Originally scheduled to open in February it has been used as a dark store with pickers fulfilling online orders but it is now the flagship - described well in the Amazon blog post and you can get a better idea of the store through this Amazon page for customers.
An Amazon native brand has gone public in China. Designed to be sold on Amazon Anker is now more widely distributed but Amazon remains hugely important to them. It's another version of a DTC brand - the entire business is designed to complement the channel to market. (I have one of the bluetooth speakers and it's as good as our Bose one)
A worry many have over this strategy is whether Amazon could copy you. Well yes, but no more than any other retailer looks for 'inspiration'. Just look at how good DTC brands are now competing with dozens of lookalikes.
The online grocery growth spurt looks set to stay in place but there is going to be some turbulence over market share. Tesco more than doubled online orders - from 600k a week to 1.5m and are hiring the people needed to achieve this. Ocado have to manage the shift from their 20 year partnership with Waitrose to the new M&S one - and are taking a pop at Waitrose;
I have made the switch to Waitrose and it’s actually fine. Just like Tesco they have thrown people at the problem and it works. But as the FT reminds us, picking, packing and delivering is expensive and the move to online will hit the margins for retailers.
Food deliveries have the advantage of being regular and the consequent high utilisation means the costs of the extra infrastructure are paid for. It’s why Amazon has to be a player in grocery - their food delivery can easily add more parcels to be delivered along their route.
The paper argued that marketers should be investing 20 to 25% of spending to finance well-structured experiments. Much of what the article says still makes lots of sense today;
One of the best ways to diagnose a marketing organization's ROI discipline is to assess the extent and quality of the media and messaging tests in progress at any given time. Some will be simple, such as testing higher levels of expenditure or new media for a proven message, reducing the frequency of mailings to see if response rates change, and testing a new advertising message in a particular region. Others, such as a simultaneous test of a new message and new media for a growing segment of profitable customers, are bigger departures from the routine. Marketers who skimp on experimentation, however, may be overtaken by changing media patterns or forced to assume large risks by rolling the dice on unproven programs when markets shift.