When hunger pangs start panging, and you reach for a smartphone or computer to summon a meal, you’ve entered one of the most competitive scrums in business today.

Companies such as Foodler, Grubhub, Caviar, Postmates, and DoorDash are all vying for your order. They’re also elbowing one another to get the best restaurants signed onto their services, with some eateries being very picky about with whom they work. And they are jostling to recruit delivery people willing to ferry boxes of lamb vindaloo or linguine carbonara around town.

It might seem like just ordering-in to you, but it’s a decision that can be life or death for delivery companies. Some have gone kaput in recent months after failing to attract enough customers — R.I.P. SpoonRocket — and there will be no doubt be more casualties.

Why all the excitement? Takeout is a big business — about $70 billion annually in the United States, according to analysts at Citi. But only about 4 percent of that spending went through an app or website in 2014, the most recent year for which data are available.

While that number is bound to grow, it’s such a contentious market that many venture capitalists are cautious. “The market is huge but saturated,’’ says Jeff Fagnan of Accomplice in Cambridge, adding that “Uber is going to be a force in this market.’’ The widely used transportation app is already using its drivers to make food appear on doorsteps in 16 cities around the world, though not yet Boston.

Still, investors put about $5.4 billion into food and grocery delivery startups around the world in 2015, according to research firm CB Insights. That’s about $1 billion more than the money poured into digital health care companies, and more than double what software-as-a-service companies collected.

Two Boston-based startups, Foodler and ezCater, seem to be holding their own. But both are privately held, so there aren’t any revenue figures we can look at.

Founded in 2005, Foodler has 85 employees, and chief executive Christian Dumontet acknowledges the “huge marketing muscle’’ of competitors like Grubhub — the industry giant that went public in 2014, acquired the Boston company DiningIn later that year, and is valued at about $2.4 billion. (Foodler has raised no outside venture capital funding, having bootstrapped its business.)

Dumontet says Foodler seeks to stand out with a “best bets’’ software algorithm to suggest not only the closest restaurants to your location but the ones you’re most likely to enjoy based on your past ordering habits. About 60 percent of orders come through the company’s mobile app, he says, so the ease of ordering with a smartphone is “another place where we try to differentiate ourselves.’’

While Foodler facilitates orders at about 1,300 restaurants in the Boston area, Dumontet notes that many of those restaurants work with other delivery services as well — which makes not only the quality of the app experience but the speed of the delivery and the accuracy of the order important, too.

EzCater, by contrast, focuses exclusively on connecting business customers with restaurants and caterers that do their own delivery. While chief executive Stefania Mallett says the corporate segment is about $19 billion — a fraction of the residential market — “it’s still quite a nice-sized market, and there are more people going after the consumers. In their business, if a sandwich doesn’t show up, one person is mad — who cares? In our case, we’re focused on completely reliable delivery since it really matters to businesses.’’

Vetting restaurants and caterers to ensure that they’ll show up with breakfast at the management off-site is a big part of ezCater’s business. “We look at their ratings and reviews on the rest of the Web, and we have a phone conversation with them,’’ Mallett says. Once on ezCater’s site, food providers accumulate more ratings from customers. And there are consequences for poor reviews, Mallett says. “If you screw up too often, or you’re not reasonable in how you take care of a situation you screwed up, then we will boot you.’’

EzCater, which employs 125, has raised nearly $35 million from investors. Mallett says national coverage is an important part of serving customers like sales reps who need to be able to make lunches appear anywhere they travel. Restaurants and caterers pay ezCater a fee when they get an order; customers “pay the same price as if they’d ordered directly,’’ Mallett says.

On the consumer delivery side, things are more confusing. Delivery services sometimes get discounted prices from restaurants in exchange for generating additional business, or mark up the normal menu prices, or tack on delivery fees in the range of $5 to $7.

And they can sometimes accept orders for restaurants with which they have no formal arrangement. Joe Barone, general manager of Area Four in Cambridge, says he doesn’t like having his restaurant advertised on services “where we didn’t know we were being offered, and they’re using wrong menus, or increased prices.’’ While Area Four does offer delivery through Caviar, it has tried to block others by refusing to accept their company-issued credit cards. “It’s important for us that the people who deliver the product have some stake in it,’’ Barone says.

Yet another delivery startup, Zoomer, is currently hiring couriers in Boston, though company representatives didn’t respond to a request for comment. Entrepreneur Michael Sheeley observes that other startups, including Maple, Sprig, and Ando, are opening their own kitchens and hiring chefs to deliver “food that you can’t get anywhere else.’’

Sheeley previously ran a startup called Chef Nightly, which raised $1.5 million but called it quits in March. “Our problem was getting the word out, and getting people to know who we were,’’ he says. Moreover, he adds, profit margins on each order weren’t very high.

Getting customers to order once is hard enough, but the real road to success in the delivery business will be regular usage, says Rob Go, an investor at NextView Ventures in Boston. “The key will be whether these companies can win over customers that are loyal and use them nearly exclusively over a long period of time,’’ he says. “Otherwise, they will never be able to get back their customer acquisition costs.’’

In other words, just like the boom and bust we’ve seen with cupcake bakeries and froyo shops, not every delivery startup will meet a sweet end.

Scott Kirsner can be reached at kirsner@pobox.com. Follow him on Twitter @ScottKirsner.