Lots of of the polls forecasted an effortless gain for Democratic nominee Hillary Clinton in the U.S. presidential election, but we now know their prediction models ended up flawed. When data experts are scrambling to figure out what went incorrect, a range of social media analytics companies are touting the point that their measurements did a greater job of tracking the situation out in the real planet, and noticed the possible for Trump’s gain perfectly in advance.

“Analysts monitoring the social media action of equally campaigns on the main social media channels noticed the result of this election coming months ago, and saved speaking about the huge silent voter base that was forming all-around the Republican nominee,” said Phil Ross, a social media analyst at Socialbakers.

“Social media analysts continually sounded the alarm that all of the polls ended up not reflecting the genuine scenario on the ground in the pre-election landscape,” he extra.

Clinton outspent Trump on Television set advertisements, set up extra industry offices, and sent workers to swing states previously, but Trump merely greater leveraged social media to equally get to and increase his viewers.

It seems, too, that Trump benefitted from that outdated adage, “any push is excellent push.”

When the Oct seventh leak of the Obtain Hollywood tape was reported as a very low moment in the Trump marketing campaign, it expanded his social media viewers extra than any other until Election Working day, Socialbakers uncovered.


In the course of the campaign, a lot of analytics companies recognized that a escalating range of swing voters ended up engaging with Trump’s messages on social media.

Of system, at the time, it was unclear that social media engagement would translate into real-planet votes that could put Trump in the Oval Business. Immediately after all, social media is usually household to extra passive participation when it will come to key matters – and routinely criticized for it.

Pundits have prolonged dismissed social media engagement as keeping small which means past elevating recognition at very best, and not usually resulting concrete improve.

In a recent example, critics scoffed at the armchair activism of the Fb Dakota pipeline examine-in motion. A million men and women virtually checked in at Standing Rock right after a viral write-up advised them it would protect the protesters in North Dakota from law enforcement surveillance. (That wasn’t essentially the situation).

In other words and phrases, traditionally speaking, no one was positive that a like or re-share meant Trump was getting a vote. In particular as the polls advised us usually.


Outside of engagement on your own, other companies like SimplyMeasured noticed that social media sentiment toward Donald Trump trended extra positively than sentiment toward Hillary Clinton on Election Working day.

Over the system of the marketing campaign, Trump noticed extra mentions on social media linked with good sentiment than his rival, as perfectly. She closed that gap in late Oct and early November, but the excellent vibes for Trump soared after once again just ahead of November 8th.

Meanwhile, although Trump had trended extra negatively than Clinton up until the last debate, unfavorable sentiment was shut for the two candidates as they headed into Election Working day.


A very similar craze was noticed by the analytics company 4C Insights, today cited by The WSJ, which uncovered that Trump had extra assist on Fb and Twitter from the beginning of Oct by November seventh. Good sentiment for Trump was fifty eight% to Clinton’s 48%, the company explained.

And eventually, Brandwatch’s assessment also uncovered that Trump had extra Twitter mentions all through most of the election cycle, like on Election Working day. Between the start off of November 8th and November 9th at 1:30am EST, Trump accrued extra than four.nine million mentions although Clinton has just above two.7 million.


But not all social media analytics firms believed their data could issue to a Trump gain. A contrarian report from CNET, for example, famous how one firm, Spredfast, believed there ended up too a lot of conflicting indications on Election Working day for them to make a prediction.

Continue to, even social media may have missed a lot of Trump voters. In the course of this election cycle, there ended up a range of silent Trump supporters, as perfectly as those people who ended up quietly leaning in his course, who may have not built their views recognised on social media.

It appears that at minimum some of these supporters only took to social media after a Trump gain appeared feasible. Immediately after Trump took swing states Florida and Ohio, good tweeting in assist of Trump spiked, Spredfast told CNET.

Even so, social media’s success probably indicates it will be offered extra awareness in long run elections. But it can’t see it all. Which is why we still need polls – preferably precise types up coming time.