You’ve got to have been living in a cave if you’ve not heard in recent years about the expected impact of AI (Artificial Intelligence) on our existence on this planet.
The impact will no doubt be far reaching, across a wide range of areas, industries and disciplines. Not least would be the impact on marketing. In fact, we’re already experiencing the impact of AI in marketing whether we know about it or not.
Right now there are a wide range of platforms in use out there that are using AI to improve results. As an agency, we actively use a number of these tools to help us get better results for our clients. A good example is ActiveCampaign, which uses Machine Learning to determine when is the best time to send a campaign to particular users or who is more likely to convert within the sales process.
We use Co-schedule which uses AI to help schedule social posts to get the best engagement. This is based on the times of day and days of the week that users engage plus the types of content they engage with.
These tools are able to get better results than we can achieve as humans.
Another tool that uses AI is, of course, Facebook. In fact, most of the big marketing platforms use it already.
As marketers these days, we are all using AI-based marketing whether we know about it or not. If you’re using Facebook Ads then you’re using an AI-based marketing platform.
Facebook uses AI in a number of areas such as:
- Textual Analysis – Facebook uses DeepText to extract meaning from words we post by learning to analyse them contextually. Neural networks analyse the relationship between words to understand how their meaning changes depending on other words around them.
- Facial Recognition – Facebook uses DeepFace to teach it to recognise people in photos. It says that its most advanced image recognition tool is more successful than humans in recognising whether two different images are of the same person or not – with DeepFace scoring a 97% success rate compared to humans with 96%.
- Targeted Advertising – Facebook uses deep neural networks to decide which adverts to show to which users. This has always been the cornerstone of its business. However, by using machines to find out as much as they can about us – and to cluster us together in the most insightful ways when serving us ads – it hopes to maintain a competitive edge against other high-tech competitors.
- Designing AI apps – incredibly, Facebook even uses AI to decide which processes can be improved by AI. It uses a system called Flow to run tests of over 300,000 machine learning models every month to find further opportunities.
What this all means for us is that we are well on our way in the adoption of AI in marketing already. We’re already using it.
If you put to one side the negative impact that this might cause, such as the loss of jobs within the industry. What it does do is create environments for improved results.
If you take learning cycles involved in marketing as an example, we’ve moved very quickly from the print-based direct mail marketing approaches of years ago – where learning cycles could be in the region of three to six months between campaigns – to digital marketing campaigns where you can review and change ads every week.
The learning cycles have reduced from months to days. It’s incredible!
With AI this can now be reduced from days to minutes or even seconds. Finding out what works against what doesn’t can be automated and your campaigns optimised automatically.
These are exciting developments in marketing. Right now we are at the onset of this technology. Machine learning and AI can help achieve better results in your marketing based on the objectives that you programme in. Therefore, it still involves humans to set up the campaigns and specify the objectives correctly to benefit from this technology.
Following the set up that we’ve outlined here will position you well to take that advantage, but don’t just set it and forget it. We still need to manually monitor and measure campaigns at this stage. Especially when first setting up a new funnel. Once our optimisation job has been carried out and the funnel is working we can then step away and let the AI start to work it’s magic. Then just dip in and out periodically to check on performance.
