Political campaigns are able to finely hone their messages based on demographic and other data, a controversial practice known as microtargeting

Online services have upended the world of advertising by enabling marketers, including political campaigns, to refine their pitches to specific groups of people or geographic areas.

Google and Facebook notably have fine-tuned their networks to allow "microtargeting," giving advertisers the ability to reach small segments of the online population at a much lower cost than for all users of a platform.

For , strategists say microtargeting is useful for finding supporters, raising funds and reaching voters likely to be persuaded by specific messages. However, critics say these narrowly focused campaigns often operate in the dark, enabling the spread of misinformation, including from foreign actors.

Here are four questions and answers on microtargeting:

How does microtargeting work?

Self-serve platforms operated by Facebook and Google allow advertisers to target messages based on users' declared interests, detailed demographics and browsing history, including pages and services they "like."

Political campaigns may be able to select, for example, suburban moms in Michigan or Arizonans who like Donald Trump, which can be far less expensive than a message to millions of people, the majority of whom may not be interested.

"Targeting allows us to really focus on voters who will determine the outcome," said Republican digital strategist Eric Wilson. "Instead of a message to the whole country, we can advertise to people in the seventh congressional district of Virginia."

Additionally, advertisers may use their own lists—which could be from voter registration databases or other public or private sources—to match people's offline and online identities to tailor pitches.

Internet platforms are struggling to find a way to deal with political ads that spread misinformation

What's the problem?

Targeted ads are not unique to the internet. Marketers who use direct mail have long gathered data to find receptive customers, as do telemarketers.

But critics of online microtargeting argue that targeting is often based on highly such as search queries, Facebook "likes" and browsing histories, often with little disclosure to users.

Additionally, platforms "facilitate and incentivize the most extreme content because they allow to speak only to the people likely to be the most sympathetic and responsive," said Daniel Kreiss, a researcher at the Center for Information, Technology and Public Life at the University of North Carolina.

Karen Kornbluh, director of the digital innovation democracy initiative at the German Marshall Fund, said microtargeting can allow campaigns to operate under the radar.

"If different people are showed different messages, it pulls the rug out from having an open debate," she said.

Why not ban political microtargeting?

Google has barred the use of microtargeting, allowing candidates only to select general categories such as age, gender or postal code location.

Twitter said it would ban political advertising, arguing that microtargeting and unchecked misinformation are harmful to the democratic process

Twitter has banned all political ads on its platform, except for non-electoral "cause-based" messages, arguing that microtargeting is bad for democracy.

The moves put pressure on Facebook, which is considering revising its targeting policy and last year removed the ability to use categories including race, ethnicity, sexual orientation and religion.

But both Republican and Democratic strategists condemned the Twitter and Google limits, warning they would help incumbents and make campaigns more expensive for newcomers.

Limits on targeting "would force campaigns to return to the old strategy of carpet-bombing the airwaves with expensive and annoying TV ads,'" said WPA Intelligence chief executive Chris Wilson, who has worked on Republican campaigns, in a Washington Post essay.

"This is how TV-heavy campaigns have been run for decades, and research shows it led to a 20-year decline in voter participation pre-microtargeting."

The Democratic political action committee Acronym agreed, saying the Google ban "means that campaigns cannot even upload their lists of supporters who have raised their hands to support their candidates."

The ban means "Google stands to make a lot MORE money off of campaigns, as we'll have to spend more to find and reach our intended audiences," Acronym said in a statement.

Facebook has said it would keep political advertising as a way to help promote debate, but is reviewing its options for how messages are targeted

What else can be done?

Some analysts call for more transparency around microtargeting instead of banning the practice.

Kreiss said platforms should be required to allow "counterspeech" to any targeted messages, giving political opponents the opportunity to respond to the same audiences.

By offering more transparency—not just for the ads, but the groups targeted—"that would mitigate some of the harm that comes with microtargeting," Kreiss said.

"Having that information out there would enable journalistic organizations to hold those campaigns to account."

Kreiss said the platforms should, instead of banning political ads, set aside any revenues "to fund electoral integrity efforts, to take the commercial element out of it."