Have you ever dreamed about learning what products your customers would be the most likely to buy in advance?
How great would it be if you could maximize your profits by determining the highest price a customer will pay for a product? What if you could optimize customer service to resolve concerns proactively before they become issues?
Do you think this sort of advanced insight would allow you to make even more money from your eCommerce efforts?
Predictive analytics is making all these dreams a reality by offering solutions for these three areas and much much more.
1. Improve Customer Engagement & Increase Revenue
Different customers engage with a retail site in different ways. Predictive analytics helps look at all the different variables to generate the desired engagement from the customer. This could mean signing up for a newsletter, clicking on a promotion or some other form of engagement.
There are several vendors that help retailers create models to track & understand customer behavior like Alteryx, Attivio, Lattice, SAS, etc. These models can then be tuned as business objectives evolve.
Lattice has researched how leading companies like Amazon & Netflix are using predictive analytics to better understand customer behavior, in order to develop a solution that helps sales professionals better qualify their leads.
This article by John Koetsier on VentureBeat, shows how venture capitalists have invested over $160 million so far in 2014 in predictive tools that help marketers understand how to sell better online and off.
Investors in this space understand the market opportunity as they see predictive tools like Lattice help inside sales reps better score leads by searching all of the publicly available information on a lead, then matching that against the qualities of a company’s pre-existing customer base.
Forrester calls the combination of predictive analytics with customer engagement – predictive apps, and calls it an era that will take digital disruption to its most logical and necessary extreme: a world of hyper-individualized experiences.
It’s because of predictive tools, like Lattice, that Dell was able to send 50% fewer leads, but get nearly double the results, according to this New York Times article.
In this example on PredictiveAnalyticsWorld.com, an undisclosed education portal that is used by 1 in 3 college bound high-school seniors, used a predictive advertising system to better match promotional offers to their existing traffic. The result was “a 25% increase in response rate beyond that generated by the existing system, which translates into an estimated $1 million of ad revenue every 19 months..”
2. Launch Promotions That Are Better Targeted For Your Customers
Promotions are a must-have for a retail business to succeed but it’s not easy to get them right.
According to this study by Oracle, 98% of fast-growing merchants feel that segmentation & targeting are important for their online merchandising strategy, yet more than half are not satisfied with the tools they have for promotions.
Predictive analytics changes that by correlating data from multiple sources to determine a personalized promotion that will work for a customer or a segment.
Macy’s has seen the benefits of predictive analytics by deploying a solution from SAP that results in better targeting of registered users of the website. Within 3 months, Macy’s saw an 8 to 12 percent increase in online sales by combining browsing behavior within product categories and sending targeted emails for each customer segment.
StitchFix is another retailer that has a unique sales model that asks users to take a style survey then uses predictive analytics to match customers with the clothes they might like. If the customer does not like the clothes they receive, they can return them for free return shipping.
Another example, outside the retail world, is Turkcell, Turkey’s largest mobile carrier, that used over 150 data points on their customers, such as usage patterns, device preferences, and location data, to deliver more relevant promotions for their customers in real-time, as well as reduce their overall churn.
However, it is important that you realize, predictive analytics tools are not just a plug & play solution where you dump data in and revenues spill out. According to a research report by Ventana, only 13% of the 2,600 businesses studied considered predictive analytics a critical element of their business intelligence strategy.
David Menninger, former director of research for Ventana, theorizes as to why this might be:
In this article on Predictive Analytics World, author Robert T. Mitchell interviews experts from three analytics consulting firms, and finds that the most common mistakes with deploying predictive analytics solutions are from not having end goals in mind, not removing junk data (due to lack of understanding) & having a company culture that is unwilling to test changes.
For example, Dean Abbot of Abbot Analytics, shared the story of a debt collection agency who wanted to determine the most optimal sequence of actions for collecting a debt, but adhered to a strict set of rules that required it’s collectors to follow the same action every time.
Fortunately, most of the experts interviewed agree that even though flawed prediction models are rarely fatal & can always be improved upon, the consensus is that building good models is a lot of work & can take a significant amount of time to get right.
For the client, this means investing a significant amounts of time & money without an immediate payoff – or worse, a completely wasted investment. John Elder says that it can take up to a year to get everything right, and for that reason even though 90% of the models built for his clients are technical success, only 65% of them ever get deployed.
3. Optimize Pricing To Maximize Profits
Traditionally retailers have used A/B or Bandit Testing to set prices for different products and come up with the optimal price that can result in maximum profits. Problem is, each price is set manually & can be far more prone to human error.
Predictive analytics takes a different approach by building a model to support real-time pricing that uses input from various sources like:
- Historical Product Pricing
- Customer Activity
- Preferences & Order History
- Competitor pricing
- Desired Margins On The Product
- Available Inventory
- & More
This video shows how Uber & AirBnB have cracked the complications associated with setting prices with different variables in the supply and demand equation and how predictive analytics has helped.
Pricing Management is an ongoing process and needs to be monitored closely to avoid automated price changes causing issues in the retailers environment.
Accenture has been emphasizing the benefits of predictive analytics for pricing management for quite some time now.
Their report states that it is never too early for a retailer to experiment with pricing using predictive analytics and, gradually, the retailer can build the analytics talent in-house. While it may not be directly related, it is interesting that since the report was published in 2011, the demand for data analysis jobs has steadily been on the rise.
4. Inventory Management – Stay Properly Stocked & Reduce Overstock
Walmart revolutionized the field of inventory management by asking its suppliers to support real-time inventory management, also called vendor managed inventory (VMI).
Predictive analytics takes it to the next level by minimizing the required / threshold inventory for a product if the predictive model sees no immediate big sales. This helps the retailers allocate their funds to buy products that have a higher demand and greater profit potential.
Last year there was an excellent research paper published on how big data and predictive analytics is changing the way inventory is managed.
Researchers from Sam M. Walton College of Business & Weber State University identified that lack of technical skills for using these kinds of technologies is the biggest roadblock to more widespread adoption of predictive analysis, but that situation is changing with several vendors offering packaged & easy-to-deploy solutions for inventory management using predictive analytics.
The chart above shows how the Google searches for the phrase ‘predictive analytics’ has grown over the years demonstrating that the skills gap is getting narrower with every passing day.
For farmers, this meant 31% less spoilage but for a typical eCommerce retailer, this means less money spent for storage, fewer things in the clearance section, or otherwise completely wasted.
This excellent podcast on Marketplace talks about handling excess inventory and how damaging too much might be for a retailer:
5. Minimize Fraud By Proactively Detecting It
Fraud is a reality of the retail business and billions of dollars are lost every year.
Any technology that can reduce losses from fraud is like a breath of fresh air for retailers. Predictive analytics solutions like those found in IBM’s SPSS suite, allow a retailer to analyze browsing patterns, payment methods and purchasing patterns to detect & reduce fraud. Some retailers are even experimenting using predictive analytics with machine learning to automatically define rules to detect & prevent fraud.
This is required as new types of fraud are being created on a daily.
In a report by Aberdeen, they analyzed the different types of fraud and the readiness to fight them.
The report also highlighted that fraud management & predictive analytics has a long way to go with only 16% of the respondents saying that detecting fraud was a primary use for their analytics suite.
Walmart demonstrated that they are serious about using predictive analytics for fraud management, among other things, when they acquired an upcoming startup named Inkiru last year.
Other retailers have also been using different scoring algorithms for fraud management for several years and now these algorithms are being enhanced to use predictive analytics to catch fraud before it happens.
6. Better Customer Service At Lower Costs
Customer Service is another one of those areas where retailers have too many questions and very few definite answers. Some of these questions are :
- Should the site only have email customer service or having phone service is mandatory?
- If phone service is required then how many reps are needed?
- Does the site also need live chat along with the phone service?
- What is the optimal hold time when someone calls?
- How do we prioritize questions from a loyal or a high-value customer?
All these questions and more can be predicted by building a model specific to the customer service needs of the retailer. The model will get refined over a period of time (just like other models) and start providing more accurate predictions to improve customer service.
Linux distributor Red Hat uses predictive analytics to improve it’s customer service by increasing what they call ‘subscriber stickiness’. According to this study, they have been able to preemptively provide customers with answers to problems they did not even know they had yet.
Hotel chains, like Marriott, are another great example of a business that are heavy users of predictive analytics to exceed customer expectations before, during and after the stay.
Luxury properties like Four Seasons & Ritz Carlton are always trying to exceed customer expectations and predictive analytics helps them achieve that. This story is one of those examples where Ritz Carlton went out of its way to exceed customer expectations but with little use of predictive analytics in this case.
7. Analyze Data & Make Decisions In Real-Time
Streaming analytics is the capability to generate insights in real-time and help retailers make decisions in real-time.
This is important as using predictive analytics with historical data does not help as a retail environment is extremely fast paced. These real-time decisions influence the best day to launch a promotion, identify products that will generate more sales, target segments with specific campaigns, etc.
Netflix is a well-known example of streaming analytics they capture & analyze all customer interactions including when did a customer pause, how many times did she pause, what was the color of the movie title that attracted the customer, etc. to help make recommendations in real-time.
Technology platform Granify also allows for this level of predictive analysis, using historical browsing data to adjust the website in real time – for example, if a visitor is showing behavior that is more congruent with someone who’s worried about sizing, it will emphasize the size chart, however if they’re more concerned about shipping, it will adjust the site accordingly.
So, How Can I Deploy Predictive Analytics?
Going back to the research from Ventana, even though only 13% reported using predictive analytics, 80% reported that they are important or very important to their business.
But before we go any further, I want you to keep in mind that simply having the platform does not guarantee success. Like John Elder was saying earlier in the article, accurate predictive models can be incredibly difficult to build & can take a lot of time and money.
To make sure your investment on predictive analytics isn’t wasted, you should be working with a skilled data-scientist to help you build your predictive models & a skilled developer for integrating with your platform.
Option 1. Predictive Tools That Integrate With The eCommerce Platform
Given the rise of predictive analytics, several eCommerce platform vendors are offering predictive tools & plugins. These should be the first choice if your business is using one of these platform as it is the easiest way to start using predictive analytics without the headache of painful integrations.
Some examples are:
- Springbot on Magento is a pretty good place to start as plans start for companies with 25,000 customers or less. (plans start at $199/month)
- Canopy Labs offers a recommendation engine to offer the right products at the right time using predictive analytics. This also is offered on the Shopify platform. (plans start at $250/month for up to 100,000 customers)
- Custora is a more robust toolset that helps increase customer lifetime value and integrates with Shopify (plans start at $3,000/month for up to 1mm customers)
Regardless to where your eCommerce business is at, proper implementation of predictive analytics into the platform can help you deliver more personalized experiences for each customer.
Option 2. Use An Open Source Predictive Analytics Product
If you already have the technical talent in-house, there are several open source predictive analytics platforms that will allow you to create more custom solutions. These platforms include:
This option will require the retailer to do the dirty work of implementing the open source solution in their environment. This means hiring the right skilled resources to implement these solutions and given these are open source products, there could be a few stepping stones (a.k.a. failures) before success is achieved with enabling predictive analytics in the retailers environment.
Option 3. Buy A Full Featured Suite
This is easily the most expensive option available – a single user license for SaS is $8,7oo – however, they also provide the most functionality for predictive analytics. There are several packaged offerings in this space like:
The benefit of these offerings is that they have pre-built models for different areas like fraud, pricing management, etc. and only require minor tweaking to make them work in a retailer’s environment.
Additionally, most vendors offer some sort of consulting services to deploy the tools versus having the retailer hire or contract the IT resources to do the work.
If you are a newcomer to predictive analytics and need to understand how to get started before selecting one of the three approaches, please review the links on this page to learn the basics & more about predictive analytics and sign up for the popular newsletters.
Predictive analytics technology is critical for retailers to succeed in today’s environment and shouldn’t be ignored.
You don’t have to enable every use case for predictive analytics but should pick the areas that will create the maximum impact by reviewing your desired targets for revenue uplift, fraud loss prevention, optimized customer service, cost savings and better insights.
Also, the benefits will be seen after a period of time so it is important to deploy the models and continue to monitor & refine them while waiting for the business to benefit from this new feature.