4 Things to Consider Before Investing in AI for Marketing

Artificial intelligence (AI), while not a new concept, has gained popularity in the marketing industry throughout the past few years. In fact, just by visiting Google Trends and searching for “artificial intelligence marketing,” you can see that within the past 5 years there has been a steep increase in searches surrounding this phrase.

This increasing interest in AI is being driven by industry leaders, bloggers, conferences, and other sources in the know telling us to prepare for the dramatic changes that artificial intelligence will make to the industry.

While we at Element Three believe the technology is going to change marketing significantly within the next few years, there are still some things to consider as your company investigates AI and the different software platforms offering it as part of their product.

Here are four considerations before diving into the artificial intelligence pond today.

1) Is your data ready for AI?

Are you capturing the right data?

When companies begin planning how they’re going to use AI to solve marketing and business problems, they need to make sure that they’re also gathering the right data to effectively train the models.

For example, let’s look at an ecommerce store looking to improve repeat sales from past purchasers by recommending other products via an email after having visited and purchased from their website. This is something that you’ll see large companies like Amazon doing. If a machine learning recommendation engine isn’t receiving very specific data (product category, product item, product price) per user, predicting any sort of outcome would almost be impossible since the engine wouldn’t know which items to recommend to users.

If transactional data or user data isn’t currently being captured, then your company may be a while away from being able to make artificial intelligence beneficial towards closing the marketing loop or providing the other business opportunities that AI provides.

Do you have clean data?

Having clean data can make or break the accuracy of machine learning outcomes, especially if you’re dealing with lower-volume data. The outcome from machine learning will only ever be as good as the data you provide. If you are not 100% confident in the accuracy of the data you have collected, then you may need to consider developing a plan to get clean data.

Key takeaway: Having clean and accurate data is going to be key to making AI an effective avenue of business value. If your data isn’t there, that should be your first focus.

2) Are the out-of-the-box solutions your business is investigating utilizing the technology to its full potential?

Element Three recently researched different companies that offer website personalization as a key strategic software feature. Many of these companies touted that their products were backed by AI and machine learning to help automate the tasks associated with personalizing a website.

After some deep vetting and dozens of discussions with various companies’ leads in this space, we were able to get one top company to admit that none of the out-of-the-box solutions truly had the types of automated capabilities that we were looking for and still relied heavily on manual user input in the backend of their software to create any sort of useful experience.

This is to be expected. Since artificial intelligence is such a hot topic now, it’s natural for these software companies to want to mention anything that will help sell their product, even if their product may be using AI in a way that’s not beneficial to the users yet, or they may not even be using AI at all.

Does this mean that these software companies won’t have a more automated solution in the future? No. In fact, large companies like Amazon and even Apple have started to open-source their software and publish their findings to help the community advance.

You can expect to see some of the top marketing software companies begin to adopt and use technologies and findings from companies like Google, Amazon, Apple, and Facebook in the near future, but be wary of how they are being applied. For the time being, it’s “buyer beware” on actual AI integrations and applications.

Key takeaway: Expect companies to try and sell you software touting artificial intelligence or machine learning. Investigate how their AI or machine learning benefits are actually applied to the software and what type of impact those benefits have.

Reports from Forrester—as well as the background research they do to back those reports—are a great resource to discover top companies within the space. You can also partner with an agency that has prior experience working with software that utilizes this type of technology to help guide you to the right solution based on what you are trying to achieve.

3) Will your marketing stack easily integrate with solutions?

Since data is such an important component to AI, you have to ask yourself: can you integrate your marketing stack with the tools offering artificial intelligence?

During Element Three’s recent investigation into different AI platforms, we found that integrations with popular marketing technologies were hit or miss. Some platforms only integrated with their own suite of marketing technologies, while others worked with only a handful of integrations that varied in popularity. With our already extensive investment in data connectors alongside native platforms, the lack of connectivity for AI tools was unexpected, and shows the relative infancy of AI-based platforms.

Fortunately, many of the platforms offered integrations via their APIs, but be prepared to have to plan the integration, the additional time it takes for a developer or software engineer to connect the systems, and also the upkeep of the connection between the different platforms. This can be supported by dedicated data connectors, but even then it should be considered a technical integration, one that likely requires both a digital marketing expert and a web developer to implement.

Key takeaway: Have a solid understanding of your current marketing technology stack and how the artificial intelligence platforms will have to connect to it. Check to see if the AI vendors offer direct integrations or have documentation that you may be able to pass to your developer or partner agency.

4) Are there other more immediate opportunities available to provide ROI?

Artificial intelligence is the shiny new toy that many companies want, but considering the fact that it can be a major investment, first consider whether there are any other immediate opportunities that have the ability to provide ROI that your company can capitalize on quickly. They may not be as fancy, but they’ll be more cost effective until new AI technology becomes mainstream.

A few cost-efficient and potentially high-ROI opportunities:

Website CRO testing

CRO testing can generate significant impacts when focused on key business aspects. Some of our clients have seen incredible results with it. Check out this case study if you’re curious to see what it can accomplish.

Lead nurturing with marketing automation

Keeping your prospects engaged through marketing automation is a great way to take an automated approach to reaching your users 24/7. Test and revamp what already exists, if you have lead nurturing implemented already.

Paid and programmatic advertising

We have seen tremendous success with paid advertising across Facebook and Google AdWords. Ad opportunities on connected TV and radio can also be worth an investment, particularly if you have the automation and scoring in place to nurture awareness-level contacts generated by such advertising.

Key takeaway: If you’re not capitalizing on some successful and fundamental tactics in marketing, you may see greater immediate success by implementing something like lead nurturing with marketing automation than going towards the big dog named AI.

Only Fools Rush In

Yes, artificial intelligence is likely to change everything—not just marketing—in the near future. And in some cases, it might be the right move for your business. But no matter where you stand, make sure you’re not just rushing into AI because it’s new and interesting. Make sure it can help you today, and provide ROI that makes it worth the effort. AI is coming, and you definitely need to be ready. But you can’t rush it. Do the research, do the legwork. You won’t regret it.