SMS Welcome Flows - the Truth

With a sprinkle of data

Author

Jon Ivanco
September 16, 2023

SMS Welcome Flows

The following is a post from Aaron Ordendorff who runs an SMS company called Recart.

I like his posts, but they tend to leave data out of things. It’s all narrative.

Context is key when reading anything and a lot of marketers don’t actually have access to data that provides context.

I’ve annotated his post with my take and follow this up with data we have from companies that pokes a few holes in some of these assumptions.

Your SMS welcome flow should … print money. If it’s not, here’s 5 things to fix right now.

SPOILER:

They’re all dead set on (what should be) the single most money-printing text inside your single most money-printing automation.

1️⃣ Offer matching

Whatever your popup incentive is, make it one-for-one with the first text you send.

Nobody misses this on email.

But it’s heartbreaking how many people forget to keep SMS spot-the-f***-on.

The offer. The exact offer. Nothing but the offer.

In fact, if you haven’t checked in a while — stop here and go make ‘em match. Then, make sure your first text is being delivered immediately.

My take on this, from data less people convert generally when asking for a phone number so in most situations if you’re presenting the same offer for subscribing to both an email and phone number you’re better off just not collecting the phone number until after they have purchased.

2️⃣ Lander or dynamic link

Are you sending new subscribers to your homepage?

Don’t. Either send them to …

- Best-selling collection
- Top product (PDP)
- Educational landing page

Or, use dynamic links to send them right back to the same page they signed up on.

My take on this is he’s correct, but most of the time it doesn’t matter, the person is already down the path of making a purchase that you can dump them where ever you like, the key takeaway from this is that you have to do this because you FORCED THEM TO LEAVE THE PAGE THEY WERE ON TO GO TO ANOTHER APP.

I feel like this is what people miss about SMS you’re telling someone LEAVE MY WEBSITE. Which is usually the last thing you want to do to someone while they are shopping. It’s highly likely that this is what contributed to us observing 10% less revenue when asking for a phone number in isolated testing.

3️⃣ Contact card

I’m pretty sure marketers are the only people on planet earth who save contact cards. But if you really want to send one:

Delay it until the third text. And do not send it alone.

Add a link + CTA … this way, that spendy MMS stands a chance of driving ROI (i.e., sales).

My take, going to be interesting to see what happens with Apple Privacy updates to how we handle SMS routing into the inbox. It’s smart, but the important call out here is actually just how expensive MMS is to send.

I wouldn’t in a welcome series at all. You’ll see why with the data at the bottom.

4️⃣ SMS vs MMS

The most important test you can run is SMS — text only, <160 characters, no emoji — versus MMS — long-form(ish) copy + an image or gif …

… and the highest-volume, highest-learnings, highest-payoff place you can put that test is at the start of your welcome flow.

My take, this is correct, but it’s the first message, that’s it. Beyond the first message your likelihood of impacting a sale falls very fast statistically speaking.

5️⃣ Exclude customers & abandoners

The moment someone does the thing your welcome flow exists to get them to do … get them out of it.

Did they buy? Then, bye.

First, add a conditional split at the very start — right after the legally required opt-in message — to exclude anyone who purchased in the last 180 days.

Send them a 2–3 text “welcome back” flow.

Second, also add conditional splits between every message in your welcome flow to exclude recent buyers and abandoners.

Buyers should go into your post-purchase flows. And abandoners into (well) your abandonment flows.

My take, you should have different flows based on different parameters, I agree that you shouldn’t send the welcome flow to anyone that’s purchased, a lot of people don’t exclude for abandoners. I’m not sure I would either, double tapping people that have abandoned for sms might be a good idea to exclude but with email, I would send both. The messaging will be very different in each.

——

Your SMS welcome flow should print money. And the very first text in your welcome flow should absolutely s*** cash. If not …

1️⃣ Match the offer
2️⃣ Lander or dynamic link
3️⃣ Delay contact card + CTA
4️⃣ Test SMS vs MMS
5️⃣ Exclude buyers & abandoners

Why I don’t like this narrative

Emphasis on narrative.

The advice is solid. Attribution for SMS is largely overinflated though pre-purchase.

SMS and the welcome flows don’t matter as much as we think they do.

Just look at the hovering open rates and click through rates across your welcome series.

How many brands increase the discount towards the end of the welcome series?

To the data

After looking at close to 100k subscribers split tested and data cleansed I can tell you that SMS is a marker of intent, but does not actually lead to increased sales during a welcome flow time period.

In fact, in our studies, even asking for SMS, not even requiring it, correlated with reduced revenue by an observed 10% and a reduced AOV, with sign up rates being equal.

Something also observed as we do multi-step forms is that people that weren’t asked for a phone number completed 4-5 questions at a 25% greater frequency than when asked for a phone number.

Very few if any platforms track the subscription to conversion rate, nor do they track the time to conversion after a signup.

We track both.

The Welcome Series whether email or SMS will print money but not for the reasons that email and SMS agencies and companies tell you.

The most amount of money statistically is always on the first email or the first text.

Why? It’s the first look at a coupon code if someone didn’t copy it on the website. Hell some people still don’t put it on the website.

Here’s a real breakdown of the speed to purchase after signing up:

Lower priced AOV product that is repeat purchased often:

Higher priced AOV produce that is repeat purchased less often:

Lower priced AOV product that is repeat purchased fairly often:

Looking at this data it’s clear that 50% of people are purchasing in less than an hour. With 75% of people purchasing at max 3.8 days. The top company does 95% of all their transactions from subscribers within 13 days.

The bottom company does 90% within 14 days.

The middle company still sees 90% of all purchases from people that showed intent just over the 45 day max time period even for more considered and more expensive purchases. (that one isn’t optimized entirely, I know from looking behind the scenes)

So yes you should mint money, but it’s not your SMS flow or your email flow that’s minting it for you. In fact, it’s likely that if you didn’t close them on the first one, you’ve missed your golden opportunity.

So many of these emails are just “welcome to the family” style.

I love a good claim as much as the next guy but I’d never seen a chart like these before we got curious and started tracking and analyzing this data.

Want to know some more interesting stats?

Here’s the time between first and second purchases for the two lower priced repeat more likely brands:

Lower priced AOV product that is repeat purchased often:

Lower priced AOV product that is repeat purchased fairly often:

Note the similarities.

The bottom one’s data is a little skewed due to being sold out of styles and sizes during this time period, but it also does not run sales. It does not run aggressive email campaigns.

The top one is always in stock but runs consistent sales and has a very aggressive email campaign program.

Changes to offers, sales, etc, will influence all of these numbers especially for repeat time purchases.

The source of truth is probably somewhere between these two companies as a lot of companies will try to upsell after the first purchase. Skewing numbers again.

So if we split the difference we get:

25% of repeat purchasers purchase within 5-7 days

50% of repeat purchasers purchase within 14-25 days

75% of repeat purchasers purchase within 60-90 days

90% of repeat purchasers purchase within 150-180 days

In the top example of a company that is always in stock and runs an aggressive sales campaign we can see that 95% of all second purchases happen pretty darn close to 180 days.

For reference that sample size is north of 600,000 email signups in the last 180 days.

Looking at this data, it’s highly likely that your welcome series isn’t actually driving your revenue at all.

A comprehensive email strategy does increase the purchase frequency by about half by being top of mind, it’s unclear if this is sales driven or simply frequency.

Another key difference about these two is that the top one has a lower AOV on the second purchase than the first - discount driven for repeat purchases.

The bottom one has nearly double the AOV on the second order as the first.

So what does this all mean?

The bulk of all “welcome series” revenue is just actually making sure that you effectively get the coupon code in front of the person to be able to use it.

That’s it. Be top of mind as much as possible during that time period of the customer showing intent and looking to make a purchase.

Is it going to take more than 160 characters for them to make a purchase, tell them why they should do it in an email.

But know that it’s more about reminding them they were already thinking of making a purchase than it is the actual content.

Does the SMS welcome series matter?

Generally speaking, I wouldn’t collect SMS during the signup or leverage it during a welcome series, it serves as an intent marker more than anything else.

It’s a quick way to spend a lot of money on people that were already likely to convert and force them to leave your website to claim an offer.

As a customer journey person, that’s the last thing I would ever want someone to have to do.

I also can’t fight with the data, I’d rather have 10% more revenue and a higher AOV allowing people to stay on the website than collecting a phone number and losing out on other data that is more important to overall strategy that can actually be used to improve the website and customer journey.

That’s my take, if you disagree, have seen different results, let me know.

Happy Saturday!

-Jon