we browse the web, making them hard to detect.

The ability to distinguish bots from actual humans, and to protect companies against bad bot activity, is the core of DataDome’s expertise.


  • The word “bot” covers a wide variety of automated programs designed to mimic human behaviour

  • Chatbots, meant to facilitate customer relationships and conversational marketing, are the next big thing in the world of business-facing AI

  • Other bots, such as web scrapers, ad fraudsters or impersonators, can hurt digital brands and online businesses

  • More than 50% of all web traffic comes from bots

3 minutes reading time

In April 2016, Facebook announced the launch of chatbots for its Messenger app. A new tool to help brands automate their customer support, or help online retailers and e-commerce websites create innovative, interactive online experiences. As they came into the spotlight, these bots have become the next big thing in the tech world – and as it usually happens, caused a tidal wave of questions and predictions about the future of these programs and their impact on the the business world.

For example, Burger King announced an experiment using the Messenger chatbot for taking orders. It is a good illustration of how this technology simplifies the management of certain tasks, by automating processes that are often difficult to manage and staff appropriately. Saving time, saving energy: bots of all kinds are there to make our lives easier.

Chatbots, the future of support

The automation of such tasks enable businesses to improve their efficiency, and represents another step towards the acceptance of artificial intelligence in current productivity practices. Bots, in all their diversity, are already delivering added value for online businesses: search engine bots, for example, are created specifically to help businesses appear in relevant search engine results.

Humans 49%
Good bots 26%
Bad bots 25%

Bots account for more than 50% of all web traffic

Warming the general public to these visible and interactive forms of artificial intelligence has certainly been facilitated by the development of virtual assistants such as Siri, Cortana and Google Now. These programs were created to automate actions that imitate human behavior as closely as possible, to such an extent that they may not only be useful for enterprises. They can also help us, as individuals, save time and energy by eliminating tedious tasks — like wishing a friend happy birthday on Facebook.

Automating intelligence

An exaggerated claim? Not according to Irene Chang, a developer who recently unveiled the Chat bot club, a prototype which can take over control of your Facebook Messenger when you’re out of time — or motivation.

Little by little, automated intelligence will begin to complement our daily lives, making its way into our conversations in order to help us save time and focus on what matters.

The question then arises: How do we make the distinction? How will we be able to differentiate between actions and messages generated by humans and those coming from automated programs? A kind of invisible Turing (or Voight-Kampff) test to help us understand who we’re really talking to.

For many enterprises, the ability to distinguish between various forms of automated activity and true human behavior gradually becomes a necessity. For while chatbots and search engine bots can help businesses prosper, all bots were not created to support and benefit everyone, from the developer to the consumer.

$7,2 billion

will be wasted by brands worldwide in 2016, due to ad fraud primarily generated by bots.

The threat of bad bots

While bots have been at the forefront in the trade media lately, the development of automated programs imitating our behavior isn’t new.

At DataDome, we have often discussed it: a significant portion of these bots, usually invisible to the public, is roaming the web in order to analyze its content, copy or capture proprietary data, or tamper with the results of online advertising campaigns.

For an online business, the ability to bring bot traffic into the light and understand its impact can become a competitive advantage: by stopping real-time capture of e-commerce price and product data, for instance, or by blocking access to proprietary media content.

Knowing and understanding bot traffic can also enable businesses to optimize their revenue, for example by protecting their advertising campaigns from click fraud.

Making the difference

Effective, real-time bot detection requires expertise and state-of-the-art technology. The methodology — our next-generation Turing test, so to speak — is based on real-time analysis of very precise technical and behavioral information, which allows us to differentiate the patterns of human web browsing from those of bots.

Contrary to chatbots, bad bots are not created to interact in a transparent way with a user — which doesn’t mean they aren’t cruelly impacting enterprises and their customers.

The mission we have given ourselves at DataDome is to provide online businesses with top-notch protection against bad bots, while allowing the good bots to operate undisturbed.

These forms of artificial intelligence, and the ways in which they are used in everyday operations of enterprises worldwide, are poised to grow exponentially in the months and years to come. In order to create the best possible environment for them, and to ensure appropriate and transparent interaction between bots and human users, it is necessary to set up efficient protection against bad bots.

Long live the bots — as long as they are working for the common good.

Don’t let bad bots harm your online business anymore. Access your personal dashboard for free in order to monitor the quality of your traffic in real time: