How to sell AI software to Customer Experience Teams

Part 1: Show Off Your Expertise

Intro

My goal in this post is to provide practical advice for founders and investors on how to build and sell Generative AI products to Customer Experience (CX) teams.

First though, a bit of my personal backstory. I've spent my entire career working at startups leading CX teams in multiple tech and e-Commerce companies such as DigitalOcean, Fab.com and Wild Alaskan. As employee #35 at DigitalOcean, I led a variety of 24x7 Customer Experience teams including Customer Success, Technical & Developer Support, and Trust & Safety - helping grow the company to 600 employees and $250m ARR. Most recently, I've been doing Customer Experience consulting for early-stage startups to bolster their processes and operations.

In my experience, companies that truly enable their CX teams to take care of the customer come out ahead against those who don't. Danny Meyer would explain this as Enlightened Hospitality because it focuses on empowering employees to make customers feel really valued. In Meyer's book, he writes:

“The service is the technical delivery of a product [...] Hospitality is how the delivery of that product makes its recipient feel.”

It's core to why I believe that CX can be a major differentiator for companies. Each customer touchpoint, from how support teams interact with customers to the policies that are developed, is an opportunity to create amazing experiences and drive loyalty.

Here is where there's a big opportunity for Generative AI to complement CX in order to drive more business value. I see that Generative AI used by CX is the merging of superior processes and customer interactions with an always-on system that every company on earth can use to drive business value.

To bring this vision to fruition, I believe that companies must do two things very well to sell their AI Software to CX Teams.

1. Companies must educate buyers to earn trust in the market. I'll discuss a few ways to do this, and discuss the mentality that founders must have to win in the long-term.
2. Once your product and positioning resonate with buyers, you must actively help them purchase and use your product. Doing so requires demonstrating value across the company, building a compelling ROI, and making integrations simple.

Creating a Superior AI Experience for CX

The best CX products I've ever used are built by founders who have a real itch to scratch. They have experienced the pain that they're trying to solve for, and their first-hand experience building solutions at a past company has given them unique insights which they combine with a long-term vision to build valuable solutions.

As we witness hype for Generative AI the topic is everywhere from LinkedIn to the Wall Street Journal, which means it has permeated into mainstream discussions much faster than crypto. Because of this, the ecosystem is exploding with many tools being built on platforms like OpenAI or hacking on Meta's leaked LLaMA model, which is truly exciting.

This begs the question: With the explosion of Generative AI tools, is it possible to build an AI tool for CX and ride this wave without experience in CX?

A former peer of mine, and the Founder/CEO of Catalyst.io, Edward Chiu, recently posted this insight on Linkedin:

“When you buy software, you are also buying expertise. No matter how intuitive the software, customers still need best practices, guidance, and direction on how to expertly use it.”

What I think he's getting at, is that a software platform must be complemented by a human's expertise on the subject. This is how you connect deeply with your customers and help them achieve more than anyone else could. You help them solve real business problems through a deep expertise in the function.

So, if you're a founder with a vision and deep experience in solving CX problems, keep building!

If you're a founder without CX experience and you really want to solve these problems, I think it's still possible. It will require lots of learning by spending time with potential customers, learning the industry, and deeply understanding their challenges. I remember back to 2013 when MaestroQA was starting, and Vasu Prathipati made it his mission to learn all about CX in order to build his company. Fast forward 10 years, he and the team have built an amazing and profitable business that raised its Series A 8 years after launching source. The difference between 2013 and today is that buyers and VCs don't have that much time for you to figure it out.

Ultimately, to win customers, I think that you need a strong connection to the problem you're solving. Otherwise, you're only selling a product feature instead of selling a vision. When thinking about scaling your company you must build a great product and attract an audience to grow, which I'll talk about next.

Educate to Attract Prospects- Product Marketing through Education

A key to showing your expertise in CX and AI is to write, produce and share content and ideas. I suggest beginning with the very basics of Generative AI, working all the way through more complicated topics and how they relate to your potential customers. This includes documenting how your own product functions. By doing this, you'll educate prospects, current customers, and potential employees by positioning yourself as an expert. There should also be a consistent CX theme throughout your content, which targets a specific audience.

If you're skeptical that this strategy will work, a perfect illustration of this is how DigitalOcean took a well-known and rigorously documented topic, Linux System Administration, and built a customer acquisition powerhouse on top of content with amazing SEO. We began by simplifying complicated topics by writing as if our users were 5 years old, and then built on top of the initial momentum by adding more and more advanced content. Do not miss the forest for the trees by thinking that someone else will write the content or that a piece of content is beneath your audience.

I think content education works for two reasons:

1. To truly learn a topic, the learner must start with a foundational and accurate understanding. Not only will beginners benefit from this information, but it will be helpful reference material for experts as well.
2. The pace of the AI industry means that we've all missed something crucial in our journey, so there's always room for questions from experts.

Going back to your expertise, content is also how you demonstrate your understanding of CX and its connection to Generative AI. Writing content that directly addresses existing customer concerns, and is written with solutions in mind, will earn you an audience. Over time, you'll be able to monetize your audience if you're also building a product that's easy to implement and solves problems.

When distributing your content, meet your customers where they are. I always advise adding content to your own site in the form of a blog or LinkedIn posts. More recently, I would recommend video content and shorter written pieces for other mediums such as TikTok. If you're already ingrained in CX, then you'll also be familiar with online CX-communities where you can share this content to a very targeted audience. By creating engaging content and linking back to your own site, you'll increase awareness for your product. This allows you to capture a prospect's intent, control the messaging, and do acquisition marketing.

As a company you'll gain a competitive advantage if you produce content from day one. Again, start at the start, be basic, and then build. If you're an investor, look for founders who are great storytellers and who can engage you with their content, this will also help in their sales process.

A Note on Market Size

Here's something counterintuitive to drive home the reason why content matters: The size of the market for Generative AI tools in CX is not the same size as the market for helpdesks like Zendesk. How could that be?

Ignoring hurdles such as HIPAA-compliance, the gap between the two markets comes down to educating buyers about how Generative AI works, why they need it, and how their company could benefit from using it.

If you fail to properly educate buyers, then they are incapable of purchasing yours or a similar product simply because they aren't aware, and you're leaving money on the table.

CX Financials, Tooling and Respect

To create the future for CX it's helpful to have context on where we've been, and I use these examples to help you help us (CX). There are exceptions to these examples, and I encourage all of us to learn from these organizations. For the rest of them, here is how I've seen companies operate.

A tight budget, even before the downturn.

The first bit of history is that most finance leaders classify CX a cost center. Said another way, putting $1 into CX isn't expected to produce any return, whereas spending $1 on another piece of marketing software… well you get the point. As a result, most companies and their financial leaders are actively trying to reduce their CX spend.

What I've typically experienced is pressure to reduce costs by increasing the number of customers per CX agent, lowering spend as a percentage of total sales, or lowering costs by outsourcing to a lower cost region.

Additionally, getting budget for new spend comes with great scrutiny. Companies that are atypical in this regard often have a co-founder with enough influence to prevent this spiral.

To illustrate this in action I'll highlight a recent example while I was working on integrating data between Snowflake, Airtable, and Stripe. I turned to Zapier to stitch them together since I'm technical enough to be dangerous, but not to write my own API connectors. Plus, third party tools like Zapier greatly increase efficiency.

As I was working on creating Stripe invoices, I ran into Stripe API issues using the Zapier connector. Since Stripe was throwing the 4xx errors, I contacted their team. While I quickly got someone on chat to help, it was clear they weren't capable of troubleshooting API issues, and worse yet, they weren't willing to dig in since I was getting errors in Zapier's platform.

Frustrated, I turned to Zapier Support. It was much more difficult to get in touch with their team since they use a lot of deflection and heavily gate their humans. However, I got a response ~48 hours later and it was thorough, helpful, and addressed my core questions.

I don't pretend to know the inner workings of Zapier nor Stripe support, but candidly, it feels like Zapier is much less cost conscience. Zapier has opted for high knowledge team members, and understands how positive CX experiences can differentiate their product in a competitive market.

Dangerously, most companies deliver a similar experience to Stripe, but they're not in the same dominant market position which positions them poorly against competition who delivers a better experience.

In Part 2, I'll provide suggestions on how to move beyond these hurdles and turn financial leaders into promoters for your product.

Tooling

The second bit of history that I want to discuss is the set of tools that have been promised are underdeveloped. For years, we've been hearing about chatbots being the future for CX. In reality, what we've received are not much more than interfaces to create If/Then workflows that are human generated and are interspersed with “does this answer your question?” prompts. For example, in 2019 Zendesk acquired Smooch.io source, which became Sunshine Conversations and is the core of their messaging product. If you view their release notes, you won't see much product development in the past year+, and this the core of their product.

Generally, CX tools promise to increase a team's efficiency and lowers costs, which allows the CFO to sign off. I'm not here to say that all tools have missed the mark, but reports of uncapped efficiencies have been greatly exaggerated. This has made the typical CX buyer very skeptical of new tools, which generally makes selling them harder.

We've already spoken about showing your expertise, so in Part 2 I'll focus on using time to value as your secret weapon to fight tooling's fraught history.

Working Cross-Functionally

The teams that garner the most attention drive revenue or create the newest product that promises to drive revenue. As I outlined, CX is seen as a cost-center, so on a historical basis CX teams are marginalized inside of their companies. Customer Success is an exception to this rule when they show that they drive revenue. It's gotten so tough in this climate, that I'm afraid retention might not even be compelling enough these days for Customer Success to earn respect.

Here's how it works today: In many organizations, when a customer has an inquiry, it's seen as theCX team's responsibility to solve it, rather than a product or organizational problem that led to the friction. A core reason for this seems to come back to tension in Engineering and Product between new product development, fixing bugs, and maintaining the infrastructure. It never seems that there's enough resources for any of the work. In my experiences, the best run organizations consistently do a few things that improve cross-functional working relationships with CX:

• Diligently track customer feedback and set expectations for follow up
• Triage urgent issues for large and small customers
• Have clear process for incorporating customer feedback into the product
• Create clarity for product/engineering resource allocation
• Have a clear product launch process which reduces downstream issues

Here's how Generative AI can help:

A great way to help CX organizations actively participate in a well-run process is to help them do these things (and more) efficiently and well. Efficiently means using Generative AI tools to dramatically increase the output and reduce the time requirements. Doing this well means helping CX teams integrate cross-functionally with their peers and their tooling.

Help us Buy Your Software!

I've spent time describing some of the raw ingredients and thinking required to build a Generative AI product for CX teams. I covered your rationale for building the product, how to build awareness, and discussed some challenges you'll face within organizations. My hope is that one or a few of the ideas resonated with you, and that you consider incorporating them into your work.

In Part 2 of this post, my goal is to help you close more deals by explaining what it's like to be a buyer in a CX org and what I think SaaS companies need to do in order to close Generative AI deals.