AI in Marketing Automation: Dreaming Small Today

When I read about artificial intelligence, I usually see a futuristic picture of amazing machines making terrific breakthroughs, defining strategies, predicting, and prescribing the best course of actions.

The problem is we won’t be able to quickly get to that point simply because… we are swamped, and that is due to very mundane, time-consuming work. It seems #AI proponents are inviting us into a super jet while we can’t get to the airport due to the long rough roads and slow horses. We lose efficiency today because marketing automation systems are still manual. Nothing wrong with that, indeed. But as wishful thinking, before we go into a brave new world of AI-driven marketing, it would be tremendously helpful to have the technology to resolve some housekeeping items:

  1. Data Management – smart and quick data normalization, mapping, and cleansing. If you ever set up all variations of job titles to be mapped to proper job levels, you know what I mean. With all the success around natural language processing, we still have to create a smart campaign to update United States to USA.
  2. Filter and Flow Steps Verification – is you have smart lists and flow steps with multiple triggers, filters, constraints, advanced logic, and scenarios, verifying all that might be very tricky. Basically, you have to look into who is left, spot check, and with trial and error fine tune your filters. The problem is tomorrow there is another use case that you couldn’t foresee, and your filters/triggers/flow steps are missing them. Being able to have some prompting and verification tool might be a huge time saver.
  3. Troubleshooting – before you manually go through all the activity logs, and check the campaign parameters, isolate numerous factors, an AI should be able to look this up and based on existing/historical data indicate possible issues.
  4. Immediate Alerting on potential errors and “suspicious” activity. Is someone trying to send an email with subject line “Update Subject Line”? Reference a wrong asset? Load a list with missing or invalid values? A smart AI should be able to prevent that.
  5. “Marketo Admin Bot.” How much time does an average #Marketo admin spend answering the same basic questions? If AI could store, process and pull the relevant knowledge, an admin would be able to enjoy a cup of coffee finally and look into all the fantastic things an AI can do.

#marketingautomation #AI #artificialintelligence #datascience #marketingoperations #marketingtechnology #martech

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s