When shares in the US food delivery app DoorDash jumped by 86 per cent on their first day of trading last month, it did not take long for sceptics to start frothing about another tech bubble.

DoorDash has never generated annual net profits. It has warned investors the losses will only continue as it prioritises growth. Analysts have questioned whether these “on-demand” ecommerce businesses can ever make money. Yet DoorDash is valued at around $50bn; Uber’s market capitalisation is now around $100bn.

With Deliveroo’s multibillion-pound initial public offering soon to follow in the UK, despite 2019 losses swelling to £318m before the pandemic boosted its profitability, I can see why some investors may look back in anxiety at the 2000 dotcom bust.

What is often forgotten about the last tech bubble is the foundation that it left behind, on which today’s more robust internet businesses were built. Billions of dollars were spent on laying fibre-optic cable and creating excess capacity that reduced broadband costs and gave a leg-up to Google, Amazon and Facebook years later.

If the legacy of dotcom mania was its digital infrastructure, today’s pandemic-driven ecommerce boom is creating a new physical infrastructure in our cities, building logistics networks that will last for decades.

The most visible manifestation is the courier and driver network that ferries our goods around. That contract-free pool of labour, while fluid and vulnerable to exploitation, is unlikely ever to disappear.

This is a boon to new entrants, such as niche food-delivery apps catering to specific cuisines or areas, which are finding it much easier to recruit couriers than DoorDash, Deliveroo or Uber did when they started up.

The less obvious part of the ecommerce infrastructure being developed is the delivery warehouses, both big and small. The vast and increasingly automated warehouses built by Amazon and online supermarket operator Ocado can cost hundreds of millions of dollars.

But that investment is vital when — as highlighted by the return of Ocado’s dreaded online queuing system with the announcement of the new UK lockdowns — capacity is often maxed out almost as soon as these facilities come online.

As well as out-of-town storage, ecommerce companies are looking to build small fulfilment centres nearer their customers. Amazon recently struck a deal with the City of London to turn 39 underground car parking spaces close to the FT’s offices into a “last mile logistics hub”. Parcels will be dropped off at the hub, then ferried to their final destination by electric cargo bikes or on foot. The City hopes to remove dozens of vans from its streets every day, reducing pollution and congestion.

Right now, delivery vans are pretty much the only traffic in the Square Mile, but those efficiencies will be all the more important when normal life resumes.

These local hubs are not reserved for companies with Amazonian resources. Delivery Hero, the German food delivery group, has created hundreds of “dark stores” — little warehouses optimised for rapid delivery in the same way that “cloud kitchens” are for takeaway meals. Venture capitalists are throwing millions at similar local grocery start-ups such as Gorillas, goPuff, Getir and Dija.

These are capital-intensive businesses to start, and many will fail. But the convenience of regular, fast deliveries becomes a hard habit for consumers to kick. Once that expectation of 30-minute gratification is created, someone will step in to meet the demand.

The most significant lesson from DoorDash’s success is that such convenience is no longer reserved for city professionals. Early on, DoorDash targeted the suburbs, where less congestion and untapped demand made for an easier sell.

In North America, it is the cities outside the 100 largest metropolitan areas that are the fastest-growing ecommerce markets. Taking a hint from its rival, Uber Eats is now available to 75 per cent of the UK population, up from 50 per cent a year ago. Increased use of robotics in warehouses will further improve these services’ viability in less dense areas.

Similar to laying fibre-optic cable 20 years ago, deploying this new logistics infrastructure takes more spadework than technical innovation. But even if the ecommerce bubble does deflate once lockdowns lift, consumers will be left with a far stronger logistics network that can only bring us all long-term benefits.

Tim Bradshaw is the FT’s global technology correspondent

Follow on Twitter to find out about our latest stories first. Listen to our podcast, Culture Call, where FT editors and special guests discuss life and art in the time of coronavirus. Subscribe on Apple, Spotify, or wherever you listen.