Scott Porad

Just a few random thoughts on Amazon.com…they’ve been on my mind a lot lately…

• Amazon Prime has changed my life.  When buying holiday gifts last winter it was cheaper to to get Prime than to pay for all the shipping.  Since then, I’ve probably ordered 100x what I would have otherwise.
• Amazon Prime Local Delivery is redonkulous!  I’ve used it twice recently to order computer accessories.  I ordered each item around 11am.  Once I got the item at 1pm, the other around 4pm.  That’s about 1 million times easier than shlepping out to some store to get them.
• Did you hear what Eric Schmidt said about the “gang of four” in consumer technology—Google, Apple, Facebook and Amazon?  Struck me as strange that Amazon was in that group.  I mean, I suppose there’s AWS, which I’m sure is a big “platform” business, but I thought they were a merchant…??  Hold that thought for a second…
• See the Kindle plus Kindle Direct Publishing.
• See Amazon Instant Video plus Amazon Studios.
• A-ha!  Own the content and the distribution.  Or, rather, distribute products that you own for better margins.  Very clever.  They’re taking it slow in this area, but they have time to be the tortoise.
• What does the market think:
  • AMZN: 50x forward-looking earnings
  • AAPL: 11x
  • GOOG: 14x
  • Facebook: N/A

How about them apples?

I regularly receive questions from readers about my experience working on the web, at Cheezburger, and so forth. I enjoy it when people reach out to me, which is why I make all of my contact information—including my phone number—available on my contact page. If you have a question for me, on- or off-topic, please don’t hesitate to send me a note.

Recently, someone who runs a medium-sized e-commerce site wrote in:

We’ve been having a debate around what our “average conversion rate” should be.  Currently, it’s 1%, but this article says we should expect 2% or higher. Does this mesh with what you saw at drugstore.com and other online retailers?  If yes, where are we going wrong?

Comparing conversion rates from one business to another is really tricky because it’s not apples-to-apples.   Here’s some factors that confound the conversion number:

• Brand — all other things equal, strong brands are going to convert better because customers know and trust them.
• Frequency — all other things equal, products where the customers shop more frequently tend to convert better.
• Price — all other things equal, products where the price is lower tend to convert better.

Take, for example, my recent experience with Zappos—they converted me on a higher price, with a low frequency item, because I trusted the Zappos brand.

It turns out, this applies to both the online and offline world.  A great example: Starbucks…great brand, high frequency, low price. How often have you walked into a Starbucks and not been converted?  Their conversion must be upwards of 99%!

All that being said, it’s a mistake to get hung up on conversion X% vs Y%.  The thing to focus on is increasing X%.

Let me make two points about increasing X%.

First, I cannot understate how much influence product and price have on conversion.  Shopping online…it’s just so easy to pop in-and-out from store to store to find the best price.  I’m not advising you to be a “low cost leader”.  But, you can fiddle with your web site a hundred ways to heaven and it will barely make a dent compared to having the right product at the right price.  In other words, get the basics of merchandising right first.

Second, the real thing to think about is not “overall conversion”…an important number, but deceiving.  You want to look at “the conversion funnel”:

Visits =>

Multi-Page Visits =>

Add to Cart Clicks =>

Checkout Clicks =>

Place Order Clicks

(The notion of a conversion funnel is an idea that I pioneered at drugstore.com many years ago, and now is built right into Google Analytics.  For free, no less!  A shout-out to Dana Morningstar who worked tirelessly to put together our first funnel.)

The idea here is that you want to look at each step in the process in order to assess where your customers are getting stuck.  Each one of those “=>” can be calculated as a ratio, and you can see…

…what percentage of visitors actually shop?

…what percentage of shoppers actually put something in their cart?

…what percentage of people with items in their cart actually attempt to checkout?

…what percentage of people who attempt to checkout actually place an order?

From there, you can see what’s working and what isn’t, then focus your changes appropriately…keeping in mind that your changes might not be development on your site, but possibly adjusting your product selection or price.

I’m have a friend with a web site who sells laptop cases online (which he designs and has manufactured in China). In the month of January, year-over-year his sales are down quite a bit, so he asked me to help him figure out why.

First, I checked to see if visits were up or down. He has an increase in visitors, and his search engine traffic is way up, so that’s not the problem.

Next, I looked at conversion. Way down. I mean double-digit down. This explains the difference in revenue.

Digging a little further, since search traffic was up, but conversion down, I took a look at his top keywords. For his top keyword, you could tell they were highly qualified customers because they were searching for “ laptop cases”. Even for these types of visitors, conversion was down.

This made no sense to me. He hasn’t changed his site, yet highly qualified prospects were converting at a much lower rate. How could this be?

The answer lied in acting like a user. In went to Google, and searched just like a user would. His site was the first three natural results–awesome, the SEO is working.

The problem was that Amazon, Target and others who resell his product were buying all the paid results. And you know what else they were doing? They were selling his product at a discount!

The exact same product that he was selling for $99 wad on sale at Amazon for $64! All thing considered equal, why would a potential customer have any reason to buy from him when Amazon, Target, et al, are completely trusted sources?

The lesson here is that not all conversion problems are solved by twiddling with your web site. You can redesign your landing pages and checkout flows and what have you until the cows come home, but it won’t make one lick of difference if the store next door is selling the same thing for less.

It reminds me of the expression: to a man with a hammer, everything is a nail. All too often companies spend time fixing something that will never, ever solve the problem because it’s the only “hammer” available. Or, if not the only, then the most convenient.

A web site is just one of the many components to an online retail business–it is the equivelant of the physical store in the real world. Price, product, and location are the other components. All of these work together to convert ptospects to customers.

This applies not only to online retail, but to all businesses that are selling goods or services. The web site is only part of the equation.