Emily Tan
Sep 13, 2012

Make digital experiences so good, customers can’t help but give you money: Google’s Kaushik

HONG KONG - Marathon Sports, Mannings and Ocean Park are three brands that Avinash Kaushik, Google’s digital marketing evangelist, loves—but not online.

Avinash Kaushik
Avinash Kaushik

Ocean Park’s SEO strategy is so bad, said Kaushik, that buying passes is a “torturous” process that makes him scream at the Hong Kong amusement park's site in anguish: “Why can’t I buy tickets? I told you exactly what I want and why. Why can’t you just give it to me?”

Marathon Sports' website’s music, meanwhile, makes him want to kill himself (because he can’t stop it) and if his first experience with Mannings was the personal-care retailer's online site, he’d never walk into one, because after all, “Watsons is always next door.”

“If your digital experience doesn’t give people tiny little orgasms, you’re doing something wrong,”  Kaushik proclaimed to a riveted audience in his opening keynote for SES Hong Kong at The Mira Hotel Tuesday. “It should make people happy.”

But far too many brands not only fail to do so, he noted, they seem to take delight in tormenting their customers. “Isn’t the point of Ocean Park’s website to sell tickets? Why do they hide it? Can’t they afford good website design and great SEO?” he wailed.

Bad data

Quotable quotes by Avinash Kaushik
"Marathon Sports - annoying music, over and over I can't stop it. KILL ME!"
"If your digital experience doesn’t give people tiny little orgasms, you’re doing something wrong.” 
"Most analytics tools are glorified data pukers."
"Find the losers, shoot them, and give more food to your pretty babies." - While talking about web pages. 
"A too-narrow view of success is telling your company to be actively stupid."
"Tweet when you have something nice to say, or just shut the hell up!” 
"Using likes as a measure of success is equivilant to me, walking in a mall in TST, and every time a girl smiles at me, I mark her down as a lover."
"I hate flash, Every time you use flash on your website, a puppy dies!"

One reason for poorly designed websites and SEO strategy could be lousy data, Kaushik said. Which is not necessarily the fault of helpless marketers; most data analytics tools, he noted, are “glorified data pukers”.

“Most of the time when we think of web analytics, the image that comes to mind is a dump truck—there’s just so much trash!” 

Data overload makes it hard for marketers to make smart decisions because they simply don’t know what measurements to look at, which is why many of them resort to the easy ones, such as website hits and clickthrough rates.

“The key is to figure out what your site is for and create a custom report that will tell you exactly what you need to know. Instead of thousands of tabs all in one place, it shows me the most interesting thing straightaway,” Kaushik said.

By giving the boss something intelligent which will immediately show which pages make the most money, the company can “find the losers and shoot them, giving more food to the pretty babies,” which will probably gain the marketers a raise, he laughed.

In these reports though, it’s important to include all three things you use to acquire traffic: owned, earned and paid media. “This shows all the channels I own and how they’re doing across the board," he said. "A balance is important! If everything on the report is paid, then you’re renting traffic, which is expensive and will get you fired. You want a nice amount of owned media—because you want a lot of free traffic!”

Embrace economic value

“The single biggest difference between companies that are going to win on the web and those that are going to lose are those who can compute economic value,” Kaushik said.

Too often, companies err on the side of over-simplicity in this area. “You spend a lot of money to acquire digital traffic," he said. "The average conversion rate for a top US website is 2 per cent. What about the other 98 per cent? Was it a waste of time?”

Most of the time, companies obsess about the people who convert and tell everyone else to take a hike, which is not only a “narrow view of success” it’s “actively stupid” Kaushik said.

“Digital marketers should look at what success means for people who don't give you their credit card, because if you focus on the 2 per cent and optimise only for them, over time, your business will shrink.”

To figure this out, the first step for companies is to identify the single biggest thing their website is trying to accomplish. Using Citibank’s website as an example, Kaushik identified its biggest goal as signing on new accounts.

“But don’t stop here, you also need to determine other success criteria your site is trying to do,” he added. For Citibank, additional success criteria include mortgages, promotional videos, credit cards, financial information—the list goes on.

"These are the things that deliver medium to long-term value that will ultimately deliver economic value," he said. "Focus on understanding holistic success. If the bank were to focus entirely on new accounts, they’d be ignoring customers who may ultimately transact more money with them than any new account owner."

Marketers need to ask themselves if they have a “now, next and long-term map” and if they are executing for that purpose, or if they’re busy painting themselves into a corner.


Brands aren’t really thriving in social, but they want to because they can sense the mass audiences there, which is the problem. The social-media strategy of electronics retailer Fortress, for example, seems to be directly imported from its TV and print executions, said Kaushik. “It’s like they’ve decided they don’t have enough channels to shout on, so they’re going to go on social media and shout at their fans there too!”

“What value is there in that?” he queried. “Why not share tips and ideas? Why not open people to the possibilities of electronics making their lives better?”

Social media was designed for personal interaction, Kaushik pointed out before wondering aloud why anyone would want to personally interact with a faceless entity shouting “BUY ME! BUY ME!” all the time.

Most of the problem brands are having with social media is that they’re not optimising for the right channels. “Optimising for more followers is stupid," he said. "Why optimise for random people to follow you? As for what time you should tweet everyday... you tweet when you have something nice to say, or just shut the hell up!

“And what’s with this crazy obsessions with likes? It’s so stupid! Using likes as a measure of success is equivilant to me, walking in a mall in TST, and every time a girl smiles at me, I mark her down as a lover.

“Measure the right things, when you have good metrics you force companies to optimise right. Measure conversation rates. That’s what God created social media for. Does anyone want to talk to you?”

Like Dr Walter Carl from an entirely different seminar (if you’ve heard it twice, it must be true), Kaushik reminded marketers to measure the amplification rate. “If I have 80,000 followers and I say something good, and they follow it, my second sphere of influence is 4.2 million.”

Likes, and +1’s, have their place as “applause rate”, he said, adding that quality is everything. “No one cares if someone is promoted in your company, but give a good recommendation and people love that," he said. "This will teach you what to say.”

Finally, although social media is “not about making money”, chances are if you’re doing things right, you will make some, so measure that too, Kaushik said. “Understand what the channel is really good at and then you will understand how to optimise your use of each channel.”

“How do we find people and brainwash them?”

Finding out where people learn about you and visit your site from is a lot less simple than tracking Google searches, cautioned Kaushik. “They may do other research on other sites.”

While studying a website to help optimise its performance, Kaushik learnt that the site’s 98,000 conversions could stem from as many as 30,000 different customer journeys. Also, users may not convert immediately, but they may eventually.

“Because about 20 per cent of the site’s visitors weren’t converting immediately, the site was about to eliminate expenditure on the source they were coming from. But that 20 per cent converted later and formed 50 per cent of ultimate conversions!”

To sum up, if there’s one big danger when it comes to data, it’s that of measuring too narrowly, said Kaushik. “Don't focus on a single visit or a single conversion.”

Campaign Asia

Related Articles

Just Published

8 hours ago

Parsing the gap: Agencies grapple with pay ...

Australia has published an employee gender pay gap report for the first time. Advertising fares better than the construction, banking, and consulting industries. However, top holding companies have gaping holes in their pay structure where equity should sit.

9 hours ago

Spikes Asia announces Innovation Spikes 2024 shortlist

This year's shortlist includes 11 entrants from six countries including India, Japan, Australia, Singapore and South Korea.

9 hours ago

Women to Watch Greater China 2024: Felicia Hartono, ...

A creative and dynamic leader, Hartono leads with flair and innovation to set new standards for client engagement.

9 hours ago

Ten questions with Ingrid Verschuren: Head of Data ...

As the generative AI boom sweeps over the media industry, Campaign catches up with Dow Jones' Ingrid Verschuren to discuss how the news and finance conglomerate is embracing artificial intelligence and navigating both its threats and opportunities.