We are marketing in an unpredictable environment. One moment, the internet appears to be on the brink of a giant leap forward. The next, China seems intent on stifling it. There are more scientists alive now than ever lived before, right back to the earliest recesses of time. We have intelligence streaming in.

There is far too much to cope with so we have to filter it. Watch this Tom Cruise movie trailer. Then answer the question, how does he filter iNBOUND iNTELLIGENCE to stay ahead of the competition?

Tom Cruise keeps ahead of the game by understanding the attackers’ personas, and predicting what they will do next. While we may not be able to rewrite the movie script, we can use predictive marketing tools to foresee a lead’s next step through the funnel.

So what is this math marketing all about then? How should we go about capitalizing on the opportunities smart marketing presents? How do you implement predictive analytics and big data?

Predictive Analysis: Offspring of Big Data and Business Intelligence

PC Mag was spot on and I confess to borrowing the ideaWe have masses of big-data-sized business information pouring in, and we have artificial intelligence on the cloud to understand what it is saying to us. These Smart inbound Optimization tools help us do predictive marketing forecasts. But wait a moment: we are not in the world of absolutes yet.

Customers get emotional when reaching purchasing decisions. By this, I mean they let their feelings guide them and these are harder to gauge than logic. Forbes comes from a slightly different angle in terms of customer relationships. It asks ‘will artificial intelligence replace human empathy in the customer service industry’? It answers no, because we do not know how to reduce understanding to bits and bytes.


Until we do – and this could take a while – iNBOUND iNTELLIGENCE processing will still stop short of absolutes, and provide probabilities instead. That said, math marketing predictive tools are becoming easier for you and me to understand. We don’t have to follow the complex algorithms either, provided we get the assessment from a trusted source.

Identifying, Capturing, Managing, & Analyzing Big Data is the Key

AI is the springboard for future business growth. Bright sparks we may be, but the volume of information is too large for us to do predictive analysis on a spreadsheet. We would waste hours and hours doing a job a supercomputer on a cloud could do in a minute. ‘To thine own self be true’ as Polonius says in Shakespeare’s play Hamlet. I would add, leave the rest to the experts and if that’s a computer so let it be.

Artificial Intelligence Marketing may well be the springboard for future business growth. However, we can’t begin to think about smart inbound optimization until we have a reliable database. Moreover, our big data comes to us through a variety of channels. Consequently, it arrives in an unorganized format.  Our data should be relevant to our business initiative. Before we can start extracting nuggets we must therefore know:

  • What data is meaningful, what are the rules for rejecting what is not?
  • Is the information complete, static or dynamic?
  • What will we do with it?
  • Which intelligence will help us achieve our business goals?

We can move on to implementing predictive marketing in our business after answering those questions. 


Predictive Marketing Is a Simple, Well-Established Process

Prediction is a key business activity although we may not have always thought about it that way. In 1893, a fellow by the name of F B Hawley challenged Karl Marx’s theory that profit came at the cost of labor. Hawley believed society compensated business for taking risks. We reduce these risks by taking well measured decisions based on what we expect will happen, and math marketing is the vehicle for doing this.

IT uses metrics from customer data to single out the most vibrant leads, so we can focus efforts on them. This again is nothing new. Marketing and salespeople have always ranked leads in their minds in terms of the most likely prospects. The difference now is lead scoring methodology is more precise.

Before attempting predictive analysis:

  • Understand the outcome(s) where you want to forecast
  • Ensure your team knows which information to provide
  • Agree gray areas and margins of error with which to live
  • Assess predictive outcomes against historic events
  • Accept that you must continuously update data forever
  • Apply predictive data analysis, review results, and learn

Practical Steps for Implementing Predictive Analysis with Big Data

Be in Control Right from the Start

Are you ready to dive in and give this a go? Artificial intelligence is not complicated. There is just a lot of it, and it piles up all the time. With the right software and advice, you could get results sooner than you expect. Big Data is complex and multi-faceted. We cannot check all the details, and hardware and software are not that bright (yet). You will sink or swim based on the quality of your data. Best start with what you already have and then build out.

Know Your Business Data Goals

They have targets in archery so they can tell who aims best in a competition. You will never know where you are going (let alone that you got there) unless you set a direction. What is your reason for doing predictive marketing? What metrics do you want to harvest? Use predictive scoring if you have a list of business data goals, and cannot decide how to trim the number down. Rank the likely outcomes on a score of one to ten. Choose the top half, or the best three maximum. The rest can wait for later. Business is for the long haul if we want to succeed in the long term.


Be Efficient from the Beginning

You may not have a great deal of marketing data when you start, but you sure are going to end up with a fully stocked warehouse somewhere. Appoint a small panel to review progress, but do share the information with those who need to know. Establish inbound and outbound channels with login protocols. Subdivide the big data into meaningful streams to obtain best predictive analysis results. You could do this by relevance to data models that predict future outcomes. You could also use automated segmentation, whereby you group leads by position in the funnel.

Continuously Launder Your Data

Data gets dirty. Products upgrade, and customers come and go. The best artificial intelligence software still can’t do its job properly with stale information. Garbage in, garbage out, here we go again! Make a habit of chucking old socks and doing regular updates. Your system will run faster and you will have cleaner results.

Smart inbound Optimization of big data from your environment is not complicated or difficult. There is software on the cloud we could recommend if you liked. The key to predictive analysis is knowing where you are going, and what you hope to achieve.