google-confirms-nov.-2019-local-search-update

Google has now confirmed the numerous reports of a local search update that began in early November. The update is related to Google “making use of neural matching as part of the process of generating local search results,” Google said via the @SearchLiason twitter account.

The name, “Nov. 2019 Local Search Update” follows the naming convention Google began using last year for Google core search algorithm updates.

The confirmation. Google confirmed it began using neural matching for local search results and subsequent local ranking changes at 12:10 PM ET Monday:

The use of neural matching means that Google can do a better job going beyond the exact words in business name or description to understand conceptually how it might be related to the words searchers use and their intents…

— Google SearchLiaison (@searchliaison) December 2, 2019

The use of neural matching in local search doesn’t require any changes on behalf of businesses. Those looking to succeed should continue to follow the fundamental advice we offer here: https://t.co/tPkyuyMjsP

— Google SearchLiaison (@searchliaison) December 2, 2019

What it means. Neural matching allows Google to better understand when users’ queries have local search intent even when the business name or description aren’t included.

There are no required changes a business needs to make as a result of this update. Google directs businesses to the basic local ranking help document published long ago.

Global release. Google posted an update saying “this was a global launch covering countries and languages worldwide.” So this impacted not just U.S. regions but globally, in all countries and languages that Google is available in.

Neural matching at Google. Google said it began using neural matching in search back in 2018 to better understand queries. Similar to BERT and RankBrain algorithms, neural matching helps Google improve query mapping to results, though it differs in its function. Google’s Danny Sullivan has referred to neural matching as “a super synonym system.” See Google’s neural matching versus RankBrain: How Google uses each in search for more details.

Why we care. We now have clear evidence that it is not just your imagination or your data showing ranking changes with the local search results. Google is saying it now uses neural matching to better understand local queries and thus, Google may show different local results because of it. Last year, Sullivan said neural matching was impacting 30% of queries. That’s surely increased with the roll-out of it to local queries.

Google’s advice here remains the same: relevance, prominence and distance are the keys to ranking well in local. It’s just that Google’s understanding of relevance has now expanded with neural matching.



About The Author

Barry Schwartz is Search Engine Land’s News Editor and owns RustyBrick, a NY based web consulting firm. He also runs Search Engine Roundtable, a popular search blog on SEM topics.

facebook-confirms-pixel-outage-may-have-impacted-conversion,-custom-audience,-dynamic-ads-reports

If you have noticed a drop in your Facebook conversion reports this week — or are seeing unexpected fluctuations with your Custom Audiences or Dynamic Ads volume — you’re not alone. The company is confirming with advertisers that its Pixel, which makes it possible to retarget audiences on the platform, suffered an outage that may have impacted standard Pixel events from being logged.

The problem. Between September 16 and and September 17, Facebook Pixel stopped logging events for some advertisers, causing reporting errors that, according to Facebook, may have impacted data connected to conversions, Custom Audiences and Dynamic Ads volume.

Andrew Foxwell, co-founder of social media advisory Foxwell Digital, posted a screenshot Wednesday of the message he received from his Facebook representative about the outage:

Unresolved issues. While Facebook claims everything should be loading as expected now, Foxwell isn’t so sure.

“Reports are coming back in order, but not yet fully recovered,” he said Thursday, “We are currently working to understand if data will back-fill into reports.”

Foxwell also said that he believes the outage impacted more than just what Facebook listed in its message. His agency has noticed reporting issues on performance as well. Another advertiser, Christian Lovrecich of Lovrecich Media, replied to Foxwell’s tweet that he was still experiencing a massive drop in performance reports across multiple accounts.

We’ve reached out to Facebook for comment on the problem and how many advertisers may have been impacted, but have not yet received a response.

Why we should care. Reliability issues with Facebook’s Ads Manager platform have been a source of frustration for many advertisers for months now. This, combined with the fact that advertisers often have limited access to Facebook representatives when a problem arises, means marketers get left in the dark when reporting is incorrect or they aren’t able to access their campaigns.

If you saw substantial downturns in performance reported earlier this week, it may have been a result of the outage. Take note of the timing for your reporting purposes.



About The Author

facebook-confirms-it-is-considering-removing-“like”-counts

Facebook is following its efforts on Instagram and experimenting with removing “Like” counts from posts. The test was first spotted by app researcher Jane Manchun Wong.

“Currently, with this unreleased feature, the like/reaction count is hidden from anyone other than the creator of the post, just like how it works on Instagram,” wrote Wong, “The list of people who liked/reacted will still be accessible, but the amount will be hidden.”

A Facebook spokesperson confirmed the company is considering hiding “Like” counts on Facebook, but had nothing further to share.

Why we should care

Social platforms are paying more attention to the benefits — or lack thereof — attached to vanity metrics. Reports that Instagram was considering removing “Like” counts from post began as early as April, with confirmation later in the year it was testing the feature in seven different countries. Last week, YouTube said it was beginning to phase out exact counts for public-facing subscriber numbers on channels with 1,000 or more followers.

News that Facebook is going down the same path it already started on Instagram increases the likelihood it could have a broader roll-out across both platforms. This move may not potentially have measurable impact for branded posts on Facebook, but it could affect the brand-influencer relationship — especially in terms of how brands and advertisers evaluate which influencers they may want to build partnerships with.

Influencers will have to craft more valuable content, no longer able to rely on “Likes” to catch the attention of a brand. And brands will need to become more invested in the evaluation process — looking beyond how popular someone’s posts may appear to be from the number of “Likes” they receive.

More on the news

  • After removing “Like” counts in Canada, Instagram confirmed to TechCrunch in July it was testing removing the metric in Ireland, Italy, Japan, Brazil, Australia and New Zealand.
  • When Instagram first began experimenting with removing “Likes,” the company said “exploring ways to reduce pressure” on the platform is something they’re always thinking about.
  • Lara Logan, a VP at the influencer agency Collective Bias, told Marketing Land that removing “Likes” opens up opportunities for influencer agencies to add value and transparency to the industry.


About The Author

instagram-confirms-it-is-testing-increased-ad-loads-in-stories
marketers who reported a noticeable uptick in ad load on Instagram, with ads accounting for as much as 22% out of 45 posts and 23% of Stories. Now Instagram is confirming it is experimenting with increasing ad volume in Stories.

“We’re always testing new ad experiences on Instagram and want to learn how people and advertisers respond to this small test,” said a spokesperson from Facebook, Instagram’s parent company.

The test involves running ads from two different advertisers back-to-back within a Story, according to a report from AdWeek. The experiment is only being performed on a small group of users, and all advertisers are eligible to be part of the test.

Why we should care

While this is still only a small test, marketers who haven’t started running Story ads should take note. If Instagram makes way for more ads in Stories, advertisers who are already familiar with the ad format and have experience running success campaigns will be in prime position to to take full advantage of the increased ad inventory.

There is also the potential for ad fatigue by users if back-to-back Story ads become a more regular experience on the platform — making it even more important for brands to focus on the creative aspect of their campaigns to differentiate themselves from competitors.

Aimclear’s VP of Marketing Strategy Susan Wenograd talked about the need for marketers to pay more attention to ad creative earlier this year when we talked to her about how the social ad landscape was changing. “It feels like we’re edging into a different era,” said Wenograd, “I believe there are totally separate effects we are feeling from these platforms maturing, competition becoming much steeper, and the demand consumers have for creative that they can connect with.”

More on the news

  • Instagram told AdWeek the goal of the test is to determine the “right frequency” of Story ads in terms of user experience.
  • When we asked Instagram about reports of increased ad load in July, a spokesperson from the company said Instagram is always improving on the ad experience and that, “Ad load fluctuates based on how people use Instagram. We closely monitor people’s sentiment both for ads and overall commerciality.”
  • Earlier this month, Kenshoo reported Instagram Story ads accounted for nearly 20% of ad spend on the platform.


About The Author

apple-confirms-$1m-reward-for-anyone-who-can-hack-an-iphone
New Apple Store Opens At The Champs Elysee In Paris

Apple is offering between $100,000 and $1 million for anyone who can hack an iPhone.

Chesnot/Getty Images

Apple has massively increased the amount it’s offering hackers for finding vulnerabilities in iPhones and Macs, up to $1 million. It’s by far the highest bug bounty on offer from any major tech company.

That’s up from $200,000, and in the fall the program will be open to all researchers. Previously only those on the company’s invite-only bug bounty program were eligible to receive rewards.

As Forbes reported on Monday, Apple is also launching a Mac bug bounty, which was confirmed Thursday, but it’s also extending it to watchOS and its Apple TV operating system. The announcements came in Las Vegas at the Black Hat conference, where Apple’s head of security engineering Ivan Krstić gave a talk on iOS and macOS security.

Forbes also revealed on Monday that Apple was to give bug bounty participants “developer devices”—iPhones that let hackers dive further into iOS. They can, for instance, pause the processor to look at what’s happening with data in memory. Krstić confirmed the iOS Security Research Device program would be by application only. It will arrive next year.

$1 million for an iPhone hack

The full $1 million will go to researchers who can find a hack of the kernel—the core of iOS—with zero clicks required by the iPhone owner. Another $500,000 will be given to those who can find a “network attack requiring no user interaction.” There’s also a 50% bonus for hackers who can find weaknesses in software before it’s released.

Apple is increasing those rewards in the face of an increasingly profitable private market where hackers sell the same information to governments for vast sums.

As Maor Shwartz told Forbes, the cost of a single exploit (a program that uses vulnerabilities typically to take control of a computer or phone) can fetch as much as $1.5 millon. An exploit targeting WhatsApp where no clicks are required from the user, for instance, can be sold to a government agency for that much, though such tools are rare. Only one or two a year will be sold, from a pool of around 400 researchers who focus on such high-end hacking. “It’s really hard to research them and produce a working exploit,” he said.

Previously, a company called Zerodium was vocal about how much it will pay researchers before handing them to its unknown government customers. In January, the secretive company announced it was offering $2 million for a remote hack of an iPhone.

Krstić said the bug bounty had been a success to date, with 50 serious bugs reported since the 2016 launch.

“>

New Apple Store Opens At The Champs Elysee In Paris

Apple is offering between $100,000 and $1 million for anyone who can hack an iPhone.

Chesnot/Getty Images

Apple has massively increased the amount it’s offering hackers for finding vulnerabilities in iPhones and Macs, up to $1 million. It’s by far the highest bug bounty on offer from any major tech company.

That’s up from $200,000, and in the fall the program will be open to all researchers. Previously only those on the company’s invite-only bug bounty program were eligible to receive rewards.

As Forbes reported on Monday, Apple is also launching a Mac bug bounty, which was confirmed Thursday, but it’s also extending it to watchOS and its Apple TV operating system. The announcements came in Las Vegas at the Black Hat conference, where Apple’s head of security engineering Ivan Krstić gave a talk on iOS and macOS security.

Forbes also revealed on Monday that Apple was to give bug bounty participants “developer devices”—iPhones that let hackers dive further into iOS. They can, for instance, pause the processor to look at what’s happening with data in memory. Krstić confirmed the iOS Security Research Device program would be by application only. It will arrive next year.

$1 million for an iPhone hack

The full $1 million will go to researchers who can find a hack of the kernel—the core of iOS—with zero clicks required by the iPhone owner. Another $500,000 will be given to those who can find a “network attack requiring no user interaction.” There’s also a 50% bonus for hackers who can find weaknesses in software before it’s released.

Apple is increasing those rewards in the face of an increasingly profitable private market where hackers sell the same information to governments for vast sums.

As Maor Shwartz told Forbes, the cost of a single exploit (a program that uses vulnerabilities typically to take control of a computer or phone) can fetch as much as $1.5 millon. An exploit targeting WhatsApp where no clicks are required from the user, for instance, can be sold to a government agency for that much, though such tools are rare. Only one or two a year will be sold, from a pool of around 400 researchers who focus on such high-end hacking. “It’s really hard to research them and produce a working exploit,” he said.

Previously, a company called Zerodium was vocal about how much it will pay researchers before handing them to its unknown government customers. In January, the secretive company announced it was offering $2 million for a remote hack of an iPhone.

Krstić said the bug bounty had been a success to date, with 50 serious bugs reported since the 2016 launch.