What The Customer Purchases Matters

Ok, back to our 18 month 1x buyer ... here, I filter for customers who bought only one (1) item in that first order, and then have not purchased for the subsequent eighteen months. I measure annual repurchase rates based on the Merchandise Category the customer bought from.

By Category:

• Category 00 = 1.85%.
• Category 01 = 4.14%.
• Category 02 = 3.50%.
• Category 03 = 3.92%.
• Category 04 = 4.63%.
• Category 05 = 5.43%.
• Category 06 = 4.84%.
• Category 07 = 6.41%.
• Category 08 = 2.13%.
• Category 09 = 2.49%.
• Category 10 = 3.47%.
• Category 11 = 3.15%.
• Category 12 = 3.02%.
• Category 13 = 2.13%.
• Category 14 = 3.06%.
• Category 15 = 3.37%.
• Category 16 = 2.99%.
  
• Category 17 = 4.76%.
• Average = 4.03%.

There are a handful of categories that cause problems ... 00, 08, 09, 13, 16. First-time buyers who purchase from those categories and then lapse tend to repurchase in the future at low rates. Going forward, you'd likely pay less for keywords associated with those categories, correct? You want fewer of those customers at a high cost, more of them at a low cost.

Ok, we've stacked a lot of wins on top of each other, with more to come.

Are You Ready For The Best Finding So Far?

Marketplaces (hint - Amazon).

So many of you have embraced Amazon.

Have you thought about the fact that Amazon is creating a reactivation problem for you?

Let's go back to our 18 month 1x buyers, evaluating annual repurchase rates. This time, we crosstab items purchased with channel responsible for the first order. Ready? This one is absolutely delicious.

Omnichannel!!  In this case, Amazon.

Look at the annual rebuy rates of 18 month 1x Marketplace buyers vs. other channels. It's a catastrophe. On an annual basis, the Marketplace buyer who purchased two items (not just one) has a 1.67% chance of purchasing again in the next year. Not next month ... next YEAR!

In other words, the Amazon buyer here is useless.

You are under no obligation to spend money marketing to unresponsive customers. In 2024, you have free choices, and free choices are there for a reason ... to allow you to have a relationship with unresponsive customers.

Meanwhile, look at the Paid Search / PLA customer row.  One-item buyers weren't very good ... but, if the customer bought multiple items, the customer has an above-average rebuy rate - with some credible numbers there. Same with Paid Social.

Almost all of you employ a vendor to help with cross-shopping opportunities. Here's a chance for your vendor to help you out ... demand that they amp-up their algorithms (#AI) to really dive into the Paid Search / PLA / Paid Social visitor who just put an item in the cart.  Ramp it up!

Naturally call center customers have better repurchase rates ... that's something I see in nearly every project. That's also something you can't really focus on going forward, it's mostly a 65-79 year old customer unless the customer needs assistance.

We keep stacking wins, readers!

• More items matters.
• Independent of items purchased, AOV size matters.
• Selling at full price matters.
• Selling best-selling items matters.
• Multiple categories (after controlling for items purchased) matters.
• Lapsed marketplace buyers are a CATASTROPHE!

Build a plan, based on what you've learned this week. How will you alter your communication/investment plan with lapsed customers?

And ... no ... this is not a plan.

Diversity of Product Matters Among Reactivation Candidates

In this example, I look at 18 month 1x customers who purchased exactly two items in their first order. Then, I layer on top of this analysis the number of merchandise categories the customer purchased from in a first order.

Ready for the results?

• 18 Month 1x Buyer with 2 Items in First Order, 1 Merchandise Category = 5.58% Rebuy Rate.
• 18 Month 1x Buyer with 2 Items in First Order, 2 Merchandise Categories = 7.28% Rebuy Rate.

Do you see what happened there?

The customer, even 18 months after buying, who purchased 1 item from 2 categories instead of 2 items from 1 category was 7.28/5.58 - 1 = 30% more likely to repurchase in the next year.

We just keep on stacking wins here.

• More items matters.
• Independent of items purchased, AOV size matters.
• Selling at full price matters.
• Selling best-selling items matters.
• Multiple categories (after controlling for items purchased) matters.

Dig into your database today (not tomorrow), create these attributes, and get busy identifying the reactivation candidates that you want to prioritize!

Look At This, Reactivation Lovers!

Have you ever noticed that a lot of songs from the 70s and 80s used the word "lovers"? Those were wild times, dear readers.

But I digress.

In order to reactivate customers at a credible level, you have to love the process. Lovers of the process stack attributes on top of each other. One of the attributes you stack is "winning items".

In this example, to make analysis easy to digest, I analyze 18 month 1x buyers who only purchased one item in the first order. Was that item a winner (graded as "A"), or a B / C / D / F item, where "F" items are items that only sold a few units during the year?

• Rebuy Rate for "A" Items = 4.55%.
• Rebuy Rate for "B" Items = 4.38%.
• Rebuy Rate for "C" Items = 3.99%.
• Rebuy Rate for "D" Items = 3.74%.
• Rebuy Rate for "F" Items = 2.99%.
• Average Rebuy Rate = 4.03%.

If the customer is a first-time buyer purchasing just one item, and that item is not a good-selling item, then the customer is less likely to purchase in the future ... even after 18 months passed.

Go look at customers who haven't purchased in a long time and segment the customers by the quality of the item purchased by the customer.

So far, we've stacked a lot of wins together.

• More items matters.
• Independent of items purchased, AOV size matters.
• Selling at full price matters.
• Selling best-selling items matters.

Dig into your customer database, collect the attributes I've described here, and do something positive with the information!!!

Reactivation Customers Buying Discounted Items

In most projects, I categorize items as selling at/above their historical average price point, or selling below their historical average price point. In other words, you have an item that sells for $50.00. Sometimes it is 10% off ($45), sometimes it is 40% off ($30.00). Average it all together, and the historical average price point for the item is $43.77. When it sells for $45.00 or $50.00, it is selling at/above the historical average price point for the item. When it is selling for $43.77, it is selling below the historical average price point for the item.

• $50 = Above
• $45 = Above
• $43.77 = Historical Average Price Point
• $30 = Below

We can evaluate reactivation candidates based on the fraction of items they purchase that sell below their historical average price point.

This is our 18 month 1x buyer segment. The number of items the customer purchased in a first order matters a lot. The share of demand from items selling below their historical average price point matters some. Look down the one item column.

• No Items Below Historical Average Price Point = 4.27% Annual Rebuy Rate.
• Item Is Below Historical Average Price Point = 3.36% Annual Rebuy Rate.

In other words, the customer that purchased an item at/above the historical average price point has a better chance of purchasing in the next year than the customer buying the item below the historical average price point.

Do you see what is happening here?

We are slowly stacking attributes on top of each other ... by doing so, we create a list of reactivation candidates who are actually more likely to, you know, reactivate!

Target Specific Reactivation Customers

Not all reactivation customers are created equally.

This week, we'll look at 18 month 1x reactivation customers ... they bought for the first time eighteen months ago and then just went dormant. This is the typical scenario ... by this point maybe 30% of the first-time buyers purchased again, while the other 70% are getting less and less like to buy again by the day.

Yes, you need to do something here.

No, you don't need to figure out the following.

Let's look at annual repurchase rates for this audience, segmented by initial average order value and initial items purchased in the first order. Tell me what you observe.

This segment of customers is not likely to purchase/reactivate in the next year, with only 5.16% of the segment doing so. This segment of customers is pretty close to dormant.

But some customers are less dormant than other ones.

Notice that customers who bought one item are less likely to reactivate ... about 4.03% reactivate vs. 6.04% of customers buying two items, vs. 7.64% of customers buying three items, vs. 8.45% of customers buying 4+ items in a first order.

Would you rather try to reactivate a customer with an 8.45% chance of reactivating, or a customer with a 4.03% chance of reactivating?

This is a pretty simple cross-tab ... one every one of you should be able to create from your customer data in about thirty seconds. Go run it today, tell me what you learn, ok?

There Isn't A Lot Of Time To Take Action

This is a VERY COMMON situation. On the x-axis, we have months since a first purchase. On the y-axis, we have the monthly conditional repurchase rate. In other words, let's say you acquire a customer, and the customer has a 3.6% chance of buying again in the first month with your brand. If the customer does not repurchase, the customer slips to a recency=2 month segment, and the monthly conditional (conditional on the fact that the customer has yet to buy again) rebuy rate slips to 1.9%.

There are three things worth pointing out in this graph (this is actual data, folks).

1. As mentioned a thousand times here, if the customer does not buy for a second time within three months, the customer quickly becomes inactive.
2. As mentioned a hundred times here, look at the peaks at months 12/24/36. This happens when a business has a strong seasonal component ... a seed business selling garden products in April ... a gift business selling stuff in November/December. Seasonal peaks are key reactivation triggers that almost nobody takes advantage of.
3. After twelve months, the customer is generally not coming back.

In this example, the first-time buyer only has a 15% chance of buying again in the next year, and only a 26% chance of buying again within four years. Hint - your business might look better than this, but not a lot better than this. Run this analysis - it's going to look similar.

Here's where marketers fail our businesses.

There should be four separate programs for these customers.

1. Welcome program for months 1/2/3 after a first purchase. If you don't convert the customer to a second purchase here, you've got problems.
2. Educational program for months 4/5/6/7/8/9/10 to keep the customer interested ... or as the kids say, "engaged". Here, you want the customer interacting with your content, given how unlikely the customer is to purchase again.
3. Anniversary programs for months 11/12/13/23/24/25/35/36/37. Push the product families the customer purchased in a first order, remind the customer of the anniversary event.
4. Low-Cost programs for all other months. Do not spend money on an unresponsive customer. Outside of email, social, and your website, do not spend money unless you can prove significant ROI. Use free content to "engage" the customer.

There isn't a lot of time to take action when you acquire a new customer. Reactivation is all about capitalizing on key windows when the fish are biting. Go fishing immediately, then be smart thereafter.

Here's The Situation

Show of hands ... how many of you think this is the marketing playbook for reactivating customers?

Here's the situation so many of us are dealing with.

Read across the "New During Year" row. New customers went from 45,000 in 2019 to 80,000 during the peak of COVID in 2020 to 65,000 in 2021. And then? Poof. Back to where counts were ... and as so many of you are experiencing, counts are down significantly in 2023 and so far in 2024. It's only going to get harder to acquire customers in the future.

New customers peaking in 2020, still high in 2021, back to normal in 2022.

Look at twelve-month buyers. Peaking in 2021, still high in 2022, back to normal in 2023. Notice the lag there?

Look at 13-24 month buyers, your most active pool of reactivation candidates. Normal counts in 2019, 2020, 2021. Peaks in 2021-2022 as the COVID customers lapse through the file. Back to normal in 2024. The lag is a year longer here.

And so it goes. Right now all of those COVID-buyers are in the process of lapsing out of your customer file. All of the sales potential of those customers was realized in 2020-2021, still contributing in 2022, then causing declines in 2023 and bigger declines in 2024. Compare counts of 25-36 month buyers, 37-48 month buyers, and 49-60 month buyers to 2019 counts ... these areas of the reactivation pool are still vibrantly ahead of pre-COVID levels ... for now.

Here's the situation ... there is another 1-2 years of mining these large reactivation pools before the COVID-buyers are gone. Once they're gone, woo-boy, is there trouble on the horizon. You're already feeling it. This is why we care about FORECASTING SEASON! Forecasting Season helps us see what is coming, so we can do something about it before it happens.

So yes, we're going to talk about reactivation.

And no, we're not going to talk about discounts and promotions ... because that is not how you reactivate a customer.

Reactivation

Remember three years ago when we talked (frequently) about the fact that the COVID-bump was ending and a reckoning was on the horizon?

As we approach Forecasting Season, the reckoning is at hand.

This is what you're telling me happened ... and the data generally backs up what you are saying.

I'll use the next several posts to discuss this table. There's just too much to share here in one post, and the content is too important to shove into just one post.

Notice a couple of things, keep these things in the back of your mind for upcoming posts.

1. The twelve-month buyer file (in this example) is smaller than it was in 2019. That's no bueno.
2. The reactivation audience is (for now) still bigger than it was in 2019.

Yes, "Reactivation" is about to become an important topic as we move into 2025. If new customers are hard to find, and if you sink a fortune into loyalty efforts that seldom generate incremental gains (for if they did generate incremental gains, your p&l would have been swimming in profit the past two years from all of the loyal buyers you generated from COVID four years ago - right?).

That last point (in parenthesis) is really important - nobody in the loyalty industry talks about it, and for good reason.

So if you struggle to generate loyal buyers ... and it is harder to find new customers ... might Reactivation be the next frontier for your efforts?

It's going to have to be the next frontier for your efforts.

Tuesday Notes

Let's try something this Fall ... send me links to articles you think are of interest to our community. If interesting, I'll publish them on Tuesdays during the Fall. If you don't see any articles published on Tuesdays, it means you aren't forwarding me anything. Also - nothing vapid ... our community doesn't need to learn "17 Tricks To Improve Facebook Advertising". Send something clever!

From one of our Readers:  Pottery Barn's Gus the Ghost. Two things here ... first, the limited run of an item (it doesn't have quite the utility at Thanksgiving, does it?) ... and second, look at how CNN is making money via affiliate links. Also notice the slight product upgrade (holding a small pumpkin) that enables repeat purchases. If you want your customers to become loyal customers and buy repeatedly, give them a reason to buy repeatedly.

From ... Me:  The Idaho Offense, the Complete Series. I did a little digging on the internet on Saturday when I noticed that Idaho was down three points to Oregon in the 4th quarter (as a 49 point underdog). Of interest was the fact that their offensive coordinator has a video series breaking down his offensive philosophy. You can have it for $129. It reminded me of how Mike Leach was willing to share his offensive philosophy with anybody who would listen (and for decades, nobody would listen ... now his stuff is everywhere and he's not with us anymore). It led me to a thought.

• If you had to create a $129 video series on the marketing philosophy of your brand, would your tactics and strategies be unique enough and compelling enough for anybody to pay $129 to learn it? Even more important, could you teach it ... could you break down your tactics and strategies (the unique ones, not the tepid ones like PLAs or affiliate links) in a manner worthy of sale?

Something For You To Read On Labor Day

Here - give this a read - it's about a Boomer music industry person visiting Harrods in London (click here).

This is where some of my readers start yelling at me ... "WE'RE NOT HARRODS, WE'RE JUST AN APPAREL BRAND FIGHING FOR OUR LIVES".

Maybe you are fighting for your life because there is absolutely nothing special about you.

"WE DON'T CATER TO RICH PEOPLE".

Fine.

Even Safeway can make you feel special. They'll put a few bottles of $89 wine out for you to look at ... some Quilt for $39 ... and then you'll buy the Chateau St. Michelle Merlot for $12.99 (in quantities of six). Yes, you get a discount ... if you buy SIX of 'em. They've got Alaskan King Crab legs, and sure, they're pricey (and yet they somehow sell through them before they spoil), so you look at them and then you buy the farm raised Atlantic Salmon, which, is not Faroe Island Salmon or Copper River King but if you don't overcook the salmon and keep it moist it will taste darn good.

What are the three or four things you sell that are unique and special, that make a customer feel good, that your competition chooses to not copy or cannot copy? You have special items. Yes, you do!! 

What are the 3-4 things you are famous for? Don't say quality, convenience, service, or prices. Or trust. Those are terms used in 1977. What are you famous for? And if you can't answer the question, you just gave your marketing executive a task for 2025, don't you think?

The omnichannel movement of the 2000s and 2010s was an unmitigated disaster. You had to have 25,000 stupid t-shirts, in all sizes, in all colors, in all channels, all at the same price, marketed to your core customer.  NONSENSE! 

Why not have one unique t-shirt and you only have 750 of 'em in a limited number of colors and when you sell out you sell out - it's over!?? Create some urgency!

Look at Headphones.com ... they actually tell you their best-selling items ... look how many are backordered or sold out! Of their top-twelve best-selling iems, five (5) of the twelve (12) are sold out or backordered.

Yeah, you better act, right now, if you want something. It might not be available tomorrow.

And the best-selling pair ... the Moondrop Variations for $520? They tell you on their website that you can get nearly identical sound from a knockoff for $159 ... they're actually steering you to another (much less expensive) item, the Truthear Nova (click here). They're dissuading you from buying the best-selling item and it is STILL the best-selling item. There's no need to discount something when you can pull that stunt off ...

... in other words, you can discount by framing $159 against $520 ... the customer perceives that the customer saved 69% and got 95% of the quality. 

That's really, really smart marketing.

There's still time to change your focus. You have three or four items you are known for. Why not make something out of that? You have categories where you have an item that sells for $79 and a comparable item that sells for $39. Why not pit the $39 item against the $79 item, allowing the customer to perceive that the customer is saving 50%?? Why are you eating margin on the $79 item at 50% off (today only, a Labor Day special) when you can steer the customer to the $39 item and then market the living daylights out of your $39 item?

You have so many opportunities to try things ... go try some stuff, ok??!!

P.S.:  A decade ago I gave talks at conferences, predicting you would become "media networks" or "sports teams" ... having hourly/daily programming that held the attention of customers, having monthly or quarterly "events" not unlike sports leagues that you'd build up to with your hourly/daily programming. As it turned out, that concept became "community" and "content". A 40% off email campaign does not fit into this world. Use your content to speak to your community ... frame your products ($39 vs. $79), create events, build excitement, give your audience a reason to care.