Gary Loveman, CEO of Caesars Entertainment Corporation, says there are three things that can get you fired from Caesars: stealing, sexual harassment, and running an experiment without a control group.
Gary, a former economics professor at Harvard Business School, spoke with NPR’s Planet Money hosts Jacob Goldstein and Alex Blumberg about the surprising results of real-world experiments that turned a casino into an economics lab.
In the podcast, he describes three strategies Caesars uses that can also help you boost your conversion rate:
Strategy 1. Identify Low-cost Incentives that are High-value to Customers for a Win-Win
A casino has a lot of ways to lure customers in: free hotel rooms, food, casino chips, limousine rides, golf time, and spa treatments.
But certain incentives, like a free meal at a restaurant, are expensive giveaways for the company. Restaurants make next-to-nothing on food, so a comped $1,000 meal costs Caesars about $1,000.
So how does Caesars get a customer’s attention without breaking the bank?
Gary knows what each incentive actually costs the company, but he also evaluates what value a customer will ascribe to each incentive. When the cost to Caesars is low and the value to the client is high, that’s the ideal freebie.
For example, in the case of the $1,000 meal, it’s high-value to the customer, and high-cost for Caesars. That’s not a good incentive for the company to hand out freely.
The cost of filling an otherwise empty hotel room, however, is low for the company, but it’s a high value for the customer. Exclusivity is a huge perk, as well. Letting a client cut line at Pure nightclub costs nothing for Caesars, but it’s extremely high-value for customers.
How to Do It: Calculate the Perceived Value Differential
The best incentive costs you very little but means a lot to your prospects.
But let’s say you have several incentives that are equally desirable to the customer based on their perceived value. How do you know which ones to select for testing?
Internet-based research lab MarketingExperiments evaluated tests and incentive offers to develop a mathematical model for evaluating possible incentives.
They developed a metric called the Perceived Value Differential (PVD), which they define as “the difference between the perceived value of an incentive, as experienced by the customer, and the net cost to deliver the incentive.”
Here’s how it works:
Let’s say you have a DVD set that costs you $12 per unit, but has a perceived value of $40, maybe because that’s the going rate for similar products.
The DVD set then has a PVD of $28 ($40 minus $12 equals $28).
Let’s also say you have a second incentive, a workbook that costs $15 per unit and has a perceived value of $35. It has a PVD of $20 ($35 minus $15 equals $20).
Presuming that the two products are equally desirable to your prospects, you’d choose the DVD set, which has the higher PVD, for experimental testing.
Strategy 2. Test Incentive Offers to Get the Desired Result
Casino restaurants, like all restaurants, don’t make much profit on food — they make money on drinks and desserts. So how can a company get their waiters to sell more drinks and desserts? Offer bonuses when a waiter’s tables order more.
But what kind of bonus should you offer? Too small, and the waiter won’t bother. Too big, and the he’ll get obnoxious trying to up-sell the customers. That’s where analytics come into play.
“You need something in the middle that’s meaningful, but not over-the-top,” says Gary, who runs tests to find that sweet spot.
How to Do It: Test Your Incentives
A/B testing can show you which incentive results in the highest percentage of conversions.
There are a few simple steps to run an A/B test:
- Assign each visitor to a variant. For example, you could test two incentives, like a webinar and a DVD.
- Make sure that the visitor always sees that same variant. You don’t want a prospect to see a free webinar offer during one visit, then a free DVD offer the next time.
- Display the appropriate variant.
- Record which variant which users saw.
- Record the target metrics that you’re interested in. In this case you might want to see which incentive yields more signups, so you also need to record signups.
- View the results and make a decision about which incentive to offer.
Note: KISSmetrics can take care of nearly every one of those steps except displaying the actual variant — you’ll need to do that!
Strategy 3. Control the User’s Experience to Make Sure It’s a Good One
Every casino game has odds, and the odds are always in favor of the house.
On average, if you put $100 in the slot machine, you get about $90 back, says Gary. That doesn’t mean that’s what happens to every individual, it’s an average.
The problem is that the new customer who immediately loses $200 and leaves is not as likely to return as the customer who had a few wins and was “up” at some point.
Why? The customer who lost $200 right off the bat had a completely negative experience, and therefore a negative first impression of gambling and the casino.
Gary wanted to change their impression of their experience, so that the new customer who lost big would return to the casino at some point.
To accomplish this, customers have a card they use in the casino, and Gary’s team is alerted when a new customer is playing. If the customer is losing, a staff member will greet her, ask about her experience, and if she is unhappy, give her a free dinner, additional coins, a free hotel night, or a limo ride to her hotel.
In his tests, he found that a first-time customer with a bad gambling experience is dramatically more likely to return if greeted by a staff member and offered a perk. (Gary does say the casino won’t offer incentives or entry into the casino to identified gambling addicts, but that addicts often won’t use rewards cards and will play at several casinos to stay under the radar.)
How to Do It: Control the User Experience to Increase Revenue
In his Warm Gun Conference presentation, “Performance Anxiety: What ‘Designing for Conversion’ Actually Means”, KISSmetrics co-founder Hiten Shah says, “It’s not about the tools — it’s all about the user experience.”
He outlines the following steps to improve user experience:
- Ask your visitors to help you improve your website to learn what visitors think you do and what they remember.
- Watch videos of people using your website and listen to their problems to learn why visitors leave and what confuses them. To do this, use an on-site task completion survey to ask questions like whether they found what they were looking for or if the pricing plan is clear.
Next, measure your results by aligning your business goals with data from your metrics and tests. Here’s how:
1. Figure out how you make money. Is it:
- Media (content consumption), such as newspapers, Hulu, and Spotify
- Transaction (purchases), such as Groupon, Amazon, and Zappos
- Subscription (continuous value), such as Netflix and Salesforce.com
- Social (communication), such as Facebook, Twitter, and Tumblr
2. Then, align your metrics with how you make money:
- Media companies focus on pageviews and need to drive traffic to increasing advertising revenue.
- Transaction companies need to focus on revenue and increasing their conversion rate and the number of transactions.
- Subscription companies should be focused on increasing the lifetime value of customers and reducing churn.
- Social companies should focus on increasing the number of users who come back regularly.
Now are you ready for some really good news?
Unlike the guests at Caesars, you don’t have to take a gamble. You can test these three strategies for yourself and let the data tell you what’s a sure bet.
Think about whether your incentives are high-value and low cost and achieving the desired result, as well as how you can improve your user experience. Then try testing one of these strategies to see if you can get an even better result.
About the Author: April Dykman is a guest blogger for KISSmetrics, you can find her on Twitter here.