Scaling Startups to $100 Million in ARR: Insights from a Customer Success Expert

With a focus on churn that parallels our customer retention goals as MSPs, I believe there’s a lot to learn from the SaaS market. So, I set out to do just that and ended up tracking down an expert in customer success and SaaS trends: Luke R. Diaz , founder of DBT Ventures (as in Do Big Things).

?

Diaz both oversees an incredible portfolio of angel and VC investments and advises on tech, software, and other innovations. He’s helped grow three startups from single-digit millions to over $100 million in ARR, which made for a fascinating deep dive on his bonus episode of The Business of Tech.

?

Keep reading for actionable ways to incorporate SaaS churn tactics into your MSP’s customer retention strategy.

?

The Past, Present, and Future of SaaS Spending

?

We began with an overview of the lay of the land of SaaS and where it’s heading in the next couple of years. In general, Diaz is seeing a lot of retraction in the market cycle due to inflation pressure. The squeeze has subsided a bit, but CFOs are continuing to prioritize free cash flow and efficiency over the ‘growth at all costs’ narrative that, according to Diaz, is now seen as unsustainable.

?

“What I am seeing is continued growth in SaaS, but at much more conservative estimates—in the 10 to 20% range over the next year and a half as the continued price pressure comes in upon renewal time,” he said.

?

There’s an overall sense of scrutiny when it comes to getting the most bang for their buck on software spend, with the usual ‘What have you done for me recently?’ question continuing to drive retention decisions.

?

Retention Tactics Innovation

?

This ongoing squeeze means that churn and retention are more important than ever. And this applies to us too; as a colleague of mine once described, the key to SaaS is not so much the signup, but keeping people from leaving.

?

So how are companies innovating to improve their renewal rates? Does that link to fundraising?

?

I ran these by Diaz, and he confirmed that there’s a direct correlation between fundraising and efficacy, all rooted in basic unit economics. The VCs he works with are looking for retention rates in the high 80s, if not the low 90s, to even consider writing a check.

?

The most clever retention solution he’s seen companies like MSPs use works best when targeting smaller businesses. When there are a lot of customers playing a small amount, people will build automated renewal flows to make engagement much more user-friendly.

?

For example, Diaz built an automated DocuSign process so customers could renew their contracts with a few clicks without any further hassle. HubSpot email communications have also been helpful in getting folks to sign, as they streamline the process with less human intervention.

?

I asked Diaz to share more about the evolution of these tactics. He explained that customer success management has many different flavors; revenue-driven methods are on one end of the spectrum, and value creation is on the other.

?

Looking forward, he believes we’ll see the revenue-driven flavor go down dramatically as the pendulum shifts to the value side of things – where CSM is all about driving business outcomes for customers to help them get the right ROI from their spending. He also predicts this trend will be coupled with more data-intensive tactics:

?

“There's become this expectation and a need for customer success teams to be data ninjas and very comfortable with SQL, leverage AI and chat generative AI tools like chat TPT, all in service of adding value for their customers,” he said.

?

AI’s Practical CSM Applications

?

The AI mention reminded me of an article Diaz wrote on LinkedIn about using ChatGPT in CSM. I asked him to expand on AI’s practical uses here, and he pointed out that although anyone can make a ChatGPT account, it’s not as easy to systematically integrate it into your workflows to actually create customer success efficiency.

?

But once the more intensive adoption is there, Diaz predicts tremendous cost savings on two fronts: information consumption and content creation. Customer success managers spend hours a week reading 10k words on a company’s financials and building custom decks.

?

“You're starting to see these pipelines or these API flows where these LLMs can populate different formats very effectively so that you're not starting from scratch. You literally have a 80 to 90% built deck for your next QBR using a fine-tuned LLM. And that gets me really excited for the next phase of this,” he said.

?

Machine Learning + Churn Prediction Models

?

Diaz is also seeing early signs of success in building churn prediction models with machine learning to calculate a portfolio’s risk with a ‘churn prediction score.’ I asked him to elaborate on that process:

?

“At the highest level, we just fed a machine learning model a ton of data. We saved hundreds of thousands of dollars because we now had a machine-learning approach to answering, where is the risk in the portfolio? We trained the models off about 400 renewals that were either successful or unsuccessful.”

?

Diaz also added a tip for the audience here: instead of forcing your team to follow an AI model’s exact advice, present it to your people as a potential red flag worth investigating further.

?

“If you have customer success managers or humans in charge of this, they won’t feel that their job's being encroached upon, but rather, hey, this is actually trying to help me find where the fires are so that I might not have seen the smoke yet,” he said.

?

The Force Reduction Question

?

Wondering if this means people will be replaced? I asked Diaz where he falls on balancing chat agents versus humans when interacting with customers, and he essentially believes that 1) the tech isn’t ready for that yet, and 2) human connection will always need to be a part of CSM.

?

“The pricing and value of Microsoft's Copilot is an interesting jump-off point. If you look at the base 365 license compared to Microsoft All AI Copilot, it's actually 6X as expensive as the base 365. Do I think generative AI is going to make all humans six times as effective? No, I don't. I think that's wildly optimistic in the short term.”

?

And although he does predict some replacement, he mainly envisions better efficiency, like having one CSM manage 300+ accounts thanks to chatbots, automated workflows, and automated product feedback.

?

Diaz’s Top Indicator of CSM Success

?

To wrap things up, I asked Diaz for his #1 indicator that an emerging organization is succeeding with CSM.

?

His answer was simple: curiosity. When a person in this role is willing to stop talking, listen to customers’ stories, and ask great questions…

?

“... then the customer success function can encode all that, feed it back to product engineering. And then all of a sudden, you have a $100 million SaaS company,” he said.

?

?

Want to learn more about Diaz’s retention magic? He published book notes from his latest research at www.dbtventures.com . You can also follow along on LinkedIn, where he writes under Luke R. Diaz .

?

That’s it for this week! As always, my inbox is open for insights, questions, or whatever else is on your mind.