Startup validation framework

Startup Validation Framework

This is a summary of tips/frameworks on how to get out of the -1 to 0 phase of building a B2B SaaS company. This has been pulled from a variety of different sources (most heavily from Lenny Rachitsky's excellent interview series on his blog) and summarized. I am in NO way taking credit for any of the following, this is merely my summarization and compilation of amazing work.

Coming Up With An Idea

Idea Ingredients (need all 3 to align):

1. This idea is important to people because…
   • Why do people care? What’s the personal psychology at play?
2. The market is underserved as evidenced by…
   • What is the current solution (scale 1 - 10)
   • What is your solution?
   • Is your solution +4 the current solution?
3. You’re excited about solving this problem
   • Could you see yourself doing this for the next decade?

Paths for Idea Formation (ideas fall into one of these 3):

1. Past pain - pain point experienced at previous company (~40% of late stage startups)
   • Anything built at previous companies that would be valuable to outside world?
   • Is there a tool you wish you had at previous companies?
   • What internal products have you built again and again?
2. Ponder and probe - pick a space and whiteboard and tinker while talking to dozens of potential customers looking for pain and pull (~30% of late stage)
   • What’s a trend emerging, but underserved?
   • What’s a transformative technology that’s emerging but underutilized?
   • How many potential customers have you spoken with about the idea?
   • How important is the problem you’re exploring (scale 1-10)
   • How underserved is the current market and where would the solution rank? Is the delta at least 4?
3. Present pull - identify something you’ve already build that’s showing signs of pull and pivot fully to that (~30% of late stage)
   • Of all the things you’ve built is there a feature that is showing disproportionate pull?
   • What problems are you having with your startup that you wish someone would solve?
   • What’s a side project you’ve wanted to build but haven’t?

Output of this step

1-2 liner of the idea, the exact problem it solves for customers, the market it serves, and why now is the right time to build it.

Idea Validation

Founders at startups that have made it to later stages spoke to a median of 30 potential customers before feeling like their idea was solid.

Outbound sales is the best signal for validating your idea (as opposed to friends, incubator batch mates, or investor leads)

Warning for Prosumer: Every prosumer collaboration product (Miro, Figma, Notion, Slack, Coda, Airtable) spent 3-4 years wandering in the dark before stumbling on something that clicked.

Validation Paths:

1. Listening path. Talk to tons of potential customers and then start building.
   • Default path unless you have a clear sense of what you need to build. Speak to at least 30 customers, friends/acquaintances make up no more than 5 of those.
2. Do it manually. Don’t build anything, solve the problem manually first for a small number of companies.
   • Do this if you’re unclear whether the problem is important or even solvable.
3. Prototype path. Start building a prototype and then co-create with a small number of design partners.
   • Do this if you have experience in the space and come into it with a clear sense of what to build. Clear founder/market fit - you’re an expert and know what to do.
4. Just launch it and YOLO.
   • This is one of two cases: founders that are super desperate and need to just try stuff or founders that have a crystal clear vision of what needs to be built.

Checklist for validation (must complete):

• Is this a hair on fire problem?
• Unit economics must be great
• Is there an existing budget for this solution or selling to a decision maker with lots of budget to throw around?
• What is the wedge (clear distribution strategy)
• Is it recurring revenue and selling to businesses?
• Is this really sticky/core and hard to rip out?
• Is the ACV large enough to sell to smaller (30 - 100 person) companies and have it be worthwhile?
• Why is now the right time to start the company?
• Is it truly a software company, not a services company?
• What is the barrier to entry? Is it a defensible business?
• What data do you generate? How can you use that data to build a moat?
• What is the competitive landscape? Are you a +4 solution?
• This idea does not have significant platform risk
• Does this require approval from just one function? Is there a single player mode?
• Is the ROI to save money or make money crystal clear?
• Is there an existing large company you can point to in this space?
• Have you spoken to at least 30 potential customers in this space?

What is the sign that there’s something there? (need at least 1):

• People we don’t have a connection to have paid us money.
• There is continued unprompted usage.
• Strong emotion. Hatred toward incumbents and strong emotional reaction to your idea (not “yeah that sounds cool”)
• We’ve gotten non-trivial cold inbound interest

Output of this step

A validated idea that is venture scale -> time to build, sell, and then raise money. Or the idea fails validation and back to square one / pivot / iterate.

Defining ICP (ideal customer profile)

Try to get as specific as possible when listing out the attributes below. Use outbound sales results as calibration for whether you have the right ICP or not. Disregard any signal from friends / close connections.

This typically has to emerge over time as you're going through initial sales motion. It’s an iterative process, so start with a hypothesis and continue to refine it as you have more conversations and see who is getting the most excited. Be careful not to pattern match too early, ensure that the great customer you spoke to actually exists at other companies and isn’t some unicorn.

Attribute of ideal customer	Examples
Company size	20 - 50 people
Job title / role / level	Revenue operations
Pain point	Don’t know why deals lost
Unique way of working	Design focused
Specific tech used	Github
Type of business	B2B SaaS
Price point	Sells software > $10k ARR
Geo	Texas
Unique place user spends time	Reddit

Different example attributes:

• Company size (e.g. 1,000 - 5,000)
• Job title (e.g. engineering manager, social media manager)
• Pain point you’re solving (e.g. compliance, internal transparency)
• Company’s unique way of working (e.g. design-driven, operationally heavy)
• Specific tech used (e.g. data warehouse, GitLab)
• Type of business (e.g. B2B SaaS, e-commerce)
• Price point (e.g. sells software that’s $10k ARR)
• Geo (e.g. urban centers, LatAm)
• A unique place the user spends time (e.g. Node.js community)

The more narrow you get the more you scope down the product which reduces time to build and speeds up the validation process. You can then outline a path to growing TAM buy removing some of the restrictions on ICP (add geo, larger/small companies, etc).

Summary of ICP tips

• Get super-specific and super-narrow with your ICP. Almost comically narrow.
• Start by paying attention to who gets most excited about what you’re building.
• Pay attention to which roles are most successfully implementing your product.
• Pay attention to the size of the company that’s the best fit for your solution.
• Look for patterns in the ways the company works.
• Be careful not to pattern-match too quickly.
• Pay the most attention to signals from outbound sales, instead of leads coming through your network.
• Keep at it—it often takes many months.
• If all else fails, focus on what you can sell this month.

Output of this step

Clear pipeline of customers and feature set needed to get to $1M in ARR. You should be able to use your ICP along with Sales Navigator or some other tool to build out your initial sales pipeline and run a motion against it.

Winning First 10 Customers

General sequence for early customers:

1. Start by reaching out to your network, looking for people who match your ICP
2. Go cold outbound, but be strategic about it
3. Tap your investors’ networks
4. Participate in relevant communities—and network
5. Put out compelling content and build a following online
6. Get press
7. Just launch

Takeaways for unicorns:

1. None of these seven strategies scale. That’s why they work. In B2B, it always starts with hand-to-hand combat.
2. Cold outbound works—if done creatively.
3. PR can work, but rarely does.
4. A surprising number of founders found their early customers by putting out compelling content online and first building a following.
5. Y Combinator (YC) especially is very effective at helping you get early users, but not the way you’d think (as you’ll see below). Also, major shoutouts for First Round Capital.
6. When joining communities, focus first on adding value to the community. No one wants to pay attention to you if you’re there just to pitch your product.
7. For your first set of customers, trust is the key.

Be skeptical of friends or people that you know until they’re willing to pay. 2nd degree connections are even better as it reduces the friction to getting in front of them, but they don’t have any history with you or a direct relationship to worry about.

For outbound, get creative. Figma used a script to identify design influencers to sell to, Retool used Crunchbase to filter down by ICP and then started emailing CTO and VP Ops at those companies, Zip exclusively used LinkedIn but focused on advice and feedback rather than selling right away. Zip made the active decision to focus on cold outbound and skip any friends.

For investor networks and incubators, ignore batchmates and only sell into alumni that are a few years out of the program.

Again, big credit to Lenny Rachitsky for the amazing interviews and write ups he's done over the years on this topic. I've just summarized the key points here, but if you want to read the full interviews check out his blog.