Outages, outages everywhere
The year is 1980. Everybody believed that we will be zapping around in flying cars and playing with toys made of Gravitonium in the near future.
Fast forward to 2019 and the reality is that we can’t even keep our websites up consistently for more than a few months.
This time, it was a Cloudflare outage, the second in the past 2 weeks, which took down multiple websites with it.
Shopify, Active Campaign, Clickfunnels, Hubspot, Discord were either unable to load properly or returned a “502 Gateway Error” for several hours.
Even websites which are commonly used to check web service disruptions such as DownDetector were affected. What an irony!
“Massive spike in CPU usage caused primary and backup systems to fall over. Impacted all services. No evidence yet attack related. Shut down service responsible for CPU spike and traffic back to normal levels. Digging in to root cause”.
So, if you were worried about your marketing cavalry not able to load up and pack a punch, it wasn’t just you, just some ‘bad software deployment’ is to be blamed.
Are you getting shadowbanned?
Instagram has been accused of “shadowbanning” users and posts starting about a year ago.
It affects your ability to reach non-followers, resulting in lower engagement and less visits to your profile.
What does a shadowban mean?
- Your posts may be visible to your existing followers, but not to anyone else.
- You might not be able to follow someone for a specific period of time.
- Your follower count is not increasing when you know there have been more people following you.
- Stories are not visible to users when you use a particular hashtag because it was banned by Instagram.
Chances are you have already observed some of these symptoms but you just didn’t know what was causing it. Maybe you googled it and Webmd said it’s cancer.
How can you tell if your account is being shadowbanned?
The Crew came across a tool which examines your latest 10 posts and tells you if you are doing anything wrong with your account or if any hashtag is shadowbanning you.
It’s called Triberr’s Instagram Shadowban Test. It checks your post’s meta data for each hashtag used. If any hashtag is missing from the source code of the post’s page URL, then the test fails that hashtag.
It’s not a perfect tool by any means but it’s better than nothing.
Check hashtag search results directly for a specific post from an account that does not follow the account you’re reviewing. If the post is visible, then it’s fine. If it is not found, then it has been suppressed/shadowbanned.
Yup, it’s a little painful of an approach because some hashtag search results are in the millions. Unfortunately, Instagram does not offer its API to perform this search programmatically.
Firefox helping users fool advertisers
Firefox doesn’t like advertisers at all. The latest update they released does more than help users to not be tracked. It helps them fool advertisers.
Yes! As our job isn’t tough enough already. Ad blockers and pixel blockers weren’t enough for Firefox. They decided to release the ‘Track THIS’ tool.
How does it work?
If you didn’t want to leave tracks on the web, you could just go Incognito. With Track THIS, instead, you will leave tracks… fake ones!.
In fact, the tool allows you to choose from 4 different profiles: Hypebeast, Filthy Rich, Doomsday and Influencer.
Once you pick your favourite, the tool will open 100 tabs according to the profile you chose.
What happens next?
You’ll start to see ads accordingly. Picked Influencer? You will get ads for clothing, makeup, skincare and services for influencers.
Are you an Hypebeast? You will enter whirlpool of Yeezys, Prada, Nike, and Canada Goose.
Well, this seems funny. But it isn’t at all.
Not only does this mess up the data that ad platforms gather on every user. It will also cause advertisers to spend money on impressions that have zero chances of landing a conversion.
However, Mozilla said that this tool is just a way to publicize Firefox’s anti-tracking features.
Let’s see what consequences this will bring. And let’s see how many users will be eager to wait for 100 tabs to load…
Maybe it’s better to have personalized ads, no?
Pushground, a 100B monthly push source, brings you 5 new ROI boosting features
The traffic turning heads and filling pockets for affiliates these past few months is push traffic and Pushground is a source you must test.
Founded in 2018 by ad-tech affiliates with over 7 years of experience, they’ve built one of the biggest push platforms with more than 10B requests per day and average CTR of 2%. They’ve also implemented a bunch of great new features since the last time we featured them:
- Creative split testing. The ability to test multiple creatives for push, where the creatives are the ones that generate the click (or they don’t) is a clear ROI booster.
- Browser targeting. Some browsers are just printing for you? Target more of that!
- OS version targeting. A pesky OS version converts really well? Target it!
- Exclude any of the targeting options available. Oh, most things are doing great for you but some are just a waste? Exclude them! You can exclude Spanish speakers in the US for example, exclude specific carriers and so on.
- New supply lists. Fresh users for fresh results, aka higher ROI. Available now, so jump in to be one of the first to try it out!
What else should you know about Pushground?
- You can filter by how long the users have been subscribed to receive push notifications (it can improve up to 6x times your performance).
- You can bid by source or supply_id.
- Target by Carrier, Country, Connection, City, Device.
- Day parting.
- True real-time data. It’s not just about clicks and deliveries but your winrate and also a breakdown for opportunities.
- 0% fraud with the integration of Opticks.io, an anti-fraud system that eliminates it before selling. Pushground eliminated 10% of its traffic with this to keep everything as high-quality as possible.
- CPC and volumes breakdown by country righ here.
Like most of the industry, Pushground will be in Barcelona for Affiliate World Europe so check them out at Booth C56!
By the way, we have a special treat for WTAFF readers. Get 10% on top of your first deposit from a minimum of $100 and no maximum by signing up and reaching out to support over live chat.
Algo update shoots health products. Lookalikes vs. Interests. Decisional tree revealed!
Do you feel lost when FB Ads performance drops? This might be a solution. FB updates its algo and cracks down on misleading health content. $1k experiment to test LAL vs Interests audiences.
Just another punch to nutra
“Misleading health content is particularly bad for our community.”
This is more or less the juice of the latest Facebook update.
Over the last month, the social network made some changes to its algorithm to downgrade these two kinds of posts:
- The ones that provide exaggerated health claims or misleading content.
- Posts that promote products or services based on health-related claims.
Well, nothing new in Mark’s reign. We already knew that FB doesn’t like this stuff. So far, it was just about seeing your ads getting disapproved.
But now, if you counted on organic reach to promote such stuff (pretty hard these days), this might be another nail in the coffin.
However, you might find your way by following FB policies and without making undue claims. It’s hard but it might help.
FB Ads decisional tree
Performance drop in ad campaigns is a common problem for FB advertisers. And the next question usually is: What caused the drop?
Do you ever struggle with this? Yes? Everybody has experienced it, right?
And that’s why Cem Verghese shared his take on how to proceed when a drop in performance kicks in.
He comes out with the decisional tree he uses to act on bad performances.
The three problems addressed are:
- Decreasing CTR.
- Increasing CPM.
- Steady CTR and CPM but drop in the CVR.
The solutions provided are many. We really suggest you to have a look at the diagram Cem shared in order to have a clear idea of what is happening with your campaign.
Interests vs. Lookalikes
Are you one of the marketers out there wondering if Interests Audiences are working better than Lookalikes Audiences (LALs)?
Well, there’s no guessing here. The only way to do it is testing. However, the AdEspresso team ran a $1k experiment to give everyone a head start. And maybe help you run smarter tests.
Let’s see the setup and the results.
The 2 campaigns looked like this
- Conversion campaigns to get registrations for a webinar.
- The pixel fires when the users fill in all the details in the landing page and register to the event.
- 1 image, 1 copy.
- USA, male+female, age 21-64.
- Exclusions: current webinar registrants excluded.
- Optimized for conversion and paid for impressions using lowest cost bidding without a bid cap.
- Placements: Facebook Newsfeed, mobile and desktop.
- Budget: $500 per campaign over an 8 days span.
Now, what was different?
- 1 campaign had a 1% Lookalike based on a 180-day custom audience created from the Complete Registration pixel event. The initial size was 2.1M people but with the 21-64 age range, it dropped to 1.9M.
- The other campaign used an Interests Audience, with the same size: 1.9M. The interests picked were the ones with the highest affinity score: Bing Ads, Digital Marketer, Jon Loomer, Social Media Today, Animoto.
The AdEspresso team predicted that the Lookalike Audience would have given the lowest CPA.
Interests Audience campaign:
- CTR: 1.10%
- CPC: $2.39
- Conversions: 158
- Conversion rate: 75.6%
- CPA: $3.16
- CTR: 1.26%
- CPC: $2.07
- Conversions: 181
- Conversion rate: 74.79%
- CPA: $2.76
The LAL Audience clearly beats the Interests Audience with a 13% lower CPA as well as a better CTR and CPC.
Although, the AdEspresso team says that with a further Interests split testing, you can still lower the CPA for an Interests Audience.
So, what’s the final take?
Lookalikes Audience are the best? Well, if you already have data, LAL can be a good choice. And the Ad Espresso team claims to use LAL 60-70% of the time.
Buuuut, (there’s always a “but”), if you don’t have data, Interests is the way to go. Also, doing new tests will still benefit your CPA.
So, take some conclusions from this experiment, but don’t take it for granted.
Cool tech, (funny) business, lifestyle and all the other things affiliates like to chat about while sipping cocktails by the pool.
When a negative review is worth $500k
Remember those grumpy customers, competitors or just random people posting fake negative reviews about your business? Well, we found some early signs of an antidote for them.
An Australian Supreme Court has ordered a woman to pay more than $500k in damages for a negative review she posted for a business on Google.
As always, the reason behind it was as crazy as it gets.
She opted for some cosmetic enhancement with Dr Kourosh Tavakoli, a surgeon who is known to be a “household name for elite plastic surgery in Australia”.
Later she claimed that the procedure wasn’t performed by Tavakoli himself. This somehow triggered her to post a negative review for them on Google.
Following that, website traffic to Tavakoli’s site dropped by more than 23% in less than a week.
The business took her to court, she lost and had to remove the review.
Things are not over yet, no, no! A week before the trial, she went on to publish yet another negative review about the procedure on Google.
Eventually, she was ordered to pay $530k in damages for defamation as well as Dr Tavakoli’s expenses for the dispute.
So, it seems you can indeed hold such people accountable for their online comments, especially if they are unfair and truly damaging.
As a user, you should also think twice before going on tirades where you make things up against other businesses.