Fail Quickly – How Google Likes To Fail, And So Should You

I just read a really interesting book called “What Would Google Do” by Jeff Jarvis. It’s worth a read if you have any involvement in digital business or especially if you’re in a sector that is threatened by the rising tide of the digital age. Among the many interesting aspects of this book, two things really struck home:

1) Google make all their key decisions based on data.
All their staff are expected to bring with them the compelling data that backs up the key decisions about strategy and product direction. No data? No Decision. You’ll notice how every new product or update is run as a beta project and for a while you can choose to run with the new interface or the old one. This is because Google will never make a decision on your behalf. Instead they let you make it. Once they have sufficient data to show that a statistically significant percentage of the user base prefers ‘X’ feature, they officially roll it out and lock it down.

2) Google has a philosophy known as “Fail Quickly”
Where mistakes do occur, they embrace it, learn from it, then move on. It’s not a blame culture, but rather an understanding that if you embrace data as the driving force for all key decisions and you still make a mistake, then either the collection of the data was somehow flawed/skewed or your interpretation of it was wrong. In either case you can adjust and rework.

Gut-Feel, Zero Data Marketing

This got me thinking about how many companies (mostly from more traditional marketing backgrounds), live by the very opposite of these two rules. Firstly, they still live in a ‘guesstimate’ and ‘gut-feel’ world, where decisions are made with partial or even zero data. In that world every decision carries a massive burden caused by increased risk. In that world every campaign is expected to work and failure is a big deal. In that world the decision to go/no-go is a big one, because once the button is pressed, it stays pressed for a long time and is only reviewed at the end of the campaign, when the budget is spent. So secondly, it means that they don’t fail quickly, they fail over a long period of time and that costs them dearly.

Take Risks

A really good example of this is Public Relations. In the old days you’d sign up to a campaign with a PR company. They would diligently work away at getting you coverage and would present you with a lovely (hopefully thick) book of press clippings. The PR company would use a funky formula to tell you the ‘value’ of all the press coverage (basically related to equivalent cost of advertising space multiplied by the inside leg measurement of the owner of the company) and everyone would pat each other on the back and say how wonderful it all is. The marketing team would then look at sales and see if they went up or down and be very happy if sales went up.

That’s the kind of incorrect belief in causal relationships and leaps in assumptions that people who dance around crystals use to prove that it “totally like helps align my chakras” (I bet you just read that in a hippy voice didn’t you?). It’s the kind of approach I used back in the day when analyzing the successes of campaigns for a range of consumer electronics products. But, in today’s digital age, there is a better way. Most PR is online these days. Even where a publication is predominantly print based, they usually publish a copy online. That means that if your PR team works with your SEO team, you can start to get really important data about where traffic has come from. So you can see which publications have driven that traffic and more importantly, which publications have driven conversions (whether enquiries or sales).

Embrace The Data…Embrace The Failures

So I am now speaking to everyone who is responsible for, or involved in, any aspect of digital marketing. Don’t think of campaigns as a single big hairy entity, but rather break it down. Not just by channel, but by stages within each channel. Digital marketing gives you the ability to test, measure and adjust as many times as you want within a campaign, so do it. I still meet marketing professionals with a set-and-forget attitude to running campaigns. It has no place in modern marketing.

What I’m trying to say is that the risks associated with making gut and long-term decisions on a £/$/€ 1M campaign are massive. But the decision to use £/$/€ 5K on one aspect of your campaign then look at the data in a week’s time, is a much easier and less onerous one. At that point you either continue, (increasing spend), or adjust and measure again. No biggie!
Even traditional media, such as TV advertising is now looking at ways to allow testing, measuring and adjustments within a campaign. It’s called programmatic advertising, it’s the next big wave and it’s delivering amazing results for those using it!

But what about those times when you do everything right and it simply doesn’t work? Even then you haven’t failed as long as you have the valuable data that tells you why. That money was an investment in your company and your customer targeting. If you can learn from it, then you and your organization are richer for it.

We Do This Because It Works…

I remember a time when a marketing plan was simply dividing your budget by region and between print ads, PR, a bit of banner advertising and some events/tradeshows. Today there are countless opportunities, some will work for you and some won’t. Many marketing directors still have a mental block about trying new channels. That’s because they are still thinking about never failing and about their big hairy campaign. If you embrace the concept of properly using the data and that failing is great, then you are liberated to find new opportunities. Want to see if Snapchat works for you? Do it. Never tried Social Advertising? Do it. As long as you’re clear about your goals and what success looks like (and you can measure it), then you’ll know really quickly if it worked or not.

On Top Of The World

Data Drives Tactical Growth, Which Drives Strategy

There’s a reason why some of the most successful companies in the world are in the digital space and are filled with 20-somethings at all levels of the organization. In the old days, it was the old-timers with all the experience that made these big decisions. But today, data outranks experience, so as long as you can get and properly interpret the data, you can drive a business forward to massive success. That makes you think differently, so with digital, the strategy can follow the minor tactical tests that make the biggest impact on sales and ROI.

You might be the person in your organization that can champion this approach, or he/she might be sitting across the room from you. But more often, the people that can truly help you is your digital agency. They are normally just itching to have a really meaningful discussion about all the things they’d do differently for you based on what they see from the data. Reach out to them, arrange that meeting. Some of what they tell you may be hard to hear, but if you embrace it, it could change your marketing and your business.

So have fun with the data and if I can completely smash a popular Trekkie quote…Fail Quickly & Prosper.

related content