The hottest topic? AI.

I mean, this was always going to be the case, right? From day one the increasingly-prevalent role of artificial intelligence (AI) in retail was the talk of the event, with queues for AI-related talks and presentations often stretching across the entirety of Olympia’s top floor. According to Kate Hardcastle MBE, the award-winning consumer insight expert who opened this year’s show, AI was “the most requested topic from the RTS advisory board” demonstrating how front-of-mind the technology is for retailers at the moment.

Some common threads on AI were focused around security and data transfer, however I did find that many of the discussions were very top-level; there were very few talks that offered practical advice on how to adopt AI in your business or the key use cases you should be getting started with.

“#AI is the most requested topic from the RTS advisory board.”

Consumer expert @KateHardcastle MBE opening #RTS2024 pic.twitter.com/Tq8dNnlHtn

— Peak (@Peak_HQ) April 24, 2024

Enhancing, not replacing

We’ve always held the belief at Peak that AI should be used by businesses to empower teams, not displace them, and it was great to see this theme carried forward by others at this year’s show. To quote Dex Hunter-Torricke, Head of Global Communications & Marketing at Google DeepMind, “AI tools can be transformative, however the human touch is just as critical — it’s a harmony.”

We heard from a few different attendees that all claimed they were looking to leverage AI as a way of saving time and resources — spending fewer hours with their heads buried in spreadsheets to drive more autonomy in their day-to-day roles and free up time for more creative thinking and larger, more complex tasks. In short, AI is being seen more and more as something that is there to protect teams whilst also protecting the business’ bottom line.

What about generative AI?

While there was significant talk of generative AI and its capabilities in talks and presentations, there were very few software vendors who were pushing this as a solution. While there’s still unrivaled levels of buzz around generative AI since the explosion of tools like ChatGPT and Midjourney in recent years, it would appear businesses aren’t yet utilizing these capabilities in the most beneficial way. We had a number of positive conversations around our own generative AI offering, Co:Driver, further demonstrating its position as a unique offering in today’s market. More on that here

Reimagining customer experience

In previous years, talk of enhancing the customer experience at these sorts of conferences usually revolved around martech tools and improving customer personalization. However, this year’s Retail Technology Show veered away from these traditional topics, instead focusing on ways of providing a better customer experience by enhancing more traditional back-office operations. There was a clear emphasis on optimizing pricing and ensuring product availability, with having the stock you need to avoid disappointing customers a key focus for many businesses. At the end of the day, customer experience starts in the warehouse — if the product isn’t available or priced correctly no amount of customer service will correct this.

Embracing marginal gains

I personally noticed more of a shift towards celebrating smaller victories and incremental wins at this year’s show. While, in the past, some vendors may have focused on shouting about huge success stories and almost-unbelievable percentage gains, this year felt different — a more nuanced approach to retail success and a departure from big-bang metrics being front and center. From a price optimization perspective, for example, Philippe Debello, UK Data Science Director at Decathlon, said that “a pricing project is not a big bang — you need to concentrate on the small wins that amount to the big wins and celebrate these.”

Looking ahead

As the curtains fell on this year’s Retail Technology Show, one thing was clear: AI is continuing to shape the future of this ever-evolving industry. By embracing AI solutions and viewing them as a partner in progress, not a replacement, retailers can unlock new levels of efficiency and customer-centricity to deal with the ongoing volatility and uncertainty still clouding the sector. To quote Zaki Hassan, General Manager at Aptos Retail, “in today’s volatile world, there is an innate need to bring some form of predictability into your business” — and AI, of course, is crucial to providing that.

Thanks for having us, Retail Technology Show, it was a blast. I’ll sign off by showing what else we got up to at the Olympia as we caught up with a number of different retail vendors in search of free swag and some fun and games — check out our thread on X below!

Location: @RetailTechShow
Mission: Have the most fun

Let’s get stuck in…#RTS2024 pic.twitter.com/EGi46AQnW6

— Peak (@Peak_HQ) April 25, 2024

What’s your business about?

A bad customer is one that gets in the way of us reaching our targets. So if you want to find your bad customers, you’ll need to know what our business targets are and how you’re tracking against them. These business targets should be specific and measurable.

For example, you might look at things like: 

• Revenue
• Transactions
• Profits
• Repeat purchase rate
• Average order value
• Cost per acquisition 
• Customer acquisition cost
• Lifetime value
• Net promoter score
• Return rate

Let’s begin our search for bad customers. We’ll start with an easy target, like profitability. Let’s say your business has a profitability margin of 40%. To find your bad customers, you’ll need to look for customers who are getting in the way of delivering this target. Here’s how. 

Bad customer #1: Fails to Pay Philippa

This is Fails to Pay Philippa. Philippa loves shopping, but hates paying — and Philippa’s not alone. She’s part of a whole segment of customers who are nowhere to be seen when the bill comes. No business wants Philippa and her spendthrift friends as a customer. 

Fails to Pay Philippa has got what she wants and she’s on the move.

Thankfully, customers like Philippa are pretty easy to find. Every business has a list of customers with a credit account, one that flags whether or not payments for their orders have been settled. So, you know where to find them. But what should you do with them? 

For starters, I wouldn’t suggest offering Philippa credit in the future. Make sure Philippa pays up front (she’s shown she can’t be trusted with a credit line). So, bad customer found and dealt with. Easy. In this case, Philippa is a straightforwardly bad customer, one that no business wants. But not all cases are this cut and dry…

Bad customer #2: Coupon Karen

Let’s look at another example, those customers hooked on discounts and coupons. Customers like Coupon Karen.

Coupon Karen smells a discount, and she’s not buying until she gets one.

Karen is keen to buy, but she’s a master negotiator. You’ll have to budge on price before she shows you the money. Karen’s highly engaged, regularly searching through your range, saving items for later and scouring every one of your marketing channels for sales or price-cutting coupon codes.

You’ve been sending her great content for weeks, but it seems like she just won’t buy. It’s clear Karen’s not blinking first. But then, as soon as it’s sale season, Karen whips out her credit card and puts her plastic to work.

Finally! You’ve made a sale and Karen’s bagged her discount — everyone’s happy, right? Maybe not. Customers like Karen have a lower lifetime value and increase your cost per sale because they need so much persuasion to purchase. 

But does it really matter? Money’s money, right? Unfortunately, it’s not that simple. Relying on discounts and sales has consequences for customers like Karen. You see, she hasn’t always been this cut-throat about cut-prices. 

You created the discount diva that Karen is today. The discounts you’ve been using to trigger a purchase from Karen got her hooked on the thrill of thrift. It’s changed her perception of your brand. Before, she’d be happy to pay RRP if she saw the right product — but now she feels ripped off unless she’s getting a discount. 

Bad customer #3: Randy the Returner

Next up, meet Randy the Returner. Randy loves shopping with you, but he returns 90% of everything he orders.

Randy the Returner can’t wait to clog up your supply chain with returns.

Let’s say you’re a fast fashion retailer that relies on a high volume of low margin purchases. It may cost you more to administer their returns than you bag in profits for the few items your customers keep.  

But, what if you’re a high-end fashion retailer that targets affluent customers? You’ll rely on fewer transactions at a high margin — and the profits earned from the high margin items those customers keep may far exceed any administrative cost of return? 

This is why knowing your business inside out is so important. A certain customer might not be ideal in the long term, but they might give your business the boost (in this case, cash flow) it needs right now. 

The key here is to know what your business needs now, and to understand which customers are delivering or distracting from it.

Segmentation: the bad customer search engine

People look at segmentation as a tool for marketing personalization and they’re right, it’s key to nailing personalization. But segmentation can do so much more than enable highly-targeted, personalized emails and ads. It can help you find your best and worst customers.

Segmentation tools that use standard approaches, which look only at predefined metrics (e.g. number of purchases or age) can’t quite get you there. Segmentation needs to incorporate a broad range of behavioral, transactional and historical attributes to help you find (and fix!) your bad customers. 

As an example, let’s catch up with Randy the Returner. We said different businesses might deal with him differently. But not every returner is created equal. Some will have a higher lifetime value and cost more to initially acquire than others.

That might mean losses incurred from their return frequency are outweighed by the value of retaining them as a customer. In other words, the short-term losses incurred by their bad customer behavior might be tolerable now, if they could be highly profitable in the long term. 

How to banish bad customers 

Why do we need to look for bad customers? There are two reasons; either we want to turn them into a good customer, or we want to prevent them from interacting with us in the future. How we deal with customers needs to be as data-driven as the approach we use to find them. 

Sometimes it can be our processes or policies that are actually making customers bad. Let’s return to Coupon Karen. Why does she need a discount before buying? 

The first place you could look is churn risk. Are you applying a single churn risk threshold across your entire customer base? If you are, you could be creating your own Coupon Karens. Churn risk should be based on a specific customer segment. You don’t need to act if a customer’s not made a purchase after an arbitrary period. Instead, you need to act when they fail to purchase within their usual buying pattern. 

For instance, Karen might’ve previously only made a purchase from you every six months.. But you’ve got a blanket business rule that says anyone who’s not purchased in three months is a churn risk. 

So you make sure Karen’s first in line for a coupon code, and it works — she purchases! But she might’ve made the same purchase at RRP only three months later. You’ve repeated this cycle and, eventually, she’s come to expect discounts, and she won’t buy without them. 

How about Randy the Returner — why is he returning so many items?

If we create a segment of customers who purchase specific SKUs across multiple sizes, we might find that the true cause of their returning habit is inconsistent sizing practices between the brands you stock. 

Without this insight, you might decide the best way to deal with the problem is to charge your customers for returns, or penalize customers who return items frequently. But this may cause valuable customers to buy elsewhere.

If we know size variability is causing returns, we can try another strategy. ASOS is a great example of this problem. In 2018, ASOS was experiencing a meteoric rise. But with the fashion market exerting downward pressure on price, its profitability was taking a hit.

One factor impacting ASOS’ margins was the high rate of returns. After looking for answers in the data, ASOS discovered size variability was generating legitimate customer returns; customers often found their usual sizes were either too large or small, depending on which brand they bought.

To tackle the issue at its root, ASOS launched Fit Assistant. Fit Assistant uses artificial intelligence (AI) to learn customer size using the customer’s previous retained orders as data. It then advises customers on what size will fit them in each brand.

‘Fit Assistant’ from ASOS

It was a huge win for ASOS and its customers. Customers were more likely to get their size right the first time, meaning less frustration for customers and fewer costly returns for ASOS. 

The truth is that there are always going to be tradeoffs when it comes to business decisions. The key is to tip the balance of those tradeoffs in your favor. It’s important not to just understand who your bad customers are, but why they’re a bad customer. You might be surprised at what’s really driving bad customer behavior.

Beat bad customers with game-changing AI segmentation

So how do you find and fix bad customers? The answer is in your data, and the most efficient way of looking at your customer data is through segmentation. But often segmentation tools are built into existing marketing platforms (e.g. social media or email marketing platforms). To truly understand customer behavior and take strategic decisions, you need to see their actions across the whole customer journey. 

Peak’s application, Audiences, can combine customer data from several sources or directly from your customer data platform (CDP), to give comprehensive segments across your customers’ entire journey. 

It uses AI to give you highly accurate customer segments based on not just historical data, but AI-driven predictions of their future behavior. 

If you want to learn more about how you can get started, you can read about the concept of ‘headless segmentation’ here. Or book a call below, and see how Peak can help you get started. 

What is customer segmentation?

Customer segmentation is the process of creating cohorts of customers into groups of similar commonalities such as demographics, product preferences and value to a company.

For many businesses, this is considered a basic principle of marketing, along with the likes of email automation and personalized targeting. The question we often get asked is that, by its nature, segmentation reduces the size of the pool of potential customers who can be targeted by you — so why would you do this?

Yes, segmentation does limit the number of customers you reach out to, however what it does do is increase relevance. By finding commonalities between your customers, you’re unearthing key connections, mutual interests or shared needs.

For example, let’s say we want to send a birthday campaign to our customers who have a birthday that falls in November. We’d create a segment of ‘Birthday Month = November’, and the campaign would then include some birthday-related content along the lines of: “It’s your birthday this month, {{Name}} — wishing you the best birthday yet!”

Now, imagine that same message being sent without customer segmentation. My birthday falls in September, so do I want to receive this message in November? Absolutely not! What does that do to me as a customer? It makes me highly disengaged with the brand and unlikely to engage with them going forward. It highlights that the brand doesn’t see me as an individual but instead as a mechanism to try and attain revenue.

What’s my likely next step? I unsubscribe from their communications, and no longer want to be contacted by them no matter how much I like their products or experiences; they’ve essentially switched me off as a customer. 

Whilst this is a very high level, tongue in cheek example the principle remains the same when it’s applied to more complex scenarios. Let’s say I’m browsing on a website, add a pair of sneakers to my basket, but I’m too indecisive so end up closing the browser. Within thirty minutes, I get major FOMO and decide, “no, I actually really need those sneakers!” so I log back in to purchase. 

The next day, I get an automated abandoned basket email: “Don’t forget to complete your purchase! Here’s an additional 10% off if you purchase today!” In this scenario, once again I’m left feeling disgruntled and irritated. Plus, all you’re doing is highlighting to a customer that if they wait to make their purchase, the likelihood is that they’ll receive a discount code. This gives the customer the ability to repurchase the item at a lower price — eroding margin unnecessarily and, additionally, causing an item return. All parties lose in this situation, so it really is important to segment, and segment correctly. So, how do you do this?

How to segment customers

You need to make sure that you find meaningful reasons to contact your customers. Step away from the ‘spray and pray’ approach to marketing communications and invest your time and efforts into creating and building relationships with your customers, not just transactions. Identify and predict what they want, when they want it.

An example of this is building a propensity to purchase artificial intelligence (AI)-driven model. You’re talking to customers who are predicted to be ready to make a purchase, so it’s on you to make sure you understand what they’re interested in and what their preferred channel is. Do this by employing an AI recommender in each communication and touchpoint, enabling you to provide personalized messages at scale and build a model to determine their channel preferences.

Identify your customers’ previous purchases and behaviors, and use these to map future communications. In the above example, where I purchase sneakers, a good example of segmentation could be a follow-up communication introducing sneaker protection spray: “keep them looking white for longer!” 

The segments to create this would be as simple as “purchased sneakers yesterday + did not purchase sneaker protection spray.” This nice touch makes a customer feel fully recognized, and ultimately contributes to a better experience with your brand — and leads to additional sales!

Finally, consider using site metrics to gauge a customer’s interest. Customers who’ve visited your site within a specific time frame indicates potential for sales. For example, pulling a segment of “has visited the site in the last three months” shows a level of interest in what you’re offering — even if they haven’t purchased anything yet. Perhaps your next promotion could be the final push they need to convert?

What are the key customer segmentation metrics?

The traditional metrics that matter when it comes to customer segmentation — those that you’d consider when determining whether or not your segmentation strategy is working — would include basic engagement metrics such as opens, clicks and conversions.

However, truly personalized segmentation needs to take this one step further. One of the most critical metrics to look at is lifetime value (LTV). High LTV customers are those customers that are predicted to be worth the most to you throughout the course of their time with you. Typically, the happier the customer, the higher the LTV. Equally, measuring much bigger strategic key performance indicators (KPIs) such as net promoter score (NPS) will be crucial when gauging whether your customer segmentation is hitting the mark!

How do teams know who to target, when to reach them, what products they’re interested in and their preferred channel?

Many businesses are already well aware of the benefits of effective, data-driven customer segmentation. For example, segmentation makes firms 60% more likely to understand customers’ challenges and concerns and 130% more likely to know their intentions. 

Here at Peak, we’re taking customer segmentation one step further. We’re doing this by pioneering a new marketing approach that we’re calling headless segmentation. What do we mean by headless segmentation? Let’s start with the easy bit: segmentation.

What is customer segmentation?

Customer segmentation is a fundamental principle of marketing. It is the process of creating cohorts of customers, divided into groups of similar commonalities such as demographics, product preferences or value to a company. So, how is headless segmentation any different?

What is headless segmentation?

When we say headless segmentation, we mean that the segmentation isn’t tied to a specific system or channel. The segmentation can therefore be used across multiple platforms and systems, to provide a consistent journey to a customer.

With headless segmentation you can join up previously-siloed sources and create new granular customer segments that give you a clearer view of the likes, dislikes and buying patterns of each individual. Gone are the days of looking at customers and their behaviors in an isolated view. To win in the modern era, marketing teams operating in every industry must focus on creating a seamless experience for their customers, across all available touchpoints — from email marketing and onsite recommendations to social media ads and the in-store experience.

The only way to effectively do this is to employ an effective headless customer segmentation strategy. By taking the steps required to remove channel bias and data silos, and looking at meaningful cohorts driven by key and integral data points, you can ensure that you’re always serving your customers in the best way possible.

Who’s responsible for headless customer segmentation?

So, who needs to drive a headless segmentation strategy? In short, everyone within your organization, not just your marketing team!

Headless Segmentation breaks down the data silos that exist between your channels and your business-wide systems. Therefore, the only way to deploy this approach successfully is for all the teams within your organization to work collaboratively, with the same common goal in mind — and to focus on managing and surpassing a customer’s expectations at every available opportunity, across all touchpoints.

Why is headless segmentation important?

Gone are the days of looking at customers and their behaviors in an isolated view. To win in the modern era, marketing teams operating in every industry must focus on creating a seamless experience for their customers, across all available touchpoints — from email marketing and onsite recommendations to social media ads and the in-store experience.

The only way to effectively do this is to employ an effective headless customer segmentation strategy. By taking the steps required to remove channel bias and data silos, and looking at meaningful cohorts driven by key and integral data points, you can ensure that you’re always serving your customers in the best way possible.

What problems can you solve with headless customer segmentation?

Peak’s Audiences application enables marketing teams to enhance their segmentation strategies with artificial intelligence (AI). It gives users the power of headless segmentation at their fingertips, creating sophisticated customer segments instantly, leveraging data taken from across multiple platforms and sources.

Let’s take a closer look at some of the most powerful use cases that Audiences is being utilized for by our customers.

How to increase customer lifetime value with headless segmentation

Headless segmentation can be used to build a picture of the key commonalities and attributes most prevalent to a specific customer segment. Additional parameters can be layered on, such as in-market likelihood, to understand not only what a customer is worth to a business, but also when they’re likely to purchase, what they’re most likely to be receptive to and via which channel. This enables you to serve the correct message via the correct channel, with the correct product, at the correct time.

How segmentation can reduce cost per acquisition

Acquiring customers is harder and more expensive than ever, with cost per acquisition increasing by an eye-watering 222% from 2013 to 2023. Combined with continued budget costs, it’s never been so vital to acquire more of the right kind of customers. 

By employing a headless segmentation strategy, you can leverage your data to create bespoke segmentation to know more about your customers than ever before. By looking at common connections between customers, you can then identify key data points to understand a business’ most valuable, profitable and engaged customers, enabling teams to drive the biggest return on investment from any campaign.

How to gain a unified view of your customers

72% of businesses describe managing data silos across multiple systems as moderately to extremely challenging. Centralizing data is the most critical step in rectifying this, and creates a highly organized way to access data and facilitate better decision making. It enables the data to be constructed in a way that creates a single source of truth for all MarTech solutions to run from. It also ensures scalability for the future, as new channels and data systems are inevitably added to the MarTech stack.

Audiences has the ability to create bespoke segmentation at an individual customer level. A key feature of this is the deployment of segment maps. Segment maps enables end users to build a grid of customer segments based on key attributes and data points, inclusive of AI predictions. Within a few clicks, you can generate highly-strategic customer segments — based on data points, KPIs, business rules and guardrails — that are entirely bespoke to that unique business and its requirements.

What is the perfect tech stack to allow for headless segmentation?

Interested in customer segmentation and marketing? Join us for a webinar on headless segmentation!

I’m super excited to be hosting a new webinar, focused around this exciting new concept of headless segmentation.

In this webinar I’ll talk to you about how artificial intelligence (AI), when applied in the right way, is transforming marketing teams — enabling scale, automation and, ultimately, personalization for your customers like never before. You can watch it on demand here.

Once you’ve got your segments set up, there are some additional considerations to take your Black Friday event to the next level.

Create a slick checkout experience! Look at payment methods to make sure your customer can check out easily and quickly. Whilst Black Friday falls on a pay day for many, some customers see “buy now pay later” methods as more valuable, as they take away an element of financial risk. Last year, Klarna saw a 141% year-over-year increase in sales, as millions of shoppers chose flexible payment solutions. Your checkout also presents an opportunity to increase basket size with final upsells like tote bags, gift wrapping and gift cards – make sure these are easy to add.

Manage customer expectations for delivery. Ensure robust delivery methods are available to create a positive experience for your customers. Understand which methods can’t be supported and remove them as an option for the customer. This is often the case for next day delivery, which can cause chaos for warehousing and logistics teams, results in unhappy customers when expectations aren’t met, and requires additional support from customer service teams to deal with complaints. Instead, take the opportunity to grow online-offline relationships during the peak season by offering delivery to store, or tempt customers with discounted postage for longer delivery windows.

Take the online-offline connection a step further. Just because a customer didn’t shop online with you, doesn’t mean they didn’t shop with you at all! Harness data from stores to build a true single customer view and create personalized messages that enhance their experience with you. For example, if a customer only shops in store, serving them messages around next day delivery may not resonate with them, instead consider click and collect options.

Consider your automated communications. Customers receive a lot of communications throughout this time – it’s fair to say the standard contact frequency rules go out the window! Figure out your strategy and align your communications to it. Switching off non-essential communications, like birthday and anniversary messages will streamline your focus to Black Friday promotions only and avoid double discounts. Some automated communications, such as abandon basket emails, can have a huge impact on Black Friday revenue – make sure they include an enticing prompt to check back in. Try something like, “didn’t quite find what you were looking for today? Check back in with us tomorrow as new deals land daily!”.

Don’t sit back when Black Friday is over! Put the reams of new data you’ve collected to good use. In particular, keep a close eye on the new versus returning customer mix as it allows you to track revenue brought in by new customers against revenue from your existing base. A higher new customer mix suggests that you’re benefitting from a powerful acquisition strategy. Prioritize retention strategies to keep these customers returning after the peak trading season has ended.

If you’re looking for a helping hand when it comes to taking your Black Friday strategy to the next level, look no further than Peak. Our platform, applications and services help retailers like Nike, KFC and Pretty Little Thing to harness the potential of AI.

Peak provides retailers with the opportunity to bring together different sources of data to power personalized, highly targeted experiences for customers, at scale, allowing your team to go from data, to real-time insight, to great commercial decisions.

Why AI? Because AI is the most powerful way to understand and act on the myriad of signals provided by each individual customer. When used correctly, AI can identify patterns and trends hidden in historical data to improve the customer experience.

More importantly, it can use that insight to predict a customer’s needs or behavior and enable the retailer to tweak the experience for both customer and commercial benefit.

However, there are very few retailers who have the ability to view, manage and harness their data in this way. Most retail or consumer businesses we meet are struggling with data silos, or systems of record that don’t talk to each other. This results in personalization being executed via a single channel – if at all.

Introducing AI into the marketing tech stack will change that.

What’s the most common approach?

For many retailers, the current approach to assembling a martech stack is clunky and suboptimal. Quite often each piece of marketing technology has its own data set, usually linked to the channels that that piece of technology is aimed at.

While this is great for looking at individual channel performance, it risks creating silos in a business – especially from a data perspective.

When data is contained within a single system (siloed), it’s isolated from the rest of the business, and it’s only ever accessed by the end user of that system. If this data is then used to make decisions, the end users lack the full picture needed to make the right judgment.

For example, in the context of customer acquisition, it is vital that every data source is used to establish high value customers. For instance if returns data is not ingested into customer profiling, the result could be a very skewed view on which customers are most valuable to a business – if only transactional data is considered, we could deem customers with a large basket value and/or frequent purchase to be customers we want to attract more of.

However if all/a high number of those items purchased are returned, not only are those customers potentially costing the business money, they are not the customers you would want to attract more of. Looking at all customer data touch points ensures that customers acquired are highly profitable, active and engaged customers. 

For example, a retailer’s marketing team sees a surge in engagement with their latest product post on Instagram. So, they promote the product further, including it in boosted posts and on their social stories.

However, what the team doesn’t consider is the number of units left in stock for that item – as they have no visibility of this information. This item sold really quickly, and now only has fragmented sizing left – so, when the majority of customers follow the link to purchase, their size is not in stock, causing huge bounce rates. 

You cannot afford to underestimate the impact that such an experience can have on a retailer’s customers. In an age where consumers are increasingly expectant of personalization, with 83% happy to share their data for a more personalized experience, if products are recommended that are unavailable then this will inevitably disappoint the customer. The breakdown in both data and cross-functional working here has resulted in a large number of customers having a poor experience and, potentially, will make them unlikely to shop again. The time to change this is now.

What’s also often forgotten about is the impact data silos can have on internal teams. Data silos lead to culture silos – and most retailers want fully cross-functional teams, all working towards the same common goal. In order to do this, there must be a single source of truth to dictate success. If this isn’t established, then teams are working in vain to try and achieve a truly omnichannel approach.

The solution is a connected approach that unites data from across the entire customer journey. By analyzing every customer data touch point together, retailers gain a complete view of behavior, this can be leveraged to increase both customer experience and commercial performance. It’s an approach that will become commonplace, not least because it offers a significant competitive advantage.

To achieve it, marketers need three core things:

1. Data unity: All data in one place, so every channel and system can use every single datapoint, both first and third party.
2. Predictive insight: Identifying patterns and trends that can inform decision making against key strategic objectives.
3. Feedback loop: Ensuring continuous learning, the performance of each action being measured and systems are iterated and improved.

Or, to put it another way…

Centralized data + artificial intelligence + execution platform

1. Centralized Data – Snowflake

A powerful data warehouse, Snowflake enables users to organize data in a structure that is most meaningful to their business. It centralizes data, creating a repository from which all MarTech solutions can run – both now and in the future.

The Snowflake Data Marketplace also provides access to third-party data, allowing users to leverage insight from the wider market as well as their businesses to enhance MarTech performance.

2. AI platform – Peak

On Peak, users can rapidly deploy multiple AI applications from a single platform, eliminating the need to leverage a complex web of individual solutions. The platform can be used by both technical and non-technical teams, giving decision makers access to a single, predictive view of their customers.

Filters and segments are fully customizable, so users can leverage the platform and its apps to deliver on the objectives that matter most to their business, for example, identifying when customers are in-market to purchase, and the specific items they’re likely to buy.

3. Execution platform – Braze

Braze is a comprehensive customer engagement platform that powers relevant and memorable experiences between consumers and the brands they love. Context underpins every Braze interaction, helping brands foster human connection with consumers through interactive conversations across channels that deliver value quickly and continuously.

The result?

A best-in-class functionality and user experience, for a fraction of the cost and complexity involved in self-building or leveraging multiple systems. Not to mention the elimination of silos across your business and increased customer loyalty. 

In this new age of changing customer expectations, the agility and personalization offered by a simplified MarTech stack will separate the winners from the losers.

Once you’ve got your segments set up, there are some additional considerations to take your Black Friday event to the next level.

Create a slick checkout experience! Look at payment methods to make sure your customer can check out easily and quickly. Whilst Black Friday falls on a pay day for many, some customers see “buy now pay later” methods as more valuable, as they take away an element of financial risk. Last year, Klarna saw a 141% year-over-year increase in sales, as millions of shoppers chose flexible payment solutions. Your checkout also presents an opportunity to increase basket size with final upsells like tote bags, gift wrapping and gift cards – make sure these are easy to add.

Manage customer expectations for delivery. Ensure robust delivery methods are available to create a positive experience for your customers. Understand which methods can’t be supported and remove them as an option for the customer. This is often the case for next day delivery, which can cause chaos for warehousing and logistics teams, results in unhappy customers when expectations aren’t met, and requires additional support from customer service teams to deal with complaints. Instead, take the opportunity to grow online-offline relationships during the peak season by offering delivery to store, or tempt customers with discounted postage for longer delivery windows.

Take the online-offline connection a step further. Just because a customer didn’t shop online with you, doesn’t mean they didn’t shop with you at all! Harness data from stores to build a true single customer view and create personalized messages that enhance their experience with you. For example, if a customer only shops in store, serving them messages around next day delivery may not resonate with them, instead consider click and collect options.

Consider your automated communications. Customers receive a lot of communications throughout this time – it’s fair to say the standard contact frequency rules go out the window! Figure out your strategy and align your communications to it. Switching off non-essential communications, like birthday and anniversary messages will streamline your focus to Black Friday promotions only and avoid double discounts. Some automated communications, such as abandon basket emails, can have a huge impact on Black Friday revenue – make sure they include an enticing prompt to check back in. Try something like, “didn’t quite find what you were looking for today? Check back in with us tomorrow as new deals land daily!”.

Don’t sit back when Black Friday is over! Put the reams of new data you’ve collected to good use. In particular, keep a close eye on the new versus returning customer mix as it allows you to track revenue brought in by new customers against revenue from your existing base. A higher new customer mix suggests that you’re benefitting from a powerful acquisition strategy. Prioritize retention strategies to keep these customers returning after the peak trading season has ended.

If you’re looking for a helping hand when it comes to taking your Black Friday strategy to the next level, look no further than Peak. Our platform, applications and services help retailers like Nike, KFC and Pretty Little Thing to harness the potential of AI.

Peak provides retailers with the opportunity to bring together different sources of data to power personalized, highly targeted experiences for customers, at scale, allowing your team to go from data, to real-time insight, to great commercial decisions.

If you’re looking for a helping hand when it comes to implementing any of our Black Friday top tips, look no further than Decision Intelligence from Peak. Decision Intelligence is the commercial application of AI to the decision making process. It is outcome focused and must deliver on commercial objectives, with the unique ability to leverage data from across an entire business, not just siloed functions. This gives retailers a truly holistic view of what’s happening across the board. 

With all of your data in one place, AI is then applied to spot patterns, trends and future possibilities; allowing teams to go from data, to real-time insight, to a great commercial decision. From a personalization perspective, for instance, Decision Intelligence enables you to power great human interactions – driven by AI – to forge closer relationships with customers and increase brand loyalty.