“It’s been crazy out there. Venture capital has been deployed at unprecedented pace, surging 157% year-on-year globally […]. Ever higher valuations led to the creation of 136 newly-minted unicorns […] and the IPO window has been wide open, with public financings up +687%”
Well, that was…last year. Or more precisely, 15 months ago, in the MAD 2021 post, written pretty much at the top of the market, in September 2021.
Since then, of course, the long-anticipated market downturn did occur, driven by geopolitical shocks and rising inflation. Central banks started increasing interest rates, which sucked the air out of an entire world of over-inflated assets, from speculative crypto to tech stocks. Public markets tanked, the IPO window shut down, and bit by bit, the malaise trickled down to private markets – first at the growth stage, then progressively to the venture and seed markets.
We’ll talk about this new 2023 reality in the following order:
- MAD companies facing a new recessionary era
- Frozen financing markets
- Generative AI, a new financing bubble?
- M&A
MAD companies facing a new recessionary era
It’s been rough for everyone out there, and Data/AI companies certainly haven’t been immune.
Capital has gone from abundant and cheap, to scarce and expensive. Companies of all sizes in the MAD landscape have had to dramatically shift focus from growth at all costs to tight control over their expenses.
Layoff announcements have become a sad part of our daily reality. Looking at popular tracker Layoffs.fyi, many of the companies appearing on the 2023 MAD landscape have had to do layoffs, including, for a few recent examples: Snowplow, Splunk, MariaDB, Confluent, Prisma, Mapbox, Informatica, Pecan AI, Scale AI, Astronomer*, Elastic, UIPath, InfluxData, Domino Data Lab, Collibra, Fivetran, Graphcore, Mode, DataRobot, and many more (to see the full list, filter by industry, using “data”).
For a while in 2022, we were in a moment of suspended reality – public markets were tanking, but underlying company performance was holding strong, with many continuing to grow fast and beating their plans.
Over the last few months, however, overall market demand for software products has started to adjust to the new reality. The recessionary environment has been enterprise-led so far, with consumer demand holding surprisingly strong. This has not helped MAD companies much, as the overwhelming majority of companies on the landscape are B2B vendors. First to cut spending were scale-ups and other tech companies, which resulted in many Q3 and Q4 sales misses at the MAD startups that target those customers. Now, Global 2000 customers have adjusted their 2023 budgets as well.
We are now in a new normal, with a vocabulary that will echo recessions past for some, and will be a whole new muscle to build for younger folks: responsible growth, cost-control, CFO oversight, long sales cycles, pilots, ROI.
This is, also, the big return of corporate governance:
As the tide recedes, many issues that were hidden or deprioritized suddenly emerge in full force. Everyone is forced to pay a lot more attention. VCs on boards are less busy chasing the next shiny object and more focused on protecting their existing portfolio. CEOs are no longer constantly courted by obsequious potential next-round investors, instead discovering the sheer difficulty of running a startup when the next round of capital at a much higher valuation does not magically materialize every 6 to 12 months.
The MAD world certainly has not been immune to the excesses of the bull market. As an example, scandal emerged at DataRobot after it was revealed that five executives were allowed to sell $32M in stock as secondaries, forcing the CEO to resign (the company was also sued for discrimination).
The silver lining for MAD startups is that spending on data, ML and AI still remains high on the CIO’s priority list. This McKinsey study from December 2022 indicates that 63% percent of respondents say they expect their organizations’ investment in AI to increase over the next three years.
Frozen financing markets
In 2022, both public and private markets effectively shut down and 2023 is looking to be a tough year. The market will separate strong, durable data/AI companies with sustained growth and favorable cash flow dynamics from companies that have mostly been buoyed by capital, hungry for returns in a more speculative environment.
Public markets
As a “hot” category of software, public MAD companies were particularly impacted.
We are overdue for an update to our MAD Public Company Index, but overall, public data & infrastructure companies (closest proxy to our MAD companies) saw a 51% drawdown compared to the 19% decline for S&P 500 in 2022. Many of these companies traded at significant premiums in 2021 in a low interest environment. They could very well be oversold at current prices.
- Snowflake was a $89.67B market cap company at the time of our last MAD, and went on to reach a high of $122.94B in November 2021. It is currently trading at a $49.55B market cap, at the time of writing.
- Palantir was a $49.49B market cap company at the time of our last MAD, but traded at 69.89 at its peak in January 2021. It is currently trading at a $19.14B market cap, at the time of writing.
- Datadog was a $42.60B market cap company at the time of our last MAD and went on to reach a high of $61.33B in November 2021. It is currently trading at a $25.40B market cap, at the time of writing.
- MongoDB was a $30.68B market company at the time of our last MAD, and went on to reach a high of $39.03B in November 2021. It is currently trading at a $14.77B market cap, at the time of writing.
The late 2020 and 2021 IPO cohort fared even worse:
- UiPath (2021 IPO) reached a peak of $40.53B in May 2021, and currently trades at $9.04B, at the time of writing.
- Confluent (2021 IPO) reached a peak of $24.37B in November 2021, and currently trades at $7.94B, at the time of writing.
- C3 AI (2021 IPO) reached a peak of $14.05B in February 2021, and currently trades at $2.76B, at the time of writing. This includes a big recent rally: as one of the rare AI pure-play public companies, it has benefited from the explosing of interest in AI over the last few months, with its stock surging over 150% in less than two months in 2023.
- Couchbase (2021 IPO) reached a peak of $2.18B in May 2021, and currently trades at $0.74B, at the time of writing.
As to the small group of “deep tech” companies from our 2021 MAD landscape that went public, it was simply decimated. As an example, within autonomous trucking, companies like TuSimple (which did a traditional IPO), Embark Technologies (SPAC), and Aurora Innovation (SPAC) all trading near (or even below!) equity raised in the private markets.
Given market conditions, the IPO window has been shut, with little visibility on when it might re-open. Overall IPO proceeds have fallen 94% from 2021, while IPO volume sank 78% in 2022.
Interestingly, two of the very rare 2022 IPOs were MAD companies:
- Mobileye, a world leader in self-driving technologies, went public in October 2022 at a $16.7B valuation. It has more than doubled its valuation since and currently trades at a market cap of $36.17B. Intel had acquired the Israeli company for over $15B in 2018, and had originally hoped for a $50B valuation, so that IPO was considered disappointing at the time. However, because it went out at the right price, Mobileye is turning out to be a rare bright spot in an otherwise very bleak IPO landscape.
- MariaDB, an open source relational database, went public in December 2022 via SPAC. It saw its stock drop 40% on its first day of trading and now trades at a market cap of $194M (less than the total of what it had raised in private markets before going public).
It is unclear when the IPO window may open again. There is certainly tremendous pent-up demand from a number of unicorn-type private companies and their investors, but the broader financial markets will need to gain clarity around macro conditions (interest rates, inflation, geopolitical considerations) first.
Conventional wisdom is that, when IPOs become a possibility again, the biggest private companies will need to go out first to open the market.
Databricks is certainly one such candidate for the broad tech market, and will be even more impactful for the MAD category. Like many private companies, Databricks raised at high valuations, most recently at $38B in its Series H in August 2021 – a high bar given current multiples, even though its ARR is now well over $1B. While the company is reportedly beefing up its systems and processes ahead of a potential listing, CEO Ali Ghodsi expressed in numerous occasions feeling no particular urgency in going public. For an overview of the Databricks story and product, see my Conversation with Ali Ghodsi, CEO, Databricks.
Other aspiring IPO candidates on our Emerging MAD Index (also due for an update but still directionally correct) will probably have to wait for their turn.
Private markets
In private markets, this was the year of the Great VC Pullback.
Funding dramatically slowed down. In 2022, startups raised an aggregate ~$238B, a drop of 31% compared to 2021. The growth market, in particular, effectively died.
Private secondary brokers experienced a burst of activity as many shareholders tried to exit their position in startups perceived as overvalued, including many companies from the MAD landscape (ThoughtSpot, Databricks, Sourcegraph, Airtable, D2iQ, Chainalysis, H20.AI, Scale AI, Dataminr, etc):
The VC pullback came with a series of market changes that may leave companies orphaned at the time they need most support. Crossover funds, which had a particularly strong appetite for data/AI startups, have largely exited private markets, focusing on cheaper buying opportunities in public markets. Within VC firms, lots of GPs have or will be moving on, and some solo GPs may not be able (or willing) to raise another fund.
At the time of writing, the venture market is still in a state of standstill.
Many data/AI startups, perhaps even more so than their peers, raised at aggressive valuations in the hot market of the last couple of years. For data infrastructure startups with strong founders, it was pretty common to raise a $20M Series A on $80M-$100M pre-money valuation, which often meant a multiple on next year ARR of 100x or more.
The problem, of course, is that the very best public companies, such as Snowflake, Cloudflare or Datadog, trade at 12x to 18x of next year revenues (those numbers are up reflecting a recent rally at time of writing).
Startups, therefore, have a tremendous amount of growing to do to get anywhere near their most recent valuations, or face significant downrounds (or worse, no round at all). Unfortunately, this growth needs to happen in the context of a slower customer demand.
Many startups right now are sitting on solid amounts of cash, and don’t have to face their moment of reckoning by going back to the financing market just yet, but that time will inevitably happen, unless they become cash-flow positive.
Noteworthy financings (excluding Generative AI):
The first half of 2022 had a good amount of funding announcements, as those often trail the closing of the actual deal by a few months. In the second half of 2022, funding announcements slowed down to a trickle.
InfluxDB, a time series database, raised $51 million in a Series E in February 2023; Anduril, a robotics and AI defense contractor, raised $1.5B at an $8.5B valuation in December 2022; Dataiku*, a leading enterprise AI platform, raised $200M in its Series F at a $3.7B valuation in December 2022; Alation, an end-to-end data platform, raised a $123M Series E at a $1.7B valuation; Horizon Robotics, a compute platform vendor for autonomous vehicles, secured $1B in financings in October 2022; Automation Anywhere, an RPA platform raised, $200M in its latest financing in October 2022; SingleStore, which provides an in-memory database, raised an additional $30M in its Series F-2 extension, valuing the company at over $1B, in October 2022; Celonis, a process mining company raised $400M at a $13B valuation in August 2022; Anyscale, a scalable computing platform, raised an additional $99M for its Series C in August 2022; Tecton, a managed ML feature platform, raised a $100M Series C at a $850M valuation in July 2022; DataStax, a NoSQL database, raised $115M in its Series F-II at a $1.6B valuation in June 2022; Cribl, an observability startup, raised $150M in its Series D at a $2.5B valuation in May 2022; Monte Carlo, a data observability platform raised $135M in its Series D at a valuation of $1.6B in May 2022; Supabase, a Postgres-as-a-service provider, raised an $80M Series B round in May 2022; Grafana Labs, an observability platform vendor, raised a $240M Series D in April 2022; Astronomer*, a data orchestration platform based on Apache Airflow, raised a $213M Series C in March 2022; Cresta, an intelligent customer service platform, raised $80M Series C at a $1.6B valuation in March 2022; dbt Labs, an open-source data transformation platform, a $222M Series D at a $4.2B valuation in February 2022; Voltron Data, built on top of the open-source Apache Arrow, raised $88M in its Series A in February 2022; Timescale, a time-series database vendor raised $110M in its Series C at a $1B valuation in February 2022; Starburst, an analytics company built on top of Trino, raised a $250M series D at a $3.35B valuation in February 2022; Dremio, an analytics platform based on a lakehouse architecture, raised a $160M Series E at a $2B valuation in January 2022.
Generative AI, a new financing bubble?
Generative AI (see Part IV) has been the one very obvious exception to the general market doom-and-gloom – a bright light not just in the data/AI world, but in the entire tech landscape.
Particularly as the fortunes of web3/crypto started to turn, AI became the hot new thing once again – not the first time those two areas have traded places in the hype cycle:
Because Generative AI is perceived a potential “once-every-15-years” type of platform shift in the technology industry, VCs aggressively started pouring money into the space, particularly into founders that came out of research labs like OpenAI, Deepmind, Google Brain, and Facebook AI Research, with several AGI-type companies raising $100M+ in their first rounds of financing.
Generative AI is showing some signs of being a mini-bubble already. As there are comparatively few “assets” available on the market relative to investor interest, valuation is often no object when it comes to winning the deal. The market is showing signs of rapidly adjusting supply to demand, however, as countless Generative AI startups are created all of a sudden.
Noteworthy financings in Generative AI:
OpenAI received a $10B investment from Microsoft in January 2023; Runway ML, an AI-powered video editing platform, raised a $50M Series C at a $500M valuation in December 2022; ImagenAI, an AI-powered photo editing and post-production automation startup, raised $30 million in December 2022; Descript, and AI-powered media editing app, raised $50M in its Series C in November 2022; Mem, an AI-powered note-taking app, raised $23.5M in its Series A in November 2022; Jasper AI, an AI-powered copywriter, raised $125M at a $1.5B valuation in October 2022; Stability AI, the generative AI company behind Stable Diffusion, raised $101M at $1B valuation in October 2022; You, an AI-powered search engine, raised $25M in its Series A financings; Hugging Face, a repository of open source machine learning models, raised $100M in its Series C at a $1B valuation in May 2022; Inflection AI, AGI startup, raised $225M in its first round of equity financing in May 2022; Anthropic, an AI research firm, raised $580M in its Series B (investors including from SBF and Caroline Ellison!) in April 2022; Cohere, an NLP platform, raised $125M in its Series B in February 2022.
Expect a lot more of this. Cohere is reportedly in talks to raise hundreds of millions of dollars in a funding round that could value the startup at more than $6 billion
M&A
2022 was a difficult year for acquisitions, punctuated by the failed $40B acquisition of ARM by Nvidia (which would have affected the competitive landscape of everything from mobile to AI in data centers). The drawdown in the public markets, especially tech stocks, made acquisitions with any stock component more expensive compared to 2021. Late stage startups with strong balance sheets, on the other hand, generally favored reducing burn instead of making splashy acquisitions. Overall, startup exit values fell by over 90% year over year to $71.4B from $753.2B in 2021.
That said, there were several large acquisitions: Grail, a cancer detection company leveraging machine learning for cancer detection, was acquired by Illumina for $7.1B; Streamlit, a platform helps turn data scripts into sharable web apps, was acquired by Snowflake for $800M; InstaDeep, an AI decision making platform, was acquired by BioNTech for ~$682M at the start of 2023; Alteryx acquired Trifacta for $400 million; Canalyst, a data vendor for public companies, was acquired by Tegus for north of $300M. Immerok, an Apache Flink vendor, was acquired by Confluent for a reported $100M. Process Analytics Factory, a process mapping company within the Microsoft ecosystem, was acquired by Celonis (which we covered for the past several years, for a reported $100M). Leapyear, a differential privacy startup, was acquired by Snowflake for an undisclosed sum.
There were certainly a number of (presumably) small tuck-in acquisitions, a harbinger of things to come in 2023, as we expect many more of those in the year ahead. For example: HPE acquired Pachyderm; Snowflake acquired Myst; IBM acquired Databand; Airbyte acquired Grouparoo; Reddit acquired Spell ; Alphabet/DeepMind acquired Vicarious.
Private equity firms may play an outsized role in this new environment, whether on the buy or sell side.
Qlik just announced its intent to acquire Talend. This is notable because both companies are owned by Thoma Bravo, who presumably played marriage broker.
Progress also just completed its acquisition of MarkLogic, a NoSQL database provider MarkLogic for $355M. MarkLogic, rumored to have revenues “around $100M”, was owned by private equity firm Vector Capital Management.
What’s in store for 2023? We discuss consolidation mostly in Part III, because Data Infrastructure feels the most ripe for it – the last few years of frenetic company creation and funding in the space has led to very crowded categories, full of still early stage startups.
Any consolidation in the near future is likely to mostly take the form of smaller deals, including startups merging as a means of survival, at least until public companies have better visibility into when their stock prices may recover.
Large, multi-billion dollar acquisitions seem less likely for now, at least as a market trend. However, given the renewed focus on AI as a top strategic opportunity by the biggest tech companies, they’re certainly not impossible. One could imagine FAANG companies spending several billions to acquire AI companies that may not be high on revenues but have strong asset value, whether AGI focused research labs or horizontal platforms a la Hugging Face. Another scenario would be the Snowflakes or Databricks of the world acquiring enterprise AI platforms to beef up their capabilities as one-stop-shop for all things data and AI.
READ NEXT: MAD 2023, PART III: TRENDS IN DATA INFRASTRUCTURE