Page 5 - AeM_Aug_2018

RESEARCH, ANALYSIS & TRENDS

Predictive analytics simply calculates the consumer Prescriptive Analytics is the final leg of your advanced
preferences: How much do your customers spend on analytics strategy. With this, and with your data and
your offers over time? Which customers are likely to marketing materials, you can finally clearly define what
respond to your offers, depending on the content? At to do to achieve your goals or predict your customers'
what point does a potential customer make his/her first shopping experience. With a “Thank you”- page or a
purchase? And when is the next purchase to be confirmation mail after the purchase, you can promote
expected from existing customers? another offer (next sell recommendation).
Prescriptive analytics: How should we respond (to In fact, prescriptive analytics not only anticipates what
those potential future events)? will happen and when it will happen, but also why it will
happen, and suggests decision options on how to take
Prescriptive analytics entails the application of advantage of a future opportunity or mitigate a future
mathematical and computational sciences, and risk and shows the implication of each decision option.
suggests decision options to take advantage of the
results of descriptive and predictive analytics. That’s Of course, prescriptive analytics can continually take in
why it is referred to as the "final frontier of analytic new data to re-predict and re-prescribe, thus
capabilities. And if you go into the prescriptive phase automatically improving prediction accuracy and
not just with predictions, but with advanced analytics, prescribing better decision options, and ingests hybrid
that's even better. Using prescriptive analytics data - a combination of structured (numbers,
algorithms allows your company to provide categories) and unstructured data (videos, images,
recommendations to customers based on the sounds, texts), as well as business rules to predict what
information in their profile, considering the context of lies ahead and to prescribe how to take advantage of
your industry (current product portfolio, inventory, profit this predicted future without compromising other
margin, etc.).
priorities.
To implement recommendations, you can e.g. suggest
a new product to replace a previously purchased If you do not yet see the value of using a more complex
product or offer a package with multiple products at a approach to performance, look at some success
reduced price. That way you get an additional buy (up- stories, where some companies reportedly have seen
sell), or you make offers that could interest your increases of up to 30% in CTR and 20% in
customers according to their purchase history. If you engagement when using advanced analytics, but let me
show frequently purchased products, you increase the tell you that limited analysis capabilities are a thing of
likelihood that cross-selling opportunities will be noticed. the past, as all tools are now interconnected. ◊

By Daniela La Marca

MediaBUZZ Pte Ltd - Independant ePublisher for Asia