Select Page

Statistics 101, Instagram TV Stats, Content Curation

In The Headlights

Get ready for Google Analytics 4 with an in-depth audit of your existing Google Analytics setup >

Every four years, the United States of America gets another big lesson in statistics. In 2000, it was a lesson on data quality. In 2016, it was a lesson in learning that forecasting models don’t work on something that’s never happened; you can’t predict the unpredictable. And in 2020, it’s a lesson on statistical significance.

Regardless of the outcome, I’ve read with dismay the analyses people are putting together that show a lack of understanding about statistical significance, especially with things like margins of error and margins of uncertainty. Let’s say you have 100 votes. You know 95 of them. 5 of them are unknown. That’s your margin of uncertainty. Because you don’t know what those 5 are, you might not be able to determine the outcome in this voting example.

Suppose Candidate 1 has earned 50 votes. Candidate 2 has earned 45. Can you declare a winner? No – because it’s possible that the 5 votes you don’t know could tie the race. Now, if Candidate 1 has earned 51 votes and Candidate 2 has earned 44, you can declare a winner, because even winning all 5 remaining votes would not push Candidate 2 ahead. The margin of uncertainty is narrower than the known outcomes.

We do this all the time in marketing, too. When we do A/B tests, we’ll see test A has 46 conversions and test B has 45 conversions and declare A to be the winner. In the case of A/B testing, we have to ask whether that’s statistically significant. How likely is that difference simply a matter of chance? If we have 91 people on our mailing list, then there’s zero chance of it being chance because we’ve accounted for every person. If we have 91,000 people on our mailing list? There’s no predictive power in our calling a winner from the test because the margin of uncertainty is much too high.

We see this most often when non-analytically-minded people attempt to interpret survey data and market research data. I’ve lost track of the number of times we’ve had to gently correct an overeager PR person’s pitch saying that “research shows an overwhelming number of consumers love our client” when in fact the data shows nothing of the sort because they omitted uncertainty and their survey was statistically invalid.

We should take two things away from the current math lesson. First, we need to understand what statistical significance is and be very cautious about how we interpret results when we don’t have all the data. Second, when we’re sharing results and reports, when it’s applicable, we should also be sharing our margins of uncertainty and tests of statisticaly significance to let our stakeholders know that a result may not be reliable.

If you’re unclear about how to make those kinds of determinations, seek out training and professional development, especially in statistics and probability. IBM offers a free class in Statistics 101 at their Cognitive Class online university as a good starting point.

The Bright Idea

In this week’s In-Ear Insights, we discuss the latest findings on Instagram engagement rates, especially for Instagram TV. You’ll learn:

  • How brands fared in 2020 with a global pandemic and multiple, conflicting crises
  • How influencers fared and whether influencers still as powerful
  • What you should do with your social media monitoring strategy in 2021

Watch/listen to the most recent episode of In-Ear Insights here

And in last week’s So What? (which airs Thursdays at 1 PM Eastern on YouTube), the team tackles setting up a content curation system to boost your personal brand.

Watch the most recent episode of So What? here!

Coming up on this week’s So What?, we’ll be tackling part 4 of MarTech basics for personal brand and small business sites, including how to set up a personal brand monitoring dashboard in Google Data Studio.

Are you subscribed to our YouTube channel? If not, click/tap here to subscribe!

Rear View Mirror Data

In this week’s Rear View mirror, we revisit a statistic we first examined in August – the appearance of new Slack and Discord servers. For those unfamiliar, Slack and Discord are private social networks; a group of people start a server and invite their friends to have conversations with them. For example, our Analytics for Marketers Slack group is a group we started to talk specifically about analytics.

Many companies have set up private Slack and Discord servers, but others have made public ones that anyone can apply to join as a way of nurturing and growing a community away from the grasp of Facebook and other mega-networks.

How popular are these kinds of public servers? To find out, we looked at the number of unique inbound links to Slack and Discord (because each server has a unique URL, either x.slack.com or discord.com/invite/x) over the last year. Note that Discord changed domains from discord.gg to discord.com in late June, hence the large orange spike.

Data

Above, Slack is represented in blue, Discord in orange. The start of the pandemic in America is noted with the horizontal grey line. The lines shown are 7-day moving averages.

What we see is that in terms of public communities, even after the domain migration, Discord continues to add approximately 250 new public servers a week, while Slack adds about 36 new public servers a week.

What’s more, these servers attract massive traffic; the top 100 Slack servers have a median of just over 10,000 visits per week to their links. The top 100 Discord servers have a median of just over 1.6 million visits per week to their links.

The key takeaway here is that private social media communities are thriving and growing, attracting millions of people every week. If you’ve been frustrated with your attempts to grow a community on services like Facebook Groups and LinkedIn Groups, give some thought to starting a Slack or Discord server of your own.

Are you a member of Analytics for Marketers? We’ll be posting the top 100 Slack and Discord server URLs in our analytics community today as an exclusive.

Methodology: Trust Insights used AHREFS SEO software to extract 11,194 unique links to Slack.com and 64,997 unique links to Discord.com. The dates of extraction are January 1, 2020 – November 3, 2020. The date of study is November 4, 2020. Trust Insights is the sole sponsor of the study and neither gave nor received compensation for data used, beyond applicable service fees to software vendors, and declares no competing interests.

In Case You Missed It
Partner Spotlight

AI Academy for Marketers is an online education platform designed to help marketers understand, pilot, and scale artificial intelligence. Join Katie Robbert, CEO of Trust Insights, and Christopher Penn, Chief Data Scientist of Trust Insights, for three separate courses in the academy:

  • 5 use cases of AI for content marketing
  • Intelligent Attribution Modeling for Marketing
  • Detecting and Mitigating BIAS in Marketing AI

The Academy is designed for manager-level and above marketers, and largely caters to non-technical audiences, meaning you do not need a background in analytics, data science or programming to understand and apply what you learn. One registration gives you unlimited access to all the courses, an invitation to a members-only Slack Instance, and access to new courses every quarter.

Join now and save $100 off registration when you go to TrustInsights.ai/aiacademy and use registration code PENN100 today.

Interested in sponsoring In The Headlights? Contact us for sponsorship options to reach over 10,000 analytically-minded marketers and business professionals every week.

Shiny Objects

Shiny Objects is a roundup of the best content you and others have written and shared in the last week.

Data Science and AI

SEO, Google, and Paid Media

Social Media Marketing

Content Marketing

Get Back To Work

We’ve changed things up in Get Back To Work, and we’re looking at the top 310 metro areas in the United States by population. This will give you a much better sense of what the overall market looks like, and will cover companies hiring in multiple locations. Want the entire, raw list? Join our Slack group!

What do you do with this information?

By looking at this data, you’ll see what the most popular titles are; use any of the major job/career sites to ensure your resume/CV/LinkedIn profile matches keywords and phrases for those titles. For companies, search job sites for those companies specifically to see all the open positions and apply for them.

You can also hit up LinkedIn and see who you know at companies listed, and see if your connections have any inside tips on hiring.

Top Marketing Positions by Count, Manager and Above

  • Marketing Manager : 390 open positions
  • Digital Marketing Manager : 248 open positions
  • Social Media Manager : 205 open positions
  • Marketing Director : 125 open positions
  • Director of Marketing : 109 open positions
  • Account Manager : 99 open positions
  • Product Marketing Manager : 62 open positions
  • Demand Generation Manager : 61 open positions
  • Project Manager : 61 open positions
  • Communications Manager : 59 open positions

Top Marketing Hiring Companies by Count, Manager and Above

  • Pearson : 163 open positions
  • Google : 134 open positions
  • Amazon.com Services LLC : 106 open positions
  • Deloitte : 75 open positions
  • Thermo Fisher Scientific : 65 open positions
  • New Relic : 57 open positions
  • Microsoft : 31 open positions
  • JPMorgan Chase Bank, N.A. : 29 open positions
  • Facebook : 28 open positions
  • Cozymeal : 27 open positions
  • Interdependence : 27 open positions
  • Verizon : 27 open positions

Top Locations of Hiring Companies by Count, Manager and Above

  • New York, NY : 582 open positions
  • Austin, TX : 322 open positions
  • Seattle, WA : 278 open positions
  • Chicago, IL : 271 open positions
  • San Francisco, CA : 271 open positions
  • Remote, NA : 268 open positions
  • Los Angeles, CA : 205 open positions
  • Boston, MA : 197 open positions
  • San Diego, CA : 175 open positions
  • Denver, CO : 170 open positions

Methodology: Trust Insights uses the Indeed.com API to extract open positions from a geographic area focused on marketing analytics, marketing, social media, data science, machine learning, advertising, and public relations, with a filter to screen out the most junior positions.

Our Featured Partners are companies we work with and promote because we love their stuff. If you’ve ever wondered how we do what we do behind the scenes, chances are we use the tools and skills of one of our partners to do it.

Join the Club

Are you a member of our free Slack group, Analytics for Marketers? Join 800+ like-minded marketers who care about data and measuring their success. Membership is free – join today.

Upcoming Events

Where can you find us in person?

  • MarketingProfs, November 2020, virtual
  • Agorapulse Twitter Summit, November 2020, virtual
  • MadConNYC, December 2020, New York City

Going to a conference we should know about? Reach out!

Want some private training at your company? Ask us!

Stay In Touch

Where do you spend your time online? Chances are, we’re there too, and would enjoy sharing with you. Here’s where we are – see you there?

Required FTC Disclosures

Events with links have purchased sponsorships in this newsletter and as a result, Trust Insights receives financial compensation for promoting them.

Trust Insights maintains business partnerships with companies including, but not limited to, IBM, Talkwalker, Zignal Labs, Agorapulse, and others. All Featured Partners are affiliate links for which we receive financial compensation. While links shared from partners are not explicit endorsements, nor do they directly financially benefit Trust Insights, a commercial relationship exists for which we may receive indirect financial benefit.

Conclusion

Thanks for subscribing and supporting us. Let us know if you want to see something different or have any feedback for us!

Pin It on Pinterest

Share This