The Wall Street Journal's Amazon CEO Andy Jassy's First Year on the Job: Undoing Bezos-Led Overexpansion covers Amazon's over-forecasting of the post-pandemic growth of ecommerce. Key quote:
"Mr. Bezos and other executives had greenlighted a strategy, guided by a revered internal forecasting tool, that overshot the long-term projections for demand from Amazon. Instead, despite early ideas among industry observers about a permanent shift in consumer behavior, the pandemic-fueled growth in online shopping has slowed as in-person shopping has bounced back."
Amazon wasn't the only one who over forecasted the shift to ecommerce.
The Information's Welcome to the E-Commerce Winter focuses on the impact of the slower-than-expected growth of ecommerce sales on startups, VCs and investors. Their key takeaway is:
"Investors and founders seized on the idea that the pandemic would permanently transform how people shop and pay. Those expectations aren't exactly living up to reality."
The chart below (click to enlarge) illustrates the cause of the ecommerce winter. The ecommerce share of retail sales has fallen from a pandemic high of 16.34 in Q1 of 2020. to 14.3% in the last quarter.
And the trend line indicates that e-commerce's share will likely continue to fall, at least in the short term.
This is an example of regression toward the mean.
While regression toward the mean has a specific statistical meaning, the concept also applies to trends.
With trends, regression happens after a shock (a pandemic, for example) accelerates a trend. After the shock ends, the trend tends to decelerate toward the original trend line.
This doesn't always happen, but with ecommerce, it seems clear in retrospect that regression would happen.
Once pandemic restrictions were lifted and the probability of catching COVID decreased, people started shopping in stores again.
Also, because people were no longer couped up in their homes, they started spending more money on services and less on products.
This combination reduced e-commerce's share of retail sales.
Going forward, we expect e-commerce's share of retail sales to level off in the 13.8% range and then start growing at a pace faster than its growth rate from 2011 to 2020.
This pattern - regression towards the prior mean but starting a new trendline above the previous level - is also common.
And many of the trend shifts accelerated by the pandemic will likely also follow this pattern.
