As with many other aspects of industry and commerce, Artificial Intelligence, or AI, is having a huge effect on marketing, bringing exciting new tools to bear on the task of promoting goods and services to varied markets. There has been a huge increase in the numbers of companies offering marketing technology. Many claim to use AI, but how many of them actually do so?

To get the real picture, we first need to be clear what AI is. One of the major characteristics of an AI solution is its use of algorithms, which could be formulae or logical decision trees – basically, a fixed set of rules to actually optimise a campaign or to do something else as part of your B2B marketing activities. The key thing about these algorithms is they’re programmed by a human and will typically have very limited ability to develop or learn.

The other side of AI is machine learning or neural networks, which is essentially computers understanding the world, and then trying to apply rules based on that understanding. The key thing about machine learning is you need to train the computer with various data sets. But once you’ve done that, the computer can do things and have insights that a human may not have, making it the most exciting part of AI.

AI can also be narrow or general – narrow AI tends to be focused around trying to achieve a particular outcome, or perform a particular activity, whereas general AI is basically the intelligent computer HAL from the film ‘2001’.

When we look at AI for B2B marketing, we’ll be focusing on narrow AI, so our definition is a computer algorithm that can learn from data to produce insights and recommendations specific to the brand or campaign. To add value, the algorithm must learn and can’t simply be a preprogrammed set of rows.

And if it’s learning from the activities for a particular brand or campaign, it should then be able to produce recommendations that are uniquely beneficial to that brand, or that campaign.

The challenges of machine learning

We need to ask if there has been a huge impact from machine learning on B2B marketing.

The CMO survey for 2019 shows that almost 60% of the respondents say they’re using AI today, although its main uses are in personalization of content, and predictive analytics, both of which tend to be algorithmic type applications. In these, the computer might be delivering different content, to different personas, but it’s based on a set of rules – not machine learning.

This means that the bar is not very high for using AI in marketing, as predictive analytics can be as simple as lead scoring algorithms. So, at the very simple level, everybody can start using AI.

If we look at some simple examples of AI in marketing, the first to consider is content creation. One of the most common uses is email subject line generators. These tend to use a very basic approach where a stock phrase is put in front of a benefit. This is not a bad way to generate ideas for a subject line, but it will be sometime before we get AI writing with the same degree of verve as a human copywriter.

Content personalization is offered by many tools, with most of them based on simple fixed rules. The most important thing here is the insights used to personalize the content, which are generated by humans – a human works out why one persona needs a particular message, and another persona needs a different message or a message phrased in a different way. So, it’s very different from something that’s truly intelligent, making it a great way to get into AI.

However, we’re still some way from computers actually running your marketing campaign and typically, most people are not using machine learning tools.

So, why don’t computers learn? Well, it’s actually quite hard to train a computer, requiring a neural network and a training set. There is also no easy answer to how much data you need, but typically, it’s thousands of data points and you may not have thousands of email subject lines you can compare to find out which will be most effective.

Although computers are pretty slow learners, they’re much less likely to jump to incorrect assumptions than humans. People may assume that the email with the highest percentage open rate must be performing better, whereas, taking into account randomness and long term performance, the AI solution will come to a more considered conclusion. But, still, computers are very slow learners, making it hard to take advantage of machine learning.

AI can be basic but very effective

There are actually fairly few applications of AI that really make a difference in marketing.

In AI for B2B, they really fall into three key areas around marketing itself. One is look-a-likes – if I know this particular profile of customer is a good customer, then find me similar people who might well be good customers.

You can use AI to predict intent, which can be as simple as lead scoring, or it can be a much more complex algorithm that learns who’s likely to buy which product.

Performance prediction is being able to assess whether a particular campaign is likely to perform well. These three areas of AI are all things that you can actually use now that will use machine learning and help you with your campaigns. But as you can see, they’re fairly small elements of the overall campaign and there’s some way to go before AI takes over a whole campaign.

Although things like look-a-like audiences or predicting intent are probably the least exciting forms of AI, they give the most benefits.

Pay per click advertising often uses AI, so if you’re using Google ads, you might be using smart bidding, which will basically determine how much you pay for an ad, using an AI.

Chatbots are growing because a lot of inquiries on a website can be dealt with fairly simple automation, but what is really needed is some intelligence to process the different ways people can ask questions.

By the end of 2021 about 40% of companies will adopt voice technology, creating things like Alexa. This does all the difficult work of understanding what the user is asking, so it’s very easy to generate these chatbots and generate voice skills.

Another application is image recognition. The huge data sets available for images make it relatively easy to train AIs to understand images, not only what’s in the image, but also to understand things like colour, and even the emotions of people in the picture.

One of the biggest is Google Cloud Vision API. This lets you send images to Google, which processes them and sends back information about them. A great example of this would be our client Censhare. They have a digital asset management system and they will automatically add more information, more tagging, and more data about an image using the Google Cloud Vision API.

Designed for marketing

But what you really want to know is, are there actually AI applications that truly optimize marketing activities, above and beyond the more generic ones? Well, yes, there are, including several designed to help you find and target the best accounts.

Once you’ve got an account list, you can use a tool like Bombora, which will identify things that drive sales, such as if they hire new people, if they announced new sales, if they perhaps are a startup, or getting new funding.

There’s also automated account discovery. With Terminus, you provide a list of accounts that you want to target, and the system then identifies similar accounts.

Then there’s digital behaviour analysis. This takes what Bombora is doing and looks at some of the contacts you’re targeting, trying to build a picture of the company and the contacts and assess the right time to approach that particular customer.

Another aspect is personalizing content. Power Factory tries to serve content so people visiting your website get the right content at the right time. You can track things like the time spent engaging on each part of the content and see if it resulted in a sale, so you can get a lot more information than just offering a PDF.

Persado discovers which phrases resonate with your audience, so you can use the phrases that work the best – pure AI powered copywriting. And then there are also products that actually identify content, such as which finds related content for your newsletters.

Still, generating natural language is not easy. I think our copywriters can relax, because it’s unlikely we’re going to see them put out of business – maybe product descriptions could be written with AI in the very near future, but long form copywriting will still be written by humans for the foreseeable future.

There are also a number of AI assistants that aim to follow up contacts you might meet at a trade show, for example – they’ll send emails that appear to be from real people, following up and trying to get someone to respond. You can get into this with a free trial with products like Conversica. The interaction is pretty limited, but it’s not a bad solution if you just want something that’s going to automatically send follow up emails.

The next area is understanding engagement, looking at how and when people engage with the content, and ultimately aiming to provide smart personalization. Again, it’s not necessarily generating the content that people are going to engage with, but it’s certainly going to help you serve the right information at the right time to visitors.

Optimizing the time to send emails as well as the frequency is another thing that AI can help with. It’s really difficult to optimize this and it requires machines to sit there and look at what works and what doesn’t – Seventh Sense is a well-known product that aims to detect and act on engagement with email.

Go your own way

We’ve talked a lot about different products that use AI, but what about creating your own? Actually, you can. If you’ve got custom data and you want to find out why one particular project is more likely to convert than another, you might choose to build your own AI rather than buying something off the shelf.

And you probably only need a relatively small number of lines of code to do that sort of analysis, although you need a data scientist to write them – you will also need a lot of data.

AI is going to help us more, particularly with image and voice recognition and natural language processing for chatbots – if you’re not using those technologies for your digital asset management, or to create chatbots, now’s the time to start thinking about it.

Campaign optimization is an area that really is about to hit primetime – you might not want to deploy it today, as it’s still expensive, but in the next few years, we’ll see a lot more people using it, particularly in terms of dynamically serving content on the website, to drive people through their customer journey.

AIs are also going to be able to give indications of performance, whether you’re likely to win a particular design opportunity, or whether your lead is a high or a low score for example.

Top tips

So finally, how can we take advantage of AI? The first thing is, don’t feel left behind. The actual use of AI is pretty straightforward, so you can get on board very quickly without needing to invest a lot.

I would certainly experiment with simple AI – probably the easiest way to do this is with Google ads. Sometimes we see great results, sometimes not but certainly experiment with AI and try to understand how to get the best out of it.

Keeping up to date is very important, to follow what’s going on and try to understand who the new vendors are in the market. Build your data sets – the more data you can build now, the more you’ll be able to use AI as those machine learning tools come on board.

And your agency should understand AI. It is going to replace everything that they do, but an agency that uses AI is going to be a more efficient agency.

Although few organizations are really using heavyweight marketing AI, there are real applications that can be delivered with very reasonable budgets – get on board now and start understanding how AI can benefit you.


If you want to find out more about AI in Marketing, who not check out our Uncovering the Truth about Artificial Intelligence in Marketing webinar.