A few weeks ago, we announced in a TechCrunch exclusive that Deepgram raised a $1.8 million seed round.
Even with the YC name behind us, raising money as an early stage company is challenging. We added to that challenge by pitching our investors “naked", and did so pretty shamelessly. Silicon Valley might have a lax dress code, but we weren't in the nude, we pitched without a deck.
This could make many feel vulnerable. For us, it was one of the best decisions we made in the fundraising process, and I'd like to share what we learned.
Why Pitch “Naked”?
This isn’t breaking news to anyone: In early stage startups, speed and efficiency are everything.
We found out that pitch decks are kind of like crutches. There’s nothing inherently wrong with using them if you need them, but they can slow you down. If you’re building something that people want, your pitch should be able to stand on its own. You should try to have a narrative arc that the investors can follow without many visual aids.
If you’re relying on a slide deck, either every investor is bound to see the same presentation every time, or you’re going to waste a lot of time changing around your presentation based on your assumptions on what each investor may want to see in your deck. Nothing about that is efficient.
How do we know this?
Before our first-ever meeting with an investor, we made a pitch deck. We poured a bunch of time into it, and the deck was actually pretty decent by the end. We sat down for our meeting and dutifully walked through the first half of the deck. We clumsily went through the motions and halfway through our planned presentation we just started telling our story and showed a prototype of Deepgram.
The story and demo is what got the investor's attention.
We didn’t go back to the deck during that meeting. We tried once or twice more with other investors and totally abandoned the deck. That worked out just so much better than our expectation.
Possible small exception: maybe bust out a graph of impressive metrics that go "up and to the right" if you have it (we didn't).
After YC Demo Day, we did over a hundred investor meetings in one month, all deck-less. This freed up time that would have otherwise been wasted on editing slides.
It also let us iteratively improve our pitch more quickly than we could have if we had to use slides, and it allowed us to personalize our pitch to each investor on the fly.
I was forced to recollect and form the story over and over then witnessed how that story affected each investor. After you do that enough times you start to recognize patterns. You'll see what things investors care about and what they don't. In fact, we found that you can repeatably lead investors through the story of your company while simultaneously heading off concerns (for Deepgram: "what about Google?"), answering their many questions before they ask, and doing it all while saving time.
How We Structured Our Naked Pitch
First we need to think about the goal of the pitch. The real goal is to get the investor interested enough after the meeting to dive into your market and see how your company fits in. Once you have them caring about what you are up to and you've got them rooting for your company, that's when you are in a place of possible investment.
Before we discuss the pitch itself, let’s start at the beginning: time management. For us, 100+ meetings in a little under a month was a huge logistics challenge. We managed it by “chunking” meetings by geography and setting up meetings only a few days in advance. So, for example, we’d take three meetings in Palo Alto and two in Mountain View, and we’d take calls from investors who weren’t based in Silicon Valley during the car rides between the towns. This allowed for maximum efficiency in terms of time, but also compelled us to maintain momentum throughout the day. Then we'd have a San Francisco day or two that week, with more calls between those meetings too.
By the end of the 100 meeting month marathon, we had the meeting process down to a science. Scheduling only a few days out and being maximally flexible allowed us to pack meetings one right after the other in time slots that investors wouldn't give up a week in advance.
Our meetings would proceed like this:
“Hey, how are you? Thanks for taking the time to talk. Did you order coffee yet?”
Purpose: Greet your fellow humans to let them know you are a nice person.
Ask Investor to Introduce Themselves:
We’d invite the investor to share a little bit about themselves and the firm they worked with (if they have one — Hi, all you angels out there!).
Purpose: get clues on how to shape what we say in the meeting. I always wanted investors to go first because
- They will talk less than you so it is good to get it out of the way and
- When you go second you can move straight into the next very important part.
Tell your origin story:
We would mention that we cofounders had worked together for years doing hard problems and fun side projects (like building drones and trading bitcoin), and describe the way we discovered the opportunity to build Deepgram (my cofounder Noah is a big proponent of life logging with audio recorders, the thousands of hours he amassed showed us the glaring problem people face in organizing audio).
Purpose: Let investors know we aren't a fly-by-night team, that we are good problem solvers, and that the idea was novel and solvable.
Deep dive on tech:
We'd go into our technical backgrounds (designing, building, and operating deep underground dark matter detectors, then analyzing the data), which showed investors that we were actually capable of building Deepgram.
Purpose: Deep proof of technical skill, show ability to communicate a complicated problem and how we solved it previously.
Explain the trajectory of the company:
Being very transparent about things that might otherwise raise investors’ eyebrows (like the fact that we went through two accelerator programs, BoostVC and YCombinator, have a great win with a customer [$500k revenue increase in our first pilot], and we'll be tackling the call center space, and need to hire key people).
Purpose: Show investors that you are aware of your company's place in the world, you are on top of problems, and you have a properly reasoned company trajectory.
We did all of this to prime investors’ interest in what we were doing, and to preemptively answer many of the questions investors may have about us or the company. This helped establish credibility up front, and positioned us as a competent team that’s capable of anticipating obstacles and executing on our plan.
And that’s the most important thing. It’s a tired cliche but at the early stage, investors really do “bet on the jockey and not the horse”. It's your job to tell them who you are and what you're capable of.
Keep in mind that those first steps would only take five or six minutes to get through. Again, speed and efficiency are key.
After that, we’d typically start answering nitty-gritty questions like:
- How much are you raising and on what terms?
- Who is your ideal customer and how are you connecting to them?
- What key team members are you missing and what’s your plan for filling those gaps?
Tell them very succinctly:
In a typical meeting, we’d spend twenty minutes pitching and answering questions, and another few minutes talking about next steps and goodbyes. By design, we didn’t leave enough time to answer everything about our company. We want the investor to leave with their wheels turning and go digging into our market.
Remember: checks aren’t signed during that first meeting. Our short-format meetings with no pitch deck created a filter for us. Leave the investors wanting more and they will follow up for more information. That's when you know they are serious.
After the Pitch:
Pretty quickly, there were investors emailing us back, asking for something they could look at, think about, or pass on to their partners. At first I resisted but I realized they need a memory aid. Instead of a pitch deck, we sent investors a bullet point list that followed the rough narrative arc of our in-person or phone pitch.
Why a bullet point list? For one, it’s very easy to modify and customize as your raise moves forward. But the format also gave investors a 200-word snapshot of what we were working on that’s visible all at once, rather than in some cut-up slide show presentation. We were also forced to formulate our value proposition to investors in concise bullet point form.
Because this list served us so well, we decided to share it for you to use and adapt for your own particular startup. This email would be sent an hour or so after your first meeting.
Here's our email, verbatim:
Hey [Potential Investor], Great meeting today! I’ve included the bullets below. ___ Deepgram Summary - Deepgram lets businesses search and analyze their recorded calls. - It is not just speech-to-text with text search added. - The traditional transcript + text search is only 20% accurate on phone calls. Deepgram search is over 80%. - The better search accuracy is a result of Deepgram's AI audio search. - Search is a tool for customer data and to help train our AI models quickly. - The team: Former particle physicists who built big needle-in-a-haystack searches for Dark Matter. Audio search was a side project outside of the lab, and Deepgram was started based on the tech. - Verticals: Call Centers (there are plenty of others, but companies with calls are the focus). - Over a billion hours of calls are recorded each year (and with services like Twilio, this number keeps growing by a lot). - The calls need analytics for QA and compliance, but this is done only to a few percent of calls and mostly by humans (expensive and bad coverage). - Deepgram is accurate and inexpensive and the first pilot increased the customer's revenue by $500k/year. - Fundraising: Raising Xmm on Ymm cap, with Zk committed
Notice that the bullets were written in the third person. That's because you want it to read like an objective list of qualities. When the list is forwarded on, the other partner will see matter-of-fact attributes of an investment worthy company. :)
Here's the line-by-line breakdown
This is what each line was attempting to impress on the investor's brain:
Hey [Potential Investor], Great meeting today! I’ve included the bullets below.
Give a nice greeting. Remind them that you met today and they asked for more info. Short, sweet, and easily forwardable.
___ Deepgram Summary - Deepgram lets businesses search and analyze their recorded calls.
One-line description of what our company does
- It is not just speech-to-text with text search added.
Why we are different from other approaches to solving the problem.
- The traditional transcript + text search is only 20% accurate on phone calls. Deepgram search is over 80%.
Details on how you are better (using cold, hard numbers).
- The better search accuracy is a result of Deepgram's AI audio search.
Give a small taste of technical “secret sauce” without going into detail. Use terms that readers can google for context.
- Search is a tool for customer data and to help train our AI models quickly.
Show how our business model complements our technical model.
- The team: Former particle physicists who built big needle-in-a-haystack searches for Dark Matter. Audio search was a side project outside of the lab, and Deepgram was started based on the tech.
Discuss our team and company origin. Be terse, but show real proof that you know what you are doing.
- Verticals: Call Centers (there are plenty of others, but companies with calls are the focus).
State your initial target market. Be specific. We could have said video editors, web search, mashup generators, eDiscovery, and podcast advertising, etc., but doing that means you'll be spread too thin to do anything. Don't try to boil the ocean.
- Over a billion hours of calls are recorded each year (and with services like Twilio, this number keeps growing by a lot).
State why the overall market for your product/service is large and growing, use brief examples.
- The calls need analytics for QA and compliance, but this is done only to a few percent of calls and mostly by humans (expensive and bad coverage).
Reiterate the expensive or painful business problem that your product or service is uniquely positioned to solve.
- Deepgram is accurate and inexpensive and the first pilot increased the customer's revenue by $500k/year.
Show how a customer has used your product and quantify the value you provide. (Use a dollar sign here. It makes the value feel more concrete—real money.)
- Fundraising: Raising Xmm on Ymm cap, with Zk committed
State the fundraising terms. (Don't use a dollar sign. This makes things feel a little more abstract.)
The overall goal is to pack as much information as possible into a few lines of text. You have to make it readable, and easily understandable with a logical flow. At all points the text should be leading the investor down a thought process:
- "What do they do?"
- "Why is it important?"
- "How are they doing it?"
- "Do they have something real now?"
- "Who are these people?"
- "Will they succeed?"
In the end, you want the investor to think: "This is promising. I'll forward it on for Due Diligence."
Conclusion: You Could Pitch Naked Too
We spent a long time agonizing over a deck we didn’t use after our first few meetings. That was time we could have devoted to building out our codebase, talking to customers, and interacting with more investors.
In the end, our decision to pitch without a deck made the whole process of raising our seed round much more streamlined, and we’d encourage any entrepreneur who’s on the fence to leave the pitch deck behind at least a few times, just to see how it feels.
One less pitch deck is one less thing to worry about. Going without reduces the chance that you lose your audience and it keeps the focus directly on the founders and key components of the company. In the end, you’ll probably find that you’re a much better communicator than you think. You can be more effective at closing because you gather and deliver information to investors “just in time”. We were.
Let me know if this helped!
Follow us to A.I. land