Last fall I discovered that voice-to-text had finally become useful on my mobile. It was 85 to 90 percent accurate in a quiet room. It felt like magic.
Since then, my morning walk to work in Barcelona has been full of dictation. No longer do ideas just drift in and out of my head. I capture them. I use those thoughts to guide my work throughout the day.
We wanted to make the experience of getting to know dynamic pricing very interactive. It should be engaging and fun. How? No one wants to look at a spreadsheet. No one wants to see another Microsoft Powerpoint presentation.
Unfortunately, passing motor scooters can bring my voice-to-text dictation accuracy down to zero percent. According to the World Health Organization, their noise pollution also makes my commute through the Example neighborhood a health hazard. These people-moving, mechanical insects (Vespa is “wasp” in Italian) are a constant presence along my 20-minute commute. They make the voice-to-text on my phone 50 percent accurate and 50 percent mysterious, on average.
One morning I arrived at my office near Gaudi’s La Pedrera building, excited by the epiphany I’d experienced on my walk. This is what I found in my dictated notes:
“When showing new product there mix heavy certain demo price. Start the timber …” Well, you get the idea. Disappointed, I rushed to my whiteboard and sketched out the ideas:
- Dynamic curves
- Shopper variable selection
- Year’s worth of data
- Minimum gig
- Revenue impact
It took four months for those ideas to turn into anything useful. That’s the rhythm of creating new things. An idea hits you out of nowhere. Then you work on it for months. They need time to take shape, to become real things.
Thomas Edison’s old ratio still holds true. Creating new products is about 1 percent inspiration and 99 percent perspiration. It’s even a higher ratio of perspiration in my case: 20-minute walk / 4 months (175,200 minutes) = 0.0001 percent.
What took months? Could it have been faster? It takes a frustratingly long time to launch new products. We took many steps and missteps. Is everything that happened during those four months essential to launch a new product?
We’re fans of the “Lean Start Up” by Eric Ries because of its systematic approach to developing new products. We love the idea of the minimum viable product and constant iterations driven by customer feedback. Here’s some insight into our recent experience of launching a new product. We’re not done because a product is never done. This is the story so far.
When I look at Apple products (like the Mac Book I’m typing on right now or the iPhone next to it), I don’t see genius.
What I see is something that meets my needs simply, without complication.
I never had a good relationship with competing Microsoft products. When I used them in the past I felt a low level of anxiety. The blue screen. The freezing. Losing work because I hadn’t hit “save” often enough. It’s strange to remember that time. I didn’t even know how much it bothered me until I stopped. Try remembering it now. Does your stomach feel a little queasy?
With Apple products I feel an emotional reaction of the same intensity. But it's in the opposite direction. Instead of randomly feeling stuck I occasionally feel delighted.
As a consumer I’d hardly associate that positive feeling with genius. That’s a big word. It's reserved for people who can rub two atoms together and create a mushroom cloud (maybe “evil genius”). Or the guys at SpaceX who can land a rocket on a platform that’s bouncing up and down in the ocean. That’s genius.
But my respect for the great intelligence at Apple is growing. I'm also trying to create new, innovative products and launch them in the marketplace. That is really, really hard to do. Apple makes it look easy. It isn’t, as other creators and starters already know.
First, it's hard to create new products that matter. Most everything you need already exists. There isn’t a lot of space in which to create.
Fortunately, in technology, we have two advantages. First we have lateral thinking. Something that works in another industry may work for you, too. Great calligraphy engages us on the printed page. Apple recognized its importance and created the first catalogue of fonts on the computer screen.
Then there is the constant forward movement of innovation. The most famous example of this is Moore’s law. Computing power gets cheaper and more powerful on a regular schedule. You can build on predictable trends.
But at the same time, the new thing that you can create doesn’t necessarily matter. There are no guarantees that it will move the needle for your customer base.
Next, new products are hard to explain because they are, by definition, new. “1,000 songs in your pocket” sounds good to a consumer. It created a lot of demand for the first iPods. Unless you have tried to come up with a similar slogan for a new product it’s hard to fully understand how brilliant that phrase is. They must have gone through a thousand variations before they hit it.
One of the new products we’re launching is dynamic pricing. It’s a combination of billing and artificial intelligence. With our dynamic pricing the benefits are clear. Give each shopper the right price at each moment and they spend more money with you.
But what about communicating it? A product that stays inside of a company isn't "launched." How do we take it out to the market?
Giving people different prices is something that is new but it’s also a practice with deep historical roots. There were no fixed prices until about a hundred years ago. Henry Ford invented it when he decided to sell his Ford Model T nationally. It was also famously available in any color “as long as it was black.” So, we’re talking about something that is old as the hills but also new.
We wanted to make the experience of getting to know dynamic pricing very interactive. It should be engaging and fun. How? No one wants to look at a spreadsheet. No one wants to see another Microsoft Powerpoint presentation.
And we didn’t want to have talk a lot about it. Could a partner of ours “get it” with almost no explanation? Was that possible?
Did we have interesting results to share? Well, we had results from our testing. It included data from the largest companies in the industry and millions of shoppers. We could show the effect of price changes on revenue. If you give certain shoppers price X you make $10,000, if you give those same shoppers price Y you make $11,000.
For people who are already selling a product it had to be very easy for them to see how their pricing structure works with different groups of shoppers.
Since spreadsheets were not an option we settled on charts that people could play with. They needed to be able to make the charts move. It had to be interactive.
They choose the shopper based on a variety of characteristics (time, place, device, etc.) then they choose a price. They find out if it’s the best price. Was $40 the best price? Nope, you lost $2,000. The best price was $35 for those shoppers and any price above or below would have lost money.
Then how do you make it all accessible? On the cloud? We would be showing it in trade shows where internet is spotty. Should we use a local version? It's a lot of data and we'd need to install specific apps like Shiny. We decided to do both. The condom and the pill. Just to be safe.
In the end we developed a demo that gives a taste of an innovative product. Thanks to lots of iterations the people who use it actually enjoy it. They often push us out of the way and grab the laptop to run the demo for themselves. That’s what we wanted. They see how it works and it makes sense to them. They “get it.”
We're happy with the reaction. But it’s not the last product we’ll ever create and launch. We’re always looking for ways to improve our processes.
Thierry Arrondo is the managing director of Vendo, which develops artificial intelligence systems that allow merchants to dynamically set prices for each unique shopper.