Enterprise sales in high gear
Gil Dibner

Beyond TTDD. Over the past year, the scale of the AI-powered vertical application software opportunity has become increasingly clear. We continue to meet companies with impressive early growth tracks, and we continue to hear stories of software companies - particularly in the US but in Europe and Israel as well - that are posting very impressive early revenue numbers. As Ed Sim pointed out recently, the “triple-triple-double-double” (TTDD) revenue trajectory that defined successful enterprise startup revenue trajectories for the past decade is being re-evaluated in the face of AI application companies that seem to be growing much faster. The new upper bound seems to be two years of 10x revenue growth (10x10x), but even somewhat more modest pathways such as Ed’s proposed “quintuple-quadruple-triple-double” (QQTD) are more frequently observed than previously. TTDD get you from $1M to $36M in four years. QQTD gets you from $1M to $120M in four years. 10x10x gets you to $100M in just two years. The list of companies that are apparently achieving results on the upper end of this scale is longer than ever before: Bolt, Cursor, Together.ai, 11x, Eleven Labs, Character.ai, Sublime, Wiz, and several more.

Observations and questions. These types of growth rates are absolutely happening and they have impacted the startup ecosystem already. They also raise some questions for which we do not yet have good answers.

• The bar is being raised for VC. Many venture investors we talk to are just not interested in stories that are showing less than 5-10x revenue growth rates when revenues are at the $1M level. Fair enough, but this does create an unattainably high bar for many companies that are good companies with strong fundamentals. This is particularly challenging for companies with a longer sales cycle (or any sales cycle, see below.). Regardless of how frequent these hyper-growers are, founders must be realistic about the fundraising climate and what VCs are hoping to see. The idea that $1M ARR guarantees a Series A round is pretty much dead right now. VCs need to see a path to $10M ARR in 12-18 months and a growth rate of 4-5X for a Series A round fundraising to be anything other than very hard.

• AI excitement is supercharging adoption. Enterprise buyers began 2024 with a tremendous appetite for adopting AI-first products across their organizations. Many of them were experimenting with this in 2023 as well. There is widespread understanding of the potential impact of AI on nearly all aspects of work. Everyone (including our parents and children) has played with ChatGPT or Gemini or StableDiffusion. We are all aware of this, and - if anything - our collective expectations of what AI can do for our companies in the immediate term are probably over-inflated. But this is leading to a very willing buying environment.

• Employee replacement has led to greater enterprise sales surface area than ever before. It’s very hard to generalize, but the way many of these products are priced and bought appears to be as a replacement for employees. AI products are increasingly speaking in the language of “agentic augmentation” of employees. If the tool makes employees in a given function 10-90% more effective, it can easily be argued that it allows the customer to avoid hiring and training at least one employee. The starting price for a lot of these tools begins around $100K and is seen as the equivalent of an entry-level employee. The economic buyer (which can be anyone with hiring authority in the organization) faces a simple choice: they can hire yet another human employee or they can buy an AI product which will enhance the entire team, reduce their management and training burden, and put them on the cutting edge of technology making them a hero in the organization to senior management which is eager to talk about AI. The choice is an easy one for most - and thus the sales surface area for these AI-first tools is simply enormous.

• These tools are bought not sold. Consequently, it can be challenging to talk in terms of “sales,” “sales process,” and “sales efficiency” in relation to many of these companies. Many of these products are being ripped out of the hands of the companies that are building them. The sales force is largely taking orders and trying to ensure customer success - but they are not forced (yet) to crack complex sales processes, overcome objections, and deal with competitive pressures. That will come, but its not a big factor right now.

• AI has eliminated technical integration challenges. One reason revenue growth is in high gear is that LLMs are particularly good at ingesting unstructured data and contributing to unstructured processes. That removed a critical friction and speeds things up massively.

• Sales durability is a massive question mark. Undoubtedly, many of these hyper-growers will evolve into successful and sustainable businesses with durable revenues and customer stickiness. But undoubtedly, many will not. It is still quite early in this wave, and we don’t yet know how revenue durability will break for many of these companies. I speak with a lot of people with pretty good information, and opinions are sharply divided even on specific companies. One person will tell me that Company X is the fastest growing company in their category; another will point out that churn is massively high; a third will point out that there is no product; and a fourth will insist that users love the product. Time will tell. The next two years are going to be really interesting.

• Good conditions for the top 20 Series A firms; seed is (as usual) harder. At Angular, we invest first (in inception rounds) when there is rarely revenue. In the 10x10x/QQTD world, this is a particularly dangerous place to invest in most categories. The Series A firms have the luxury of waiting to see the revenue trajectory around the $1-2M mark, and can snap up the 10x10x/QQTD hyper-growers. For them, there is little incentive to fund before that trajectory is clear - and zero incentive to invest in 2-3x growers at the $1M level if they believe they will find others that will grow faster. These conditions are actually optimal for the top 20 Series A VCs in the world who can reliably access and win these opportunities. For any Series A firm not in the top 20 (and this would include most of the Europe or Israeli series A firms), the situation is darker. As a seed fund, we need to focus on opportunities where we believe we can predict large outcomes at the inception stage: sometimes that’s going to be 10x10x/TTDD-style growth, but not always. We will continue to focus on backing deep technical defensibility that will drive long-term growth, breakeven potential, and breakout revenue growth down the line.

• Product appears to be the missing ingredient. Across the range of early-stage companies we are meeting - including those with very impressive early revenue traction - the missing ingredient appears to be robust product design. If I had to bet on what will separate winners from losers over the long-run, it would be a range of things all related to product rigor and depth. The double-edge sword of LLMs and GenAI is that they make it very easy to build a product that is 80% of the way to great, but getting to truly great remains very difficult - maybe more difficult than previously. In a world where nearly anyone can deliver an 80% good solution to customers, I believe the best strategy for a seed fund is to identify situations where founders are obsessed with delivering a product that can get to 100% - even if it takes longer and even if the initial growth looks a bit shakier.

For a lot of reasons outlined above, enterprise sales for many AI-first vertical application companies appear to be happening in high-gear. Like a car speeding downhill on a smooth highway, the transmission is in high gear and everything is spinning super fast. Customers are adopting so fast that even high churn is not a problem. It’s easy to overlook competition, product problems, or sales inefficiencies - because all of these just disappear into the blur caused by the overwhelming speed of the growth engine. But when a car in very high gear encounters any disturbances, the transmission can begin to slip - sometimes with catastrophic or unexpected results. I am not at all arguing that all this growth is fake. It’s very very real - but it’s an entirely new world of revenue growth and sales that founders and VCs are now navigating. Knowing when to downshift - and knowing to avoid upshifting too early - may be the key to survival and scale for some of us. Some companies were born to run in high gear from day one. Some companies should start out in first gear because they need to climb uphill on the dirt road for a while before they get on the highway to scale. Knowing which is which will make all the difference.

—

A quick note on this newsletter. This is our 250th edition. We’re excited by the large and growing community that this newsletter has built. This will also be our last edition for 2024, but we’ll be back as usual in early January. In the meantime, have a wonderful holiday. To all the founders and early employees of high-risk businesses: remember to do your best to disengage a bit if you can. See you in 2025!

Am I Thinking About AI the Right Way?
Gil shares the four AI themes and questions he's thinking about.

The Venture Apocalypse
The venture world is deeply debating its future, but core principles remain unchanged.

No Sleepwalk to Success
Engineering success in a technical startup.

Revenue Durability in the LLM World
Everything about LLMs seems to make revenue durability more challenging than ever.

More blogs

ENTERPRISE/TECH NEWS

Running out of data. A recent piece in Nature discussed a pressing AI challenge: AI developers are running out of quality internet data to train LLMs. “Researchers at Epoch AI, a virtual research institute, projected that, by around 2028, the typical size of data set used to train an AI model will reach the same size as the total estimated stock of public online text. In other words, AI is likely to run out of training data in about four years’ time.” How are the large AI companies, like OpenAI and Anthropic, addressing this issue? A OpenAI spokesperson told Nature: “We use numerous sources, including publicly available data and partnerships for non-public data, synthetic data generation and data from AI trainers.”” Another possibility is that LLMs won’t actually need more data. “It’s possible that LLMs, having read most of the Internet, no longer need more data to get smarter. Andy Zou, a graduate student at Carnegie Mellon University in Pittsburgh, Pennsylvania, who studies AI security, says that advances might soon come through self-reflection by an AI. “Now it’s got a foundational knowledge base, that’s probably greater than any single person could have,” says Zou, meaning it just needs to sit and think. “I think we’re probably pretty close to that point.”

AI in the real world. Two new reports, a research paper from Claude and a new Goldman Sachs report are sharing data on how AI is actually being used in the real world. Currently only 6.1% of companies are using AI, up from 5.9% last quarter. However, the companies are reporting productivity gains of close to 30%. The most popular use case is web and mobile app development. Additionally, the finance and insurance industries have seen the fastest AI usage growth and SMB adoption has doubled year over year.

HOW TO STARTUP

CEOs calling it quits. Jason Lemkin highlights a new trend of founder CEOs quitting their startups. “Quitting Culture is a real thing, it’s all over Tik Tok, it’s become endemic in sales teams in particular. Part of it is blaming someone else for your own shortcomings. Blaming the VCs, the CEO, the markets, the “downturn”, anything but yourself. So I recently noted that 2 founder CEOs I’ve known for years that raised $50m+ each just … aren’t CEOs anymore. They weren’t forced out by their VCs, at least not directly. They just had enough. One is doing YouTubes. The other seed investing. Hooray. There’s also a most invisible version. Two CEOs I invested in basically just gave their startups to other companies, even with many millions left in the bank. So they could just give up. Perhaps this is just part of the normalization of start-ups. The only thing I can tell you, as someone who has thought about quitting a few times but never did … is you can often fight your way back.”

Opening a new office. Ludmila Nikitina, Latvia country manager for Estonia-based fintech Wallester, shared her top tips for opening an office in a new location. The entire post is worth reading for founders looking to expand, but this section in particular is key: “Pull in HQ. The local team can’t — and shouldn’t — do everything alone. Lean on the established processes, expertise and experience that you have across the company. You need strong backing to get moving quickly and efficiently — not siloed teams trying to figure out a new office on their own like a brand new startup. Our experience from other expansions meant that we could plan better and minimise logistical challenges. You’ll want clear communication channels and regular check-ins for the most effective support.”

HOW TO VENTURE

Picking a lane. In Ed Sim’s latest newsletter, he reflects on the current state of the venture capital industry, specifically the two extremes of seed rounds and the implications for small seed funds. "Either you have first-time founders who want to raise as little as possible to get started […] Or you have second- and third-time founders raising, well, as much as they can to an extent - let’s call this the jumbo-ification of Inception with rounds >$10M and upwards of $20M+. Smaller seed firms need to clearly pick a swim lane- do I compete on small rounds only or go head-to-head with the multistage firms leading these $8-10-20M rounds? Well, you know my answer - if you have the will and brand and experience, then by all means you should compete with the larger funds and back those second and third timers…as long as you think the valuation premium at Inception leads to that elusive, HUGE multi-fund returning outcome."

The a16z vs Benchmark approach. The New York Times recently spotlighted the two opposing strategies of venture capital, the paths taken by Benchmark and a16z. Benchmark, since its inception, has stuck to the traditional playbook of venture capital: “It would write tiny checks to invest in private technology companies and help them succeed with guidance and connections. It would resist the urge to become bigger over time.” Meanwhile, a16z has expanded in every direction, from raising massive funds, to hiring 80 investment partners, to creating a media empire with several newsletters and podcasts. Many funds have since copied the a16z approach and hired teams to help their portfolio companies with marketing, recruiting, and other services. While staying smaller as venture capital fund is risky, as “the profits have to come from the investments, not the fees”, the track record of sticking to the basics of venture capital, as we do at Angular Ventures, can be unmatched. “Benchmark’s first eight funds, which were raised and invested from 1995 to 2019, generated returns of more than 7.5 times the money invested, after deducting fees and profits taken by the fund”.

Aquant was named to Fast Company’s list of the Next Big Things in Tech.

Beebop’s Olivier Deckers shares the engineering practices driving Beebop’s success in energy innovation and enabling them to deploy Europe’s largest virtual power plant.

Aquant
Senior Enterprise Account Executive (NYC)

Firebolt
Software Engineer, Query Planner (Munich)

SOURCIX
FS Engineer – Frontend (Tel Aviv)

Tensorleap
Algorithm Developer (Tel Aviv)

More jobs