Talk to Your Customers

Thomas Bisgaard Kjølhede (a seasoned SaaS CMO) posted this on LinkedIn the other day:

The bar is really low in B2B SaaS at the moment...

Even though they have raised (too much) VC money, lots of data, lots of customer insights, lots of resources, talent, etc...

… then there are way too many who:

• Answers you with AI.
• Are obsessed is ROI on everything.
• Saves their prices for you.
• Think marketing = advertising.
• Pitch relaxes you EVERY time they have the chance.
• Claiming they are "number 1" or "the favorite".
• Believe content marketing written by AI actually works.
• Do not care about people's private lives and the rules about this.
• Selects keywords based on volume instead of intent.
• Judge the channel dead because the first test did not work.
• Talk about new ideas on slides instead of executing on them.
• Copying their competitors (not just inspiration, but directly copying)
• Doesn't have control of their ICP and therefore has a muddy user journey and messaging that no one understands.
• And I could go on forever...

Who is it that wrote that playbook? Because it is admittedly poor.

On the other hand, it gives you a great opportunity to take the lead and stand out positively:

• Talk to your customers regularly and keep track of your ICP.
• Strategy first, tactics second. And not the other way around.
• Build the entire user journey and your messaging according to your ICP and strategy.
• Execute continuously and adjust as you become wiser and get more data.

It may seem nerve-wracking, but it actually works and it doesn't take much to stand out positively.

The best way to grow your B2B SaaS startup is to get to grips with the boring stuff that actually moves the needle (and your AI chatbot on the website rarely does!).

Try it, it almost feels like magic.

This is brilliant. It applies to B2B SaaS marketing as well as every business and industry. 

Herodesk is in a super competitive space. There are tons of alternatives that customers can choose between. Herodesk is also “the new kid on the block,” but instead of trying to catch up in a never-ending feature race and blindly copying what all the others can and do, I’m taking a different approach. 

The lighthouse - my core believes

I have some core beliefs about what Herodesk should be and should do. 

• Herodesk must make it easy to give customer support
• Herodesk must be super simple and intuitive to use

It’s as simple as that. Whenever I consider a feature, large or small, I hold it against those two parameters. 

Will this feature make it easier or more difficult for my customers to use Herodesk to support their customers?

Will this feature make it easier or more difficult to use and work in Herodesk?

If the answer to either one is “no”, it’s back to the drawing board. 

Assuming it’s yes…

The strategy - what is needed?

I have some clear thoughts on what I want Herodesk to be and what’s needed in a helpdesk product to meet our customers' (the ICP!) needs. 

Rome wasn’t built in a day, so it’s not all there yet, but most of it is prioritized and planned. From there, it’s a matter of grinding and building. 

The next big thing I’ll work on (and write about) is Artificial Intelligence (AI). It is indeed “topic of the year”. Everybody is talking about it. Chat-bot services are popping up everywhere. It’s not that it’s difficult to make a chat-bot. What’s difficult is to make a chat-bot that gives usable (and not outright wrong) answers! 

I see a long list of areas where AI can help Herodesk become better in regard to the two lighthouse targets: Make it easier for our customers to help their customers and make Herodesk easier to use. 

AI is but one example, and I’ll write more later about my thoughts about AI in relation to customer service and how it affects SaaS pricing.

Listening to Your Customers

There’s this (in)famous quote from Henry Ford:

“If I asked my customers what they wanted, they’d ask for faster horses!”

… which implies: Don’t listen to the customers. Build your vision. 

This is true to some extent, if you’re building something novel. I’m not, however. I’m trying to build a new, simpler, and easier version of something that already is, while rethinking processes and workflows to make it easier for my customers to get the job done: Help their customers. 

In doing so, I must listen to our customers to ensure that the product actually meets their needs.

That doesn’t mean we’ll implement their suggestions 1:1. Often, their idea of how to achieve a certain goal isn’t the way it’ll be implemented. But when I learn what they are trying to achieve, I can hold that goal up against how I envision Herodesk, and they’ll often be able to achieve that goal, although in a different (but better) way than they originally imagined.

Three examples of this… 

1) Forwarding a conversation

I thought it wasn’t necessary to build a function that lets you forward a conversation in Herodesk to an external recipient. I was wrong. Lots of users requested it! 

However, this feature isn’t something that’ll directly impact how easy it is to get the job done or Herodesk usability. So, I looked at it from another perspective:

• A “forward conversation” feature is something that everyone can use (I won’t build a feature specific to one or two customers - this is the main reason I reject feature requests).
• If implemented correctly, it won’t make Herodesk less intuitive to use, and it won’t make it more difficult for my customers to support theirs.
• It is already requested by many people, and it will probably be requested by even more in the future.

With this in mind, I decided to go ahead and implement the feature. 

2) Custom sender domain

The majority of messages sent from Herodesk are via e-mail. I originally thought it was sufficient that these e-mails were sent from Herodesk system addresses, as long as our customers could set the sender name. 

Turns out I was wrong. 

The emails sent would look something like this: From: Anders from Herodesk <si2hrsidks@inboxes.herodesk.io>. 

That didn’t work. 

Some of my customers wanted to be able to send from their own domain out of aesthetics (it looks nicer). Others had it as a requirement in their organisation. And then there were some who got feedback from their customers that they thought it was a spam mail because of the sender address. 

Ouch! 

This was a no-brainer, based on direct customer feedback that proved my initial idea wrong. We needed support for sending emails from custom domains, the sooner, the better! (this feature was made available two weeks ago).

3) New design/layout for Herodesk

Even though the old Herodesk design was nice, it had some flaws that I only learned after talking to a lot of customers and viewing how they worked in Herodesk.

• It didn’t work well on smaller screens, as a lot of vertical space was wasted on unimportant things
• It wasn’t intuitive how to get started
• The color scheme used too many different colors, making it confusing and unintuitive

Based on all that feedback I decided to spend a month re-doing the design. This is another example of a project I wouldn’t have started on my own, but that got high priority thanks to customer feedback.

My lighthouse principles guide the overall direction of the project. The concrete features and functions that we implement are based on what I believe the product should be capable of to solve our customers' tasks. Finally, by talking to as many customers as possible, I get invaluable feedback that finds its way into the product roadmap in different ways. 

This approach helps me find the right balance between how I believe Herodesk should look and work to be as great as possible, and at the same time be capable of doing what the customers need.