Dharmesh Shah (Hubspot) - #MastersofSaaS

Videos · Business

This summary was largely done for my own note-taking, sharing it just in case it adds more value to other people.

I have no affiliation whatsoever with anyone in this note. This is a summary largely taken for my own reference, and may contain errors :)

Highlights

Had little entrepreneurial bent early on. Starting a company wasn't on the agenda
Was passionate about computers and realized that he wanted to do software, and the large company he initially worked for wasn't great at dealing with devs. So started his first company in 1994 in Alabama without any business background. It was Pyramid Digital Solutions, an enterprise software startup in the financial services sector
Sold the company to Sunguard in 2005
Initially told his wife he wouldn't start another company
The second company he started just went sideways. He was much more knowledgeable for the second one and did things differently –but it failed miserably

Functional decomposition — being able to take a large problem and break it up into pieces. If you learn and try with iteration and some kind of loss function, you’ll figure out the right way eventually. As long as you get better and there’s a short feedback loop
Talent vs Skill. Talent is the ability to acquire skills faster than the average person. All skills are acquirable, it’s a matter of time

Problem of hiring for skills in a startup is that you may not know what skills are even needed
Skills are acquirable. Talent and temperament are mostly innate
Temperament includes humility, intellectual curiosity, empathy for customers and coworkers

Not being mindful around team diversity. Hired for convenience. You hire for people in your network that are like you, which can make your team narrow
Culture debt. It's analogous tech debt, and he didn't recognize this early on

Thought that consumer companies had bimodal outcomes (huge or bust). But B2B companies could have more "normally distributed" outcomes
Thought that a B2B company for SMB kind of gives you the best of both worlds. Bottoms up growth with stable revenue
Realized that there were a ton of tools for SMBs, but very few SMBs were taking advantage of them. This was because it was very difficult for a normal person to make all of those tools work together. Putting Wordpress + Google Analytics + a bunch of tools together is hard if you're not very technical — particularly back then
This is why Hubspot's product was so broad at the onset of what they were doing. This also why their product sucked at the beginning because they were so broad with limited engineers

Essentially created the Inbound Marketing category. It was exceptionally hard, expensive, and fraught with risk
It worked out, and this is rare. Advise to entrepreneurs is not to immediately jump to the notion of being a category creator. Instead, fit your product into an existing category first to get initial adoption. Then, if your product is indeed very far removed from the category you're fitted into, try to create a new category
For Hubspot, their "initial category" was marketing automation. But because their product was the antithesis of this, they had to create a new one
For creating a category, you need to have elements of your category description that not everyone agrees with. Be opinionated and be open to offending people. Like "no one wants to see ads, you need effective content to attract an audience". Create some controversy and polarisation
If you're doing category creation, you'll need resources. Bootstrapping is not going to work

You don't want to say "I have a category creating product". Instead, say "the world is broken in this way". The former is about yourself, the latter is about a real problem that (hopefully) a lot of people face
Then, articulate your product in a simple enough way that people can actually understand. This is hard with a new product. So spend a lot of time on storytelling

Frank Slootman of Snowflake thinks that you need to win the enterprise to win a meaningful market. But Hubspot has won with relatively low ACVs in SMBs
People have traditionally used SMBs as a way to get to the enterprise. Historically, the money has been in the enterprise
Startup incentives are geared towards selling to enterprise, because that makes their numbers look better
They resisted going to enterprise until very late, because:
- Selling to SMBs is increasingly becoming a viable strategy because the number of SMBs that can use your product is increasing
- Competition in the enterprise is far more cut-throat. It's hard to differentiate as a startup
- The sales cycle is super long in enterprise and speed of iteration is lower. It's much shorter in the SMB sector
- Feedback cycles are much shorter in the SMB sector – people vote with their dollars much more quickly

The core GTM machine is "put $X of customer acquisition cost, get $Y in LTV back", where Y > X
Getting the machine right > getting the product right
Hubspot's initial product sucked, wasn't the best for many years
But if you can get the organisational machine right, it can pull the product up. The opposite is much harder
Look at instrumentation. Make sure that the machine is working (don't treat it like a black box)
Know your knobs and dials (input variables) and know how they map to the output variables
What are the tradeoffs that you must make?
Churn is the biggest problem in a SaaS company. Know what causes retention and what causes churn. Retention is the most important thing for net growth

Finds continuous blocks of time. Long stretches of times for meetings, and long stretches of undisturbed time to get into flow states
As few meetings and phone calls as possible

Short
Too many founders make the mistake of waiting too long before asking for money to make the product
It doesn't matter if the product sucks. Your job as an entrepreneur is to mitigate risk. If you want to understand if you are solving a real need, ask people to pay
If you're smart enough, you'll be able to build the product. The harder thing is being able to find a market for what you want to build
You'll also get more honest answers from customers if you ask people for money

Very few. Only a couple things
Eventually, if you don't make the product great, your company won't get to where you want it to be. You can start off with optimizing the marketing part of the machine, but eventually you'll have to optimize the bottlenecks (product quality) when you have real customers coming in and growing