The Hidden Economics of a Slow Website
Economists tend to study visible markets. Demand curves, labor supply, and the price of capital are all comfortable to model because the underlying numbers are reported somewhere. There is one quiet market that rarely shows up in the literature, and it sits inside almost every business that sells online: the internal market for attention on a slow website.
I lead a small team that audits website performance for businesses across sectors. From that vantage point, the economics are surprisingly consistent. A site that is a few seconds slower than its competitors will, on average, convert a smaller share of its traffic. The traffic itself does not change. What changes is the marginal shopper or reader who decides, often without conscious thought, that the site is not worth waiting for.
Friction as a price
It is useful to think of load time as an implicit price. The visitor pays in attention rather than money. Like any price, the elasticity is not uniform. Some visitors are highly motivated and will wait. Others, especially those comparing options across tabs, are operating on near-instant timeframes and will leave at the first sign of friction. The aggregate effect on a publisher or retailer is a smaller realized share of the visitors that paid acquisition channels delivered.
What makes this market peculiar is that the price is invisible to the seller. The advertising platform reports the visit. The marketing dashboard reports the session. Nothing reports the visitors who silently abandoned because the page took too long to render on their phone. The entire transaction happens before the analytics scripts even finish loading, which is part of why it has remained understudied.
Why the loss is uneven
The economic loss is not distributed evenly across an organization. It concentrates in three places.
First, on mobile devices, where networks are slower and processors are weaker. A page that loads in two seconds on a desktop in the office may take six on a mid-range phone on a commuter train. Mobile traffic is now the majority of consumer-facing traffic for most businesses, so this gap matters disproportionately.
Second, on the highest-intent pages. Checkout, application, and contact forms are usually the heaviest pages on a site, because they carry the most third-party scripts. They are also the pages where each lost visitor is most expensive, because that visitor was furthest along the funnel.
Third, on paid traffic. Organic visitors arrive with a slightly higher tolerance for friction because they chose to seek out the brand. Paid visitors are often midway through other tasks, and they bail more readily. The cost-per-acquisition equation is therefore especially sensitive to milliseconds. A small improvement in page speed can translate into a meaningful improvement in marketing efficiency, with no change to the underlying campaigns.
Treating performance as a market input
From a strategy perspective, this is an unusually good cost line to attack. Site speed work is a one-time investment that pays back across every future marketing dollar. It does not require a new vendor relationship, a brand pivot, or a hiring round. It is closer to fixing a leak in the pipe than to building a new pipe.
It also has a social dimension that economists may find interesting. Faster websites consume less energy per visit. A page that ships smaller assets and runs less JavaScript draws less compute time on data centers and less battery life on devices. The aggregate is small per page but adds up across an internet that runs trillions of page views a year. The same work that improves conversion also reduces the carbon footprint of the visit, which is a rare alignment between commercial and environmental incentives.
A quiet line on the balance sheet
Most businesses already track conversion rate and customer acquisition cost. What they tend to miss is the implicit tax that load time places on both metrics. Adding performance as a measured input, alongside copy, design, and pricing, reframes the website itself as a market with its own efficiency. Once that frame is in place, the work to improve it stops feeling like a technical project and starts feeling like what it actually is: an economic decision with a measurable return.
