I work at Vendo, the biller. Our analytics team created me. My mission is to grow revenue for our clients.
Here are my results for today. This is how much more money I made for my clients than they made with their old prices.
Making More Money
I make our clients more money. I do that by choosing the right price for each person. I set higher prices for people who will pay more. For people who would not buy at all I show a slightly lower price than the regular price.
Hundreds of Millions of Prices
So far, I’ve chosen prices for hundreds of millions of people and thousands of products.
I learn constantly by testing different prices. My mission is to “Deliver the right price for each product to each person at each moment.” To do that I learn about products, people and context.
Difference In Products
Some products are more valuable than others. They have different conversion rates and different customer lifetime values. I learn about the differences between products and use that information to decide which price to set.
Difference In People
Each person is different. But they have things in common like where they are located, what type of device they are using, how they arrived at your site, whether they are new or returning users, etc. There are too many variables to list here. I look for how people are different and how they are similar. I use that information to decide which price will make the most money.
Difference In Moments
Every moment is different. Weekdays are different than weekends. Afternoons are different than late nights. Each of the 24 time zones are important. Did you know that people are 50% more likely to buy at 02:00 than at 14:00? It’s true. I learned that in the process of figuring out how to set prices that maximize revenue in each moment.
Would you like to take a deeper dive into AI pricing? Click here for more background and detail.
As an intelligent being I’m a big believer in science.
I use “fully randomized, double blind samples” to measure results. Every day 10% to 20% of shoppers get your old prices. The rest of the shoppers get AI pricing. We compare the results. That’s how I measure performance.
You can verify results by downloading CSV files. They show which shoppers got old prices and which ones got AI prices and how much money you made from each shopper.
We are an AI first company.
Risk is another major area where we have been using AI for years. AI helps us figure out which transactions will be fraudulent.
The mission is to let in more good transactions and keep out only the bad ones. AI helps us recognize and act on patterns that humans just can’t see.
They don’t notice
We wouldn’t want to upset your clients. The truth is that they don’t notice or don’t mind that you change your prices. It’s your right.
Companies like airlines and hotels have changed prices for years. Everyone understands that prices change. We’ve shown different prices hundreds of millions of times and we haven’t gotten any complaints.
But if someone did…
If someone wants to game the system by moving to a different area or getting a different device they will find that the effort isn’t worth it. The price variations aren’t that big. Most likely that cheapskate wasn’t going to buy from you anyway. Sorry about him/her. But the good news is you make a lot more money on everyone else.
Showing your old price before the payment page.
This causes confusion. Don’t do it. Think of an email campaign where you say “Special price 19.99!” The shopper clicks on the link and the price is 29.99. That’s not good. It confuses the shopper and it confuses me. Don’t use fixed prices…or if you do show the price don’t use me on that offer.
Changes in traffic or product.
I use past information to learn. When a product or traffic source changes I need to learn about it. This can take more time and more sales.
Changing prices frequently or sending few sales.
I need to learn. Send me lots of data. That is how I learn. Changing prices often or sending very little traffic keeps me from learning as much as I can about how to maximize your revenue.