Shadow brands

The best performing digital campaigns test thousands of ad variations to see what works. Yet traditionally brand advertising is known to work by “refreshing and building memory structures”, so the advice is to stick to your distinctive brand assets. These two jobs to be done by marketing seem to be at odds. The difference in approaches stems from different company stage and goals. Direct response channels capitalize on existing demand, so you’re optimizing to what triggers immediate action. Larger companies are well known, so are already in the running for every purchase happening today. Their focus is generating long term demand from customers who won’t be in market for months or years into the future.

Most consumers just don’t care much about most purchases in most categories. So they buy the brand they think of first, or the one they recognize on the shelf, or whatever has the most reviews (i.e. largest market share). With massive sustained investment in branding over a long enough period, it’s possible to win enough share of mind to exclude your competitors from the consideration set entirely. With most of the battle already decided before consumers decide it’s time to buy, the remaining scraps of undecided comparison shoppers are left for the smaller brands to fight over. Even if a large brand got very good at performance marketing, and won most of these in-market skirmishes, it wouldn’t be enough to make their quarter: if they lose on brand, they lose market share, and lose their jobs. So it’s no wonder they focus so much on brand advertising and care little for direct response.

For startups the calculus is different. If you only have 100 customers, geting 100 more is an achievement. What might be a drop in the ocean for an incumbent brand, might be more than enough to get you to your next funding round. You could double every year for the next 5 years before the in-market audience even starts to get saturated. A startup is “a company in search of a business model” and the statistics say 9 out of 10 won’t find one. When you might soon be dead, it makes perfect sense to focus on the short term, and be aggressive in your grid search. If you find product-market fit, all manner of sins will be forgiven. If you don’t, you’re onto your next project anyway. You can afford to throw spaghetti at the wall and see what sticks. When you’re a 30 year legacy brand that sort of behavior would startle your dinner guests. This relative velocity of testing gives startup an advantage. Most of what they try will fail, but if they find something new that works, they can race far ahead of the game, before their larger rivals can react.

If there a way for big brands to get some of the speed of a startup, without rocking the boat at head office? Is it possible to test new ideas at high tempo and continue to innovate, without breaking the association between your brand and its distinctive assets? The solution is a little startup-inspired smoke & mirrors. ‘Smoke testing’ is a way to validate ideas before building them. Add a button for a new feature, and nobody clicks it, you don’t have to waste time building it. Of course you couldn’t do that on your production website (it’d take months to make it into a release anyway), but you could spin up a ‘shadow brand’ that isn’t associated in any way with the mothership. For example Google created a fake pizza brand for Nestle ‘Dr. Fork’ to test creative ideas that would normally struggle to get approval. FreshBooks built a startup ‘Billspring’ which accelerated learning to improve the legacy product. You may not even need to go to all the hassle of making up a fake product. Tim Feriss spent $200 testing six different book titles just on Google Ads, seeing what got the most clicks before settling on ‘The 4 Hour Work Week’. It’s hard to imagine he would have sold over 2 million books with his first choice of title “Drug Dealing for Fun and Profit”. If you’re struggling to test all of your good ideas, consider creating a ‘shadow brand’ to move faster.

About the Book 📕

← Back to Essays 📝