You’ve likely been collecting and analyzing mountains of customer data and unfortunately, still don’t have reliable answers to key questions listed below that can help you generate new revenue and increase your profits.
- Who is our next best player?
- Who is about to churn?
- How to convert free players into payers?
- What are his/her preferences?
- Which players should be selected for contact/promotion so that the campaign/marketing can have the maximum impact and achieve the maximum net profit?
Why is that? In my view, the answer is simple. Your use of Analytics is bounded to the Business Analytics area, which is to me, use of simpler manipulation technique on past data to answer questions useful for decision making in business, leading to an “average customer” business strategy, vs. “individual customer-centric” approach.
For me, an average customer doesn’t exist. Your player base is made up of lots of different players’ types, which experience the same game differently, behave differently and contribute to your business differently. As such, knowing each player’s behavior and player’s profile is essential, however predicting your players’ behavior and their value is even more vital, especially considering relatively short lifetime of your customers.
The good news is that you can quite easily do it by adopting Predictive Analytics techniques, which is an extension of the understanding gained from backward Business Analytics to predict future events or behavior, applying machine learning and statistical modeling algorithms on your historical and current data.
By using these methods, you’ll be able to get numerous actionable insights that will help you to gain ROI from each of your customer related activity and make more money. I have pointed out only 5 of them, which should be the most appealing for you, based on my best practices in iGaming Business:
- Customer Lifetime value (LTV) – by predicting the future value of your each player, you’ll be able to manage your customer acquisition and retention costs (CAC & CRC) more wisely and to verify that your CAC or CRC < LTV.
- “Likelihood to be VIP” – by predicting a probability of a player to be a VIP customer, before he becomes VIP, will enable you to provide your potential VIP customers with a special personalized prize, encouraging more deposits.
- “Likelihood to churn” – by predicting your potential churners, before they defect, you’ll be able engaging them proactively with personalized appealing prizes and reduce churn rates tremendously.
- “Likelihood to respond” – by identifying customers with a higher propensity to respond to an offer, you’ll be able to focus your marketing efforts and use your budgets on those that will say “YES” to you.
- “Likelihood to be woken-up” – by identifying former customers (dormant players), whom is worthwhile to retrieve, you’ll be able to optimize your marketing budgets.
I assume that the next question would be, which prediction accuracy could you expect, if applying those methods and how much money will be made. Well, I believe that the numbers below speak for themselves; however the numbers will vary from business to business.
- LTV prediction: about 75% accuracy after 12 hours following registration and over 90% accuracy after a 2 days of activity.
- Boost in revenues:
5% revenue uplift from existing customers through data driven up-sell/cross-sell activities.
Doubled revenues from Churn prevention and reactivating dormant customers on some products line. - Savings: over 50% reduction in marketing costs.
- Response rates: Response rates are up to 3 times higher than traditional mass marketing campaigns.
Consequently, if you want to increase your revenues and manage ROI driven business, based on facts and not guesswork (and who wouldn’t), I would strongly recommend exploring Predictive Analytics based solutions and start making more money.