Here’s what I learned this week

There’s a lot of marketers that misinterpret data and signals.

Last week I saw two prime examples of this.

I figured I’d take this weeks email to talk about data and context and how they need to work together.

Short term gains v. long term growth

The first post that caught my eye was from Jordan West.

Here’s the post:

These 2 things more than doubled our email +sms opt in rate on our popup 👇

Background…

We were struggling big time with New customer aquisition (probably not the only ones haha)

Our email opt-in rate was tanking unless we used spin to win (who hasn't??)

It generally performs decent but I hate it from a branding perspective and because there are a lot of junk emails that come through for the discount

And so…

1. We changed our email opt-in lead magnet from “15% off your first order” (and hostley - who cares about discounts like these) to

“enter our $300 monthly giveaway”

2. We switched to using Amped pop-up from the Sendlane popup - don’t ask me how… but their popup performs amazing 🤯

I like it because 👇

1. Not training our customers on discounts

2. $300 is substantially less than the discounts we were giving

3. Way more subscribers and they seem to be just as engaged

Thoughts? Questions? 👇

I saw this while I was playing golf and promptly responded with:

“Do you really think you’re gonna get less junk emails doing a giveaway? Best of luck, my friend.”

Yeah it was a little snarky, but in my years of doing popups and collecting emails giveaways tend to provide you with the lowest quality audience, more on that from another post I made last week in a bit of a response to this.

Jordan was nice enough to tag me and share the reporting data:

Detailed Analysis of Pop-Up Performance
Summary Statistics

New Pop-Up (March 15 onwards):
Total Revenue: $172,953.16
Average Revenue per Day: $2,789.57
Standard Deviation of Revenue: $4,662.13
Previous Pop-Up (Before March 15):
Total Revenue: $162,005.19
Average Revenue per Day: $2,655.82
Standard Deviation of Revenue: $4,088.14
Key Metrics Comparison:
Total Revenue:
New Pop-Up: $172,953.16
Previous Pop-Up: $162,005.19
Average Revenue per Day:
New Pop-Up: $2,789.57
Previous Pop-Up: $2,655.82
Standard Deviation of Revenue:
New Pop-Up: $4,662.13
Previous Pop-Up: $4,088.14
Revenue Difference:
Absolute Difference: $10,947.97
Percentage Difference: 6.76%

So there’s two things missing from this number of orders and subscription to conversion rate.

It’s not Jordan’s fault, 95% of marketers would have missed this.

In the post he said they changed from a 15% off coupon code, but the delta in increased performance is 6.76% based on revenue.

But remember there should be a 15% difference if it was coupon code related.

No coupon means that revenue should have gone up by 15%.

Yet it only went up by 6.76%.

So effectively I can infer that there are less sales which means less product in hand which means less repeat orders long term.

My goal is for everyone reading this to be able to spot these things quickly and get in the habit of paying attention to what data is shared and what it really means.

There’s a lot of short term gain porn on LinkedIn and Twitter and Social Media without understanding the broader business implications.

Hacks don’t work.

This is also why just focusing on isolated statistics for performance doesn’t work.

Popup up testing, why opt-in rate rarely matters

I really do believe that people miss the logical reasoning element of strategy when it comes to doing things in ecommerce.

Often we overcomplicate things or make assumptions without fully considering how things interact with each other.

Which brings me to another interesting interaction on LinkedIn that happened late in the week.

David Wardenga posted about popups and tests he was running around opt-in rates as the main KPI.

I watched his video on his methodology but noticed that just relying on the opt-in rate doesn’t provide the context of whether or not something is actually more successful as that alone is dependent on the quality of the audience being driven to the website.

I wrote him in a DM that it’s tough to just rely on opt-in rate alone without looking at revenue, subscription to conversion rate, and other variables.

Inquiring about if those were part of his research of methodology.

The response,

I didn’t have the heart to explain to him that he didn’t drive the revenue, the brand’s reputation drove the revenue, he just babysits a list that is super high intent.

It was a weird interaction.

Look I get it, people get defensive when you ask questions and question the logic behind tests that they are running when they are really proud of results that they are finding. I’m all about testing.

But the truth is most tests people run are inherently flawed and too focused on individual KPIs that can’t be measured in a vacuum and are often miscalculated.

But this theme had me noticing something…

Simple messaging is often repeated by people without asking questions about it

So in both my interactions Amped.io was brought up.

I think they make a pretty product, I like that they are moving into zero party data too, it’s smart, I don’t view them as a competitor though.

Formtoro.com is a data intelligence platform that uses popups to collect data to build out marketing strategy.

Amped is a platform that pushes data into your ESP. In truth, anyone can do that, it’s a commoditized service.

They are riding a wave of old school hack tactics.

1. Full Page Popups

2. Load Immediately when someone hits the website

3. Micro Opt-in or question first

4. Move the close button and change the contrast to make it difficult to spot

5. Focus only on opt-in rate and revenue generated

If you do that above, your opt-in rate will go up, but you’re really just giving away margin to everyone without a real plan.

I know, I’ve run the tests.

Hint, your popup didn’t generate the revenue, I wish I could claim that as the case, but the truth is you just traded margin to more people with a higher opt-in rate.

But ecommerce loves the idea of silver bullets and quick wins and big promises.

Ecommerce doesn’t like old school proper customer research and strategy to optimize entire funnels.

Real marketing.

A lot of “self proclaimed experts” in this industry have very limited knowledge about broader business practices.

Software is 100% to blame for this too.

Lofty claims and reports that hide or simply lack the context necessary for you to make appropriate well thought out decisions.

Almost all internal software dashboards cherry pick metrics that make the software look good.

The truth is data is tough, interpreting it isn’t easy, it requires knowledge, understanding, and experience.

More emails does not mean more money

I’d love for this narrative of more emails = more money to die.

More emails = more money.

Have you heard this before?

If this was true every ecommerce website would only run giveaways.

But websites don't do that.

Why you might ask yourself?

Because it creates a list full of people who's incentive has nothing to do with actually making a purchase.

Which lowers the quality of your list.

80-90% of your revenue comes from people that have signed up from your popups in the last 45 days and people that have already purchased your products.

That other percent, large sales, new product drops, or back in stocks.

So rather that focusing on a KPI of list size, you need to also focus on the subscription to conversion rate of that list size.

Here's an illustration for you.

List size of 1,000,000 people where 30% have purchased before.

List size of 500,000 people where 80% have purchase before.

Which list makes more money?

I'll do the math for you, the second list has 100,000 more buyers on it, it will make you more money.

My frustration with the ecommerce industry is that it's very good at short catchy phrases that are completely devoid of logical and statistical basis.

Outlandish claims with cherry picked stats rather than actual business fundamentals. It drives me nuts.

Email is one of the worst offenders as a category.

The vast majority of whether or not your emails perform isn't based on creative or timing or offers.

It's based almost entirely on what percentage of your list is an existing customer and had a good experience with your product which is balanced with their perceived value, use of the product, need of more of your product, and coupled with relevant timing to purchase another one of your product.

You want to build an email list that converts, it's worth more and it compounds over time.

But hey if you "just need more emails" then run a giveaway.

If someone told you that you need to just be collecting more emails as a means to drive massive growth, they're lying to you.

You could run a popup to new customers with a coupon code in it without the email and you'd probably get better results.

Because it's the offer that creates the action, not the action of providing an email.

Somehow, sometime ago, we started to make up things entirely based on these catchy narratives that are linked directly to the services provided by the companies that sell this shit to you.

We're a popup company and I've never cared about the amount of emails or sms people are collecting, I only care about the % of the list that actually converts, if it doesn't make money it doesn't matter.

The whole reason we emphasize zero party data and analytics is to use that data to attract a higher quality audience that has a higher likelihood of converting.

Retention begins before the first purchase.

I actually wrote another article about offers called:

Why you should always use a % off discount for your popups

I’ve gamed pretty much every KPI at this point in my career, you can’t view any of them in isolation or you’ll miss bigger impacts.

Ok value time, I posted this last week on LinkedIn but I feel like it’s worth sharing again with more screenshots and context added.

How to use data to improve your ads

On your landing pages, create a popup form with an offer that's really good.

Use a Formtoro form so that you can ask a few questions and include a coupon code that's auto-applied to the cart.

Then ask a few questions with our live data collection.

"What about our ads made you visit today?"

and we ask

"Why are you in the market for our product category?"

Then we take those data points and look at what answers are driving the most revenue and conversion rates along with...this is important, this is the contextual information you need to make these decisions, this is the gap of most every software out there. (I'll keep screaming about this!)

"What matters most to you in our product category?"

Guess what ads we create more of?

People make marketing way harder than in needs to be, like miles harder than it needs to be, if you have this information, create these ads, have a good offer, and you're still not getting sales, you don't have a feasible business.

This is a closed feedback loop that hedges against most risk of testing and worthless spend.

It's not an attribution thing either, on the vast majority of paid networks you can't actually control who sees your ads, that's up to the platform.

Your goal is just to speak to the highest quality audience, with the proper messaging.

That's it.

If the messaging is good, it should work, if it aligns with what drives revenue and conversion but still isn't showing results, tweak the messaging within the parameters to unlock something better for the audience.

If you follow the above outline, you will make money, if your ad team isn’t doing this, they aren’t being efficient or they have room to be more efficient.

Also guess which network we’re going to start running ads on?

Hello Reddit.

It’s been an interesting process so far, but it’s untapped in terms of what’s possible. I’m working on documenting the experience as I learn.

Special bonus - work in progress preview

I’m putting together a 90 day plan that’s replicable for every ecommerce brand around how to optimize your popups to maximize gains.

Conservatively in our testing this increases your revenue from first time shoppers by up to 20% over 60 days before you start collecting data.

Any brand will be able to roll out this strategy.

I’ll probably end up filming a series of videos explaining how to implement it.

The Takeaway

Pay attention to what people are talking about, but ask questions, ask for the raw data too when possible, you’ll start to notice trends where, not everything is as it seems.

Have a great week!

-Jon

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

You can learn from me: jonivanco.com