Salesforce Connections 2024 – “Trade should not be an island”

As I write this, Salesforce Connections 2024 is already halfway through, but we can of course say that this is the year of Copilot.

With Commerce Cloud Copilot scheduled for general availability in June 2024 and Copilot GA Marketing Cloud in July 2024, you couldn’t go too far into the McCormick Conference Center without hearing how these Copilots will change your daily life. Day.

However, I see a different theme in this year’s exhibition. But for now, what do customers think? First, the press reports:

There were tons of keynotes on Wednesday for various products. Needless to say, I have a lot of annoying questions. I’m focusing on Commerce Cloud news here because I spoke with Alex Bucher, vice president of product management. First, details about Commerce Cloud:

Payment in Salesforce: “Salesforce Checkout, a unified one-click checkout platform connected to customer data, is built on the Einstein 1 platform to provide shoppers with a more consistent experience across multiple channels throughout their purchasing journey”
Headless Commerce Opportunities for B2B: “Developers can implement headless commerce using pre-built APIs in Commerce Cloud that help sellers create engaging B2B sites that compete with traditional consumer shopping experiences“
Composable Commerce Improvements for B2C: “Developers can combine or create templated, composable and customizable, headless approaches to building B2C e-commerce sites with new, powerful Salesforce APIs and ready-to-use personalization features“

Corporate retail experiences – are brands getting closer?

I see a theme running through this message: whether it’s B2B or B2C digital commerce, it’s time to meet/exceed the high bar Amazon has set for direct-to-consumer personalization and logistics, but with enterprise-grade solutions and hopefully human touch when needed – an area where brands can (and should) strive to do better than Amazon. Oh, and these enterprise solutions should also be available to smaller clients with lean teams. Being “headless,” they should integrate with Amazon, Shopify, SMS, and more.

But this is my view. So I asked Bucher: What is his message for those who are not here in Chicago? As he told me:

What I hope people take away from today’s experience is that it’s really a combination of data, AI, marketing and commerce all really coming together. I know this sounds a bit like marketing speak, but it actually builds this end-to-end journey. When you think about commerce, it’s part of the deal, the bottom of the funnel – but you’re implementing so many elements of the full customer lifecycle. And that’s where a lot of our innovations are taking us.

Commerce Cloud Copilot isn’t GA yet – but we still need to talk about the results. After all, Salesforce has been developing AI-related platform improvements for 12 years. So we shouldn’t wait to talk about digital commerce success. During the Commerce Cloud keynote, Salesforce cited the example of Siemens Customer 360: >23% win rate and a twofold increase in the number of cases resolved the first time. Transitioning from legacy systems pays off: according to Salesforce Commerce metrics “customers saw a 29% increase in digital revenue as a result of the transition from legacy channels”

Key Customer Demo – Einstein Copilot for buyers in action

Salesforce did something different during the Commerce Cloud keynote: the customer provided context for the demo. Jenna Flateman Posner, COO of Solo Brands (pictured above), discussed a scenario with Einstein Copilot for Shoppers features: personalizing the shopping interaction on WhatsApp.

The next challenge for this demo? Showing how personalization can work when you don’t have your own “logged in” customer data. This demo involved clicking on a personalized ad for the Solo Brands Portable Mason Oven on Instagram. Clicking on the Solo Brands website displays a 3D version of the furnace. However, the demo buyer is distracted and does not add the item to the cart. But that’s not the end. As Salesforce explains:

However, some time later, I receive a notification on my phone from WhatsApp – it’s the Solo bot, powered by Copilot for Shoppers, which re-engages… Solo was able to recognize who I was, even though I was shopping as a guest and never logged in. They are able to piece together my purchase history, my previous interactions with their website, and my engagement on Instagram. And on top of that, since I’m a regular customer, they offered me a 20% discount.

But will this stove fit in my yard? We ask the Co-Pilot:

We take a photo of the patio and send it to Copilot, and Copilot provides the product specifications. He also recommends Solo Adirondack chairs to help complete the set and sends a demo photo of the chairs and stove on the back patio:

So we ask Copilot how much it will cost and we get an answer with a ready-made card and a promotion applied. In this case, the order is processed by logging in to our existing Amazon account. So how does this Copilot work for Solo Brands? As Posner told participants:

Let me tell you what really inspired me. First, in no particular order, the most important thing is the ability to leverage the loyalty and trust that Amazon has generated among its consumers and transition into our environment.

“We have a lot of terrible AR – it’s nice to see a really good one.”

Posner says there’s a “nuanced value proposition” here involving stocks. This enables Solo brands to make better use of existing inventory and potentially undertake larger shifts across larger categories and risk styles – with the understanding that the B2C environment “can help increase throughput”.

For Posner, it’s not just about Copilot for Shoppers, but how all the pieces fit together:

One-to-one personalization, product reviews, upselling and cross-selling opportunities, all at the intersection of AR. I know we have a lot of terrible augmented reality (AR). So it’s nice to see a really good one. Placing our product in the context of consumers’ purchasing process and convincing them in the sales process that this is the right product for them will certainly help with conversions.

Opening a new, highly interactive channel like WhatsApp is also a big win:

Even though we are all betting on AI, using it aggressively as much as we can (combining one-to-one and one-to-many communications) via email and texting, it’s hard money. It’s starting to get a little commoditized. So the idea of ​​having a new communication channel that involves communicating with each other and having such an intimate engagement and communication with the consumer is a really great opportunity to build customer affinity and engagement, which will really enhance the brand advantages.

My opinion – trade islands and killing all B2B portals

Salesforce has greatly improved AI communications; I hear less hyperbolic statements like artificial intelligence eliminating user interfaces. Instead, Salesforce AI demos were comprehensive, with detailed elements such as disclaimers notifying users that they had generative AI results to review. These are the first I’ve seen that offer prompt templates, allowing business users to import relevant data into a prompt query (using RAG methods, etc.) – without coding.

However, not all customers will be ready for Copilots yet – there are also pricing issues to consider (e.g. a Commerce Cloud customer would purchase consumption-based Data Cloud and Einstein “credits”. To me, the most important topic of this program is not really AI, but , how far customers can take their business away from siled ways of working. According to the Commerce Cloud keynote: “Trade should not be an island.As Bucher told me:

When you think about marketing, commerce, data, and servicing all of these areas together… You have to take commerce off that island and think about the full outcome of the funnel, the full customer lifecycle.

An even stronger message came from a B2B client who was quoted (anonymously) during the speech

Kill all portals.

I can’t blame the client for feeling this way; The traditional B2B shopping portals are long gone. Bucher adds:

Customers expect a direct, consumer-level experience regardless of what they buy online, including when shopping for business. Think of a showroom complementing your products. Why should this experience be any less personalized than if you bought the same products yourself in a retail store?

Hyperpersonalization is an important topic in all of this. During a recent podcast, I shared some of my complaints about the hype around hyper-personalization, including the “scare” factor and the difficulty of predicting consumer context in real-time, given that our lives and priorities can change in the blink of an eye. Salesforce has some thoughtful responses to my complaints here. Short version: Salesforce believes it can overcome the downsides of hyper-personalization with human oversight of AI systems. In some cases, AI can help in a real-time context, not by trying to predict it autonomously (which worries me), but by pulling real-time data from the consumer into the scenario.

One key example: customizing an outdoor gear website with family-friendly gear in real-time when a shopper tells an AI bot they’re shopping for the whole family. This is actually the right context, something that hyper-personalization fans don’t always understand. There seems to be a better understanding here of how to design AI to take human feedback into account rather than fantasizing that we can live without the human element by correcting/modifying our interests and preferences. It’s encouraging to see designs that derive information about human preferences, but in a clumsy way.

But of course, human-in-the-loop projects have implications for calculating AI ROI. As it turns out, humans can be quite useful creatures, but maintaining them isn’t always cheap. So we need to look carefully at the return on investment in this type of AI using customer data. During today’s CMO panel, Salesforce CMO Ariel Kelman made the case that using this type of customized prompting with customer data via RAG could reduce or eliminate the need for enterprises to train/tune their own models, a potentially huge expense. I also heard good things about the tangible benefits of generational AI in these scenarios from other CMOs on the panel, but this will need refinement. That’s another time. Our team will soon be able to learn more about Salesforce Connections.