Here’s what I learned this week

My week continued much like last week a bit of frustration with the claims being made by people in our industry.

I consider myself an ethical data driven person.

But truth be told in business, everyone is just bending the rules.

We praise leaders that are more like cult leaders, we are happy to manufacture interactions to hype things.

We change accounting definitions to make things look better than they are.

The list goes on.

Which is a good segue into what I want to talk about this week.

Dark patterns and gaming KPIs.

Gaming Opt-in Rates via Dark Patterns

The majority of agencies and software hyper focus on one KPI.

Klaviyo - Email driven revenue

Attentive - SMS driven revenue

Recart/Amped - Collect more email and sms opt-ins

Formtoro - Subscription to conversion rate (not nearly as sexy and way more complicated)

(Maybe I should change this to subscriber revenue)

The myopic focus on one KPI makes it easier to measure success in a vacuum.

It’s abundantly clear that these narratives work.

It’s why we hear things like “30% of your revenue should come from email” and “You should get between 10-12% opt-in rate on your popups”

Lots of people repeating things that aren’t necessarily true but works for selling this downstream to brands.

I heard this from an agency last week:

"We used Amped for pop-ups primarily, have an average of 12% opt in rates. Pop-ups are our #1 focus since it’s TOF. Double the people entering your email list, double the revenue you generate over time."

This statement stood out to me, because then I looked at the popup in question.

It’s riddled with dark patterns.

• Full Screen

• Moved close icon to the top left

• Images added near it to distract from noticing it

• No clear terms listed

• SMS terms and privacy don’t link to anything

Let me be clear.

I don’t like this trend.

Double the people entering your email list double the revenue you make over time isn’t remotely true unless they all had the same level of intent and were at the same portion of their journey and all had the same goals.

The danger in oversimplifying these sorts of messages is that in encourages behaviors that are often short sighted and can be long term detrimental.

We looked at this last week with real data examples provided by our friend Jordan West.

There is no certification to be an email marketer, there is no test to pass that proves you understand data and statistics.

You don’t have to explain what a confidence interval is to an examiner and how it’s calculated.

I think what they meant to say is the higher the subscription to conversion rate or the more people that convert the more money you make over time.

Because honestly, that’s how these things work.

A list that is large but largely doesn’t convert, isn’t worth much.

The fact is most people still don’t track subscription to conversion rates.

It’s like measuring your success based on number of Tinder matches rather than actual conversations or dates.

Hint they vary based on the form, device, timing, offer, and intent.

What I’d driving at here is that when you evaluate the performance of something you can’t just rely on a single data point.

A few years ago we created a flow that added a gift card to a cart $10 free money they just needed to checkout to claim it.

40% cold traffic opt-in rate.

I was really proud of it, until I realized that we were creating massive liabilities on our books, that we couldn’t restrict what people used the discount on, and that tracking it wasn’t exactly straight forward.

We stopped it after a month when we also started tracking subscription to conversion rates and realized they were being skewed because people were “converting” to just claim the gift card.

It also wasn’t driving the sales we were looking for, was it worth trying, yup 100%, would I do it again, nope.

Data, Statistics, and how to Measure your email team’s actual impact

The fact is that email is really hard to mess up.

The list is all high intent, about 50% of your sales come in from the actual transaction and purchase checkout flow.

Assuming the other 50% of sales come from 20-30% of your email sign up forms it means you have a healthy amount of people in your lists that have actually experienced your products.

If someone doesn’t purchase within 45 days of signing up to your list, the odds of them purchasing drop to less than 1% for the vast majority of stores.

Retention by and large is a myth in terms of someone’s ability to exert a big force over it.

Retention is instead driven by perceived value, use of product, need for more product, product availability, and timing.

This is real data about how people shop:

Above you can see the pretty standard results that you would expect from most stores, in fact across most of our stores, these things look very similar.

The reason, sign up rates are very high intent shoppers already.

This is why email falsely claims all their revenue generated from the Welcome Series.

These people are already super high intent, with the vast majority of them purchasing without receiving a second email.

Being an email marketer has to be one of the easiest jobs in the entire world.

You literally setup a few flows, then you build a template for products and collections, then populate them.

You send two emails per week and maybe toss in a third one on the weekend that is more lifestyle focused.

There’s only a few types of emails that move the needle and the vast majority of the time that needle is being moved from people that have already purchased from the brand.

Types of Emails:

1. Sales

2. New Product Drop

3. Back in Stock

4. Company Update

5. Customer Profile

6. Weekly Lifestyle Newsletter

1, 2, 3, 5 are all product or collection based emails.

4 is just a monthly update to tell people what you’re working on.

6 is just a templated email that allows you to show off the brand’s personality, because you’re looking to create a tribe around your brand of other likeminded people.

The only lever you have in email marketing to drive an increase in sales or to alter timing and need is to adjust pricing to improve perceived value of the goods that you’re selling to the point where you can impact or change the timing and need portion of decision making.

A deal too good to pass up because it doesn’t come along very often.

Think Black Friday.

To all my email marketers that are going to say product education emails and post purchase customer journey emails matter, they often don’t.

The education stuff should be on the product pages in video form.

The problem with email marketing is the mass amounts of efforts have to happen because of brand laziness on product pages.

Build a good product page that’s thorough and all of a sudden creating emails from existing assets becomes a breeze.

I’m not going to lie, I don’t really read emails from brands.

Instead, I just look at the pictures to see if anything looks interesting.

I treat my emails more like social media feeds, can you tell me why I should open the email in the subject line then can you follow through with good images that paint the picture of why I should care.

Here’s how you test people:

Want to know if you're email marketing team is actually driving revenue?

Here's the exact parameters for running a simple test.

Create a Segment - Non-purchasers, signed up more than 45 days ago

Launch a campaign without any sales or offers.

This is how you tell if it's actually the email that's converting someone or it was the initial acquisition channel and your website that is converting someone.

Now when you get sales, some of them are going to come from people that purchased under another email, you won't be able to tell these easily, but a percentage will be from them.

Know your numbers, understand the actual impact.

Why 45 days is the magic number for purchase cycles

Ben Dunlap asked me “Why 45 days specifically? Or should we adjust that number based on our own metrics?”

It’s statistics.

You can adjust based on your own metrics but 45 days tends to play out for most ecommerce brands.

This was a stat from Amazon and Google back in the day and holds true in most of our testing.

Think of it like this:

Someone signs up for a discount on a website.
They don't purchase from that website.
Standard purchases happen within 7ish days.
So they purchased from somewhere else.
Then you look at the max time of being 7 days plus 5 days for shipping, so we're at 12 days, then they have about 30 days to return the product they purchased if it doesn't work for them.

So the evaluation process is max 30 days.

12 + 30 = 42 days

So 45 days is about the max for someone to comeback or change their mind and instead convert with you.

I’m pretty sure that I’ve never come across any email agency that uses this logic to take a different approach with new signups on their list.

It’s low hanging fruit if you can sway the evaluation period.

This is higher level marketing though that requires having and actioning on data that isn’t super easy to come by.

Marketing is suggestive behavioral changes

The vast majority of all marketing is just making someone question their existing behavior.

Are you over the age of 35?

Are you still wearing tennis shoes?

Keep your comfort and up your style with xyz.

Honestly, I don’t even know what shoes are being advertised but if I like the way they look in the photo I’m probably clicking on that ad.

Why? Because it’s talking to me without talking to me.

It knows who the audience is that they want to speak to, narrows it down via the messaging, highlights an existing behavior in terms of the target demographic and calls out a behavior they are trying to change while still providing the benefits behind the reasoning for the existing behavior.

Far too often we don’t go granular enough in advertising.

We have a tendency of selling into pain points rather than behavioral changes.

There’s no real pain for a lot of products as most are commodities.

Everyone is trying to solve for the same problems which means eventually no one has anymore problems.

I found a cool app

It’s not perfect yet, but the fact that the concept exists means that cool stuff could be done.

In truth, this is something that would almost make sense for us to build into Formtoro to upgrade the experience piggybacking on our latest release of auto-apply discounts.

I’d love to have a coupon auto-applied then see the reduced prices as I shop rather than having to add things to a cart.

Less friction = better experience.

So what you’re seeing is:

When we auto add a coupon code it live edits the price of the products the code applies to and then tells people how much extra they are saving as a result of it being auto-applied.

The below result is still massively hacky for now and no where near flawless, but it’s a step in the right direction.

Matt Bahr almost didn’t make it into this email

This is more bullshit. The how did you hear about us does not impact the utility of the product and lead to a higher LTV.

See beginning of this thread around retention.

Here’s the text of the post:

“So excited for this roll out! Comparing HDYHAU data to LTV is incredibly valuable –– in the example below, looking at holiday 2023 performance, Influencer/Youtube/Podcast is driving a much higher LTV over the past 8 months than channels like FB and Instagram AND has similar first order AOV 📈📈📈.”

Correlation and causation baby, at first glance this sounds like great news!

But guess what, Influencers, YouTube, and Podcasts are all related to lifestyle segments which means that their audience is more niche and focused than say Facebook Ads.

Of course the LTV is going to be more if you make a good product, they are already more dedicated to the lifestyle that these channels cover.

Any channel that is more tailored to the lifestyle will out compete broad channels of distribution.

LTV has nothing to do with the source of the traffic, rather the correlation is that people that willingly seek out those channels are more connected to the types of goods the brand is selling.

So just be where people consume content around the stuff you sell.

The better question will always be:

Where do you spend your time online?

Ironically, this is exactly why you would want to ask that question rather than where did you first hear about us.

You want to be where people hang out and target niche communities on those social networks rather than just tracking where people heard about you.

The goal is to arbitrage places where people already hang out, not just collect a ledger of what you’re currently doing.

The Takeaway

Some people messaged me about last week’s email. I don’t call people out to be mean, but rather put together the type of data driven analysis to serve as a teaching moment to help people better understand how to view and evaluate data.

I’m always going to be data and logic first.

Have a great week!

-Jon

Catch up on past posts: https://ecomwithjon.beehiiv.com/

You can learn from me: jonivanco.com