Why You Should Join Pavilion
Why You Should Join Pavilion
Modernizing how the government buys everything is a 2 trillion dollar opportunity.
Welcome to “Why You Should Join,” a monthly newsletter highlighting early-stage startups on track to becoming generational companies. On the first Monday of each month, we cut through the noise by recommending one startup based on thorough research, original analysis, and inside information from partnered venture firms. We go deeper than any other source to help ambitious new grads, FAANG veterans, and experienced operators find the right company to join. Sound interesting? Join the family and subscribe here:
If you enjoy our work and think others would find it useful, we’d appreciate you telling your friends! We recently launched a referral program and are giving out branded Patagonias as our first reward. Get one by generating a referral link here :)
Why You Should Join Pavilion
(Click the link to read online).
When it comes to spending taxpayer money, no one beats Turkmenistan.
In 1998, then-dictator Saparmurat Niyazov unveiled a 40ft, gold-plated statue of himself atop a 250ft rotating pedestal which moved to always face the sun. In 2020, his successor Gurbanguly Berdimuhamedow unveiled a 20ft, gold-plated statue of his favorite dog breed, also atop a giant pedestal (this time featuring a 15ft, 360-degree LED screen playing 24/7 dog footage). This past year, his successor and son Serdar commissioned a (probably giant, probably gold-plated) statue of he and his father’s favorite horse, who holds the Guinness World Record for fastest 10m on hind legs and is apparently a national hero.
Other countries should take notes.
So why doesn’t the US have giant statues of Joe Biden and his dog guarding Washington DC?
There are many reasons, but a major one are all the safeguards we have on how our government spends money.
Unlike Turkmenistan, spending taxpayer money on anything in America is a highly regulated process called public procurement. From fixing roads to purchasing new playground equipment, buying anything over a certain price (i.e. over $25,000 for goods, over $65,000 for public works in Palo Alto) requires using special contracts generated from formal bidding processes.
These formal bidding processes are complicated affairs that specialized procurement professionals are responsible for: they do market research, develop a scoring rubric, write a solicitation, publish and market that solicitation, wait for bid submissions, review and evaluate those bid submissions, handle negotiations, award the contract, and execute the final purchase. The formality of this process prevents corruption and wasteful spending, but it also means it usually takes 4 to 24 months for agencies to get what they need.
This results in major inefficiencies at every level of government.
The problem is, these inefficiencies quickly add up: public procurement is the biggest and most important buying process in the entire country.
Each year, 90,000 state and local government organizations spend 1.5 trillion dollars on contracts with private businesses, with 81% of that going through contracts generated using the process above. The US federal government spends 649 billion more, meaning public procurement in the United States totals over 2 trillion dollars annually. That’s almost 10% of our GDP.
As a result, optimizing public procurement might be the single highest-leverage action there is to help small businesses, government agencies, and the broader economy. Building a way to make procurement less lengthy and labor-intense (while keeping the same rigor and transparency) would be a unique opportunity to achieve both broad social impact and incredible financial upside — you could seriously change how every every level of government operates.
And that’s exactly what Pavilion’s done.
They’re modernizing how the government buys everything.
Background
Pavilion is revolutionizing public procurement by accelerating the adoption of a relatively new, rapidly growing procurement method called piggybacking. Piggybacking is a popular form of cooperative purchasing (anytime multiple governments work together to buy something) where instead of generating a new contract with a traditional bidding process, state and local governments use contracts other governments have already generated to make their own purchases. The city of San Francisco, for instance, might purchase forklifts by “piggybacking” off a contract Sacramento already generated instead of running their own bidding process.
As a practice, piggybacking is both faster (traditional bidding processes take 4-24 months) and cheaper (smaller governments can leverage better rates their larger counterparts have negotiated). It’s legal in all 50 states, in use by over 95% of governments, and already represents up to 19.8% of all SLED (state, local, education) procurement spend.
But piggybacking isn’t without limitations.
A major problem holding the practice back is how hard it can be to find the right contract to piggyback off of.
There’s no centralized database containing the country’s many cooperative contracts. Rather, contracts are siloed across the databases of city governments, county governments, state governments, and about two dozen intergovernmental purchasing cooperatives. As a result, procurement professionals still mostly find agreements to piggyback by emailing colleagues from other cities or posting on forums asking for help. 73% of procurement professionals still rely on “word of mouth and networking” alongside basic web searches for since there’s no “Google” for cooperative contracts.
That’s where Pavilion comes in.
Pavilion has built the world’s first search engine for cooperative contracts. They’ve acquired, parsed, and indexed over 80,000 of the country’s active cooperative contracts, packaging them into a clean, easy-to-use, and personalized search engine for procurement professionals.
Instead of calling his colleagues in neighboring cities or asking on a forum, Michael from the picture above can just search for staffing-service related contracts to piggyback on Pavilion:
This revolutionary approach to cooperative procurement creates value for governments in three ways:
It’s fast. Finding a contract with Pavilion is much faster than running your own search over forums, phone calls, and email chains. Built-in search filters allow you find contracts that meet specific requirements right away (i.e. purchasing from a CA OSDS-recognized minority or women-owned enterprise).
It’s exhaustive. Searching manually by relying on your immediate network makes it easy to miss contracts from cities, counties, or states where you don’t know anyone. Pavilion solves this; by aggregating over 80,000 contracts from 400 sources across all 50 states (a number that’s constantly growing), they’ve built the most comprehensive source out there.
It’s personalized. Each government entity has its own rules and regulations on what they can and can’t buy. Pavilion’s engine automatically tailors results around what a buyer can use (i.e. compliance) and what they’d like to use (i.e diversity goals) so they don’t have to read through the details of the contract themselves.
Better yet, Pavilion is completely free for governments and procurement professionals to use, something which has allowed them to grow adoption at an unprecedented rate. They’re not charging yet, but we don’t think they’ll have trouble coming up with a business model: the US government is the biggest buyer in the world, and Pavilion is on track to be the first stop for all of its shopping trips.
(We discuss opportunities to monetize further below).
The company was founded in 2018 as CoProcure and rebranded in 2023 as Pavilion. Since launching publicly in late 2019, they’ve absolutely taken off: usage among governments is growing 400% year over year. Almost every single government who tries the product remains active after 6 months (we’ve seen their retention metrics) and the system now drives WELL north of the $100 million in annualized GMV reported in last year’s Series A announcement (we’ve seen their engagement metrics). All of this has been achieved primarily via word of mouth/SEO and without dedicated GTM motions (which the company is now ramping up).
We verified the purchasing community’s love for Pavilion ourselves by applying for and successfully infiltrating nsite (a gated forum for purchasing professionals run by the National Institute of Governmental Purchasing) pretending to be a middle-aged purchasing assistant from New Jersey named Jack Brown. We screenshot every post we could find mentioning Pavilion, and found all of them to be positive:
The company was co-founded by Mariel Reed and Alicia Chen. Mariel was previously a Senior Innovation Strategist for the city of San Francisco focusing on procurement and talent, an early employee at Coursera where she led expansion into China and India, and the Valedictorian of her class at Georgetown. Alicia was previously the first female engineer and Head of Desktop Infrastructure at Dropbox, where she helped scale the company from 70 to over 2000 people.
To fuel their growth, Pavilion raised a 22 million dollar Series A in March of 2022 led by Forerunner Ventures. Before, they raised a 2.4 million dollar Seed round led by Neo and a 1.5 million dollar Pre-Seed round led by Leadout Capital. Other investors include Forum Ventures, Marco Zappacosta (Thumbtack), Dan Lewis (Convoy), and Katrina Lake (Stitch Fix).
Pavilion’s excellent product, strong traction, and stacked team make them an exceptional startup. But are they a category-defining business?
We think they have the potential to be.
In this piece, we’ll dig deeper to see why this is the case. We’ll look at how they’re targeting a massive, globally important market. We’ll look at how they’re well-positioned to become a leader in that market. We’ll look at how they stack up against the competition, and why their team is perfect to execute on this opportunity. And we’ll do it all as thoroughly as we can.
Ready? Let’s begin.
The Opportunity
We’ll begin our analysis of Pavilion like we do for all of our companies, from a first principle:
Large companies lead massive, growing markets.
Many companies with clear product-market fit don’t become truly massive because they don’t meet this condition. Thus, we must establish two things:
Pavilion operates in a massive, growing market.
Pavilion will become a leader in that market.
A Massive, Growing Market
Public procurement, the general category Pavilion operates in, is just about the largest market there is. State, local, and federal government agencies spend over 2 trillion dollars a year on contracts with private businesses — improving the efficiency of this process by any amount would create an incredible amount of value Pavilion could capture a slice of.
Cooperative procurement, though, the specific segment of that market Pavilion has started in, is a little tougher to quantify.
It’s hard to put an exact number on its size and what percentage of it is piggybacking versus cooperative bidding (the other form of cooperative procurement where a few governments run a normal bidding process together), but we think it’s large and growing.
Consider the following data points:
Focusing just on SLED entities, we know that out of the $1.5 trillion all 90,000 of them collectively spend on contracts, about 20%, or 300 billion, is cooperative spend. Anecdotally, we’re told there’s a roughly even split between piggybacking and cooperative bidding, meaning governments spend roughly 150 billion each year piggybacking off each other’s contracts.
Breaking things down further, we can consider how much individual cooperative contracts are worth. On average, a single cooperative contract lasts for 4.8 years and brings in $2.07 million annually, generating lifetime sales of $9.9 million. The median purchase off a cooperative contract is a whopping $128,000 and the average purchase is an even greater $541,131. Only 8% of purchases fall under $20,000. A lot of money flows through these contracts.
Breakdown of purchases by size. Source. A strong proxy for growth in piggybacking volume would be growth in the transaction volumes of national purchasing cooperatives. National purchasing cooperatives are public, non-profit, or for-profit entities that work with different governments as “agents” to help create and discover cooperative contracts. Each maintains their own collection of cooperative contracts pooled from partnered governments: when a partnered government is looking for a contract to piggyback, one option is purchasing through their cooperative. We can consider the purchasing volume these cooperatives support as a proxy for overall piggybacking volume:
Growth in CoOp-facilitated purchasing volume. Source. As we can see in the chart above, major cooperatives facilitated $29.4 billion worth of piggybacking spend in 2015, with a projection to grow to $61 billion in 2025. Cooperative contracts are getting more popular with time.
But there’s more.
Beyond general growth in the practice of piggybacking, growth for a tool like Pavilion specifically will be driven by a number of naturally aligning tailwinds:
We’re experiencing a generational shift in government workers. 30% of the government workforce will hit retirement age within the next two years. 2022 saw the most federal retirements in over a decade, with over 114,000 workers calling it quits. Governments are now putting a lot of effort into recruiting younger workers, and a big part is investing in the newer technology and tools younger generations expect. The federal government has already spun up several initiatives to drive this:
The Technology Modernization Fund was created by the Modernizing Government Technology Act of 2017 and is dedicated entirely to helping government agencies adopt new technologies.
18F, TTS, and Centers of Excellence are all internal consultancies run by the government that help fellow agencies modernize their workflows.
Younger procurement professionals expect a more Google or Amazon-style experience when searching for products, which is exactly what Pavilion provides. Tools like Pavilion are built to serve the next generation of government workers.
POV: you’ve been issued a government work laptop. Congress recently passed the Infrastructure Investment and Jobs Act, a bill which includes $550 billion of new, above-baseline spending for state and local governments to upgrade their roads, energy systems, and communications infrastructure. Here are 50 fact sheets breaking down what each individual state gets.
It’s the responsibility of state and local governments to actually deploy that capital, which means procurement professionals have their work cut out for them over the next few years. Piggybacking will be an important tool in helping governments handle the sudden flood of infrastructure-related demands and responsibilities without being overwhelmed.
Yet across all of this, the biggest factor for growth might be the arrival of a tool like Pavilion itself.
A major reason piggybacking isn’t used more is how difficult and labor-intense it can be to find a good contract. Pavilion eliminates much of this difficulty with their search engine, making it way easier to adopt an already-lucrative system.
This will almost certainly have enormous implications for the amount of total piggybacking out there. In the same way Uber dramatically expanded the market for cab rides by making it easier to call one, the very existence of Pavilion will expand the market for collaborative contracts.
Market Leadership
What does it mean for Pavilion to become a “market leader”? As a free tool, what does “market share” even represent?
In Pavilion’s case, “market share” means the proportion of purchasing professionals who use them as a regular part of their procurement workflows. Intuitively, this represents how much value Pavilion creates for the procurement industry (value they can later take a slice of).
Let’s go over how Pavilion can gain market share, how they can monetize that market share, and how they can defend that market share.
Gaining Market Share
To gain significant market share, you have to be much better than what people are currently using in some way. If you’re not “10x” better in some way, no one will bother switching over.
What does Pavilion 10x for procurement professionals?
They 10x the search experience for cooperative contracts. As we’ve mentioned, finding cooperative contracts before meant relying on forum posts, calls/emails to colleagues in neighboring cities, and relationships with different purchasing cooperatives. The process was slow, labor-intense, and ultimately a shot in the dark - there was no guarantee you’d stumble upon a contract you could use.
Compared to all that, Pavilion is:
Faster. No more calling, emailing, or asking 10 different sources. Search just one with Pavilion.
More Personalized. No more reading through contracts to see if you can use them. Pavilion tailors results around both what you can use (i.e. compliance) and what you’d like to use (i.e diversity goals).
More Thorough. Get the best deals by looking through 80,000 contracts at once instead of just the handful your colleagues or purchasing cooperative have heard of.
Pavilion’s 10x is apparent in how fast they’re gaining popularity among procurement professionals. Like we mentioned earlier, usage among governments is growing 400% year over year, almost every government who tries the product remains active after 6 months, and the system is driving way, way more than the $100 million in annualized GMV reported in last year’s Series A announcement.
They are gaining market share really fast.
But the amazing thing is they don’t have to stop at just search.
There’s an opportunity to modernize the entire procurement workflow, and Pavilion’s growing dominance at the top of the purchasing funnel (search) positions them perfectly to do it.
As their product becomes more and more ingrained in the workflows of their users, Pavilion can begin building features to optimize the rest of the procurement process: they could build payments (another major pain point for buyers), collaboration (to help procurement professionals work together), and a whole host of features for the suppliers on the other side of the table. There’s an opportunity to build a next-generation operating system for public procurement.
Search is the biggest pain point and the perfect wedge to eat the rest of the procurement workflow. Pavilion is starting with piggybacking, but the rest of the industry could be theirs to take as well.
Monetizing Market Share
Pavilion is a free tool for governments, and they’ve pledged to keep it that way forever.
This is totally fine because procurement professionals might be the most lucrative search users in the world: they’re super high intent (it’s literally their job to buy stuff) and they wield massive budgets (everything they purchase is on the scale of a city at minimum). Capturing their attention is orders of magnitude more valuable than doing the same for the average Google user.
To wit, Pavilion’s dominance in this sector gives them several paths to building a giant business:
Fees. Pavilion could take a small platform-wide commission on all purchases sourced through their platform. Alternatively, they could leverage search data to create new contracts to fill gaps in the market, taking an industry-standard 1.5%-4% commission on purchases made through these. There’s lots of room to experiment here.
Advertisements. Within search rankings, Pavilion could prioritize certain contract offerings on behalf of paying suppliers. Boring but classic.
On-Ramp. Pavilion could serve as an on-ramp for businesses looking to sell to the government for the first time, guiding companies through the cooperative contracting process and making it really easy to get in front of purchasers.
None of this is even considering all the additional features they could build on top of search. Becoming a full service “operating system” for public procurement would offer even more opportunities to monetize.
The bottom line is we’re not worried about them figuring something out. The government is the biggest buyer in the world, and Pavilion is capturing and centralizing their attention at an unprecedented rate.
Defending Market Share
To defend your market share, you need moats to stave off copycats. Let’s have a look at three of Pavilion’s:
Data moats. Most cooperative contracts are publicly available, but Pavilion’s scale and popularity has given them access to several proprietary sources of data. They can leverage these to build a more unique and valuable product:
Contract data - we’ve verified several purchasing cooperatives and government entities (including some of the largest in the country) now exclusively use Pavilion to publish their cooperative contracts (meaning you can only find them there).
Search data - beyond analyzing search trends to improve search quality and user experience, Pavilion can use it to identify gaps in the procurement market to create new contracts (exclusive to their platform) to fill those gaps.
Supplier data - as Pavilion grows, suppliers are contributing their own data to help attract buyers (i.e. areas served, specific products offered, diversity credentials). This data is often not on contracts of any kind, but helps buyers make better decisions.
Switching costs. It’s really hard to get the government to adopt new technology, but Pavilion is one of the few to have figured it out. Going forward, this will work in Pavilion’s favor against new entrants.
Technical moats. Pavilion has two: the technology itself and the team that built it.
Technology - Pavilion’s data acquisition and personalization systems are non-trivial to build. They’ve developed a state-of-the-art contract ingestion (uploading/scraping, parsing, indexing) system that continuously monitors over 400 different public and private sources. Search results are personalized for individual users based on a combination of domain expertise, platform data, and ranking algorithms.
Team - We go into this more below, but Pavilion’s team has unusual technical depth for a government-facing company. Most would-be competitors just don’t have the technical firepower to build a comparable product. Great tech employees aren’t usually attracted to govtech.
Competitive Landscape
Now that we have an understanding of Pavilion’s product and approach, let’s see how they compare against other players in the world of cooperative procurement.
Most of these aren’t direct competitors (or even technology companies) but they’re all tangentially related in some way. Future competitors would likely spawn out of one of the following categories.
Purchasing Cooperatives
National purchasing cooperatives are governmental (i.e. Sourcewell), non-profit (i.e. E&I), or for-profit (i.e. Equalis) entities that work with governments to create and find cooperative contracts. When a partnered government is looking for a contract to piggyback off of, in addition to running their own search, they can reach out to a CoOp who can help. They’re major players in helping agencies find contracts to use, but they have major differences from a tool like Pavilion:
They’re still siloed. CoOps centralize contracts across partnered governments, but are still siloed from each other: there are about two dozen major CoOps that maintain their own, separate lists of contracts that they refer partnered governments to. As a result, none have as much data as Pavilion.
They’re not technology companies. As a result, none have the technical capabilities to build a tool like Pavilion.
The biggest moat Pavilion has against them, however, is structural.
There actually has been one prior attempt to build a search engine for cooperative contracts. It was called ProcureSource, and it was built by a purchasing cooperative called U.S. Communities.
It failed.
It failed because other purchasing cooperatives didn’t like it (the tool was designed to promote U.S. Communities contracts, so other cooperatives thought it was biased), because it didn’t align incentives (the tool charged other purchasing cooperatives to list their contracts, which disincentivized sharing), and because it wasn’t built by a technology company (it was a purchasing cooperative’s side project).
U.S. Communities was subsequently acquired by OMNIA Partners, a different purchasing cooperative :(
To wit, no individual purchasing cooperative is positioned to build a search engine for cooperative contracts. You need a third party, tech-first company like Pavilion to unite them all.
E-Procurement Workflow Tools
E-procurement workflow tools (i.e. Bonfire, Periscope, OpenGov Procurement) are software tools that help in the contract creation process (i.e. managing the public bidding process discussed above). They monetize by charging government users subscription fees.
These are software products built by technology companies, but they aren’t directly competitive with Pavilion: they help you create your own contracts, but they don’t let you search through the contracts of others.
Pavilion has actually partnered with several players in this space (i.e. Bonfire, OpenGov Procurement) for distribution purposes.
Startups
The final category of potential competitors are fellow startups who also develop free tools for procurement professionals.
There are two players of note:
Procurated - Procurated is building Yelp for government suppliers: they allow public sector employees to review contractors they’ve worked with. They recently raised a $10 million Series A from Greycroft. They aren’t direct competition for Pavilion and it’s unlikely they will be: reviews are way less of a pain point than search (you have to figure out which suppliers you can even buy from before reviews matter).
CompareCoOps - CompareCoOps is a platform allowing you to search through contracts from different CoOps and reach out to them for quotes. It’s a subsidiary of VendorPanel (an Australian company focused on procurement software) which raised a $3 million Series B in 2021. Compared to Pavilion:
Their focus isn’t on search but rather on automating the quoting process. As a result, they have less personalization and significantly fewer contracts for you to search through.
You can’t try their product without going through a much longer sign-up process which contains an application that has to get approved.
Execution
Govtech isn’t something hotshot new grads usually flock to.
In fact, it’s not something anyone usually flocks to.
That’s why the team Pavilion has assembled might be their strongest weapon. The quality of people they’ve managed to recruit makes them an anomaly for govtech — they combine domain expertise with best-in class technical depth and product-building experience.
Consider a sampling of the roster for yourself:
Mariel Reed - Co-founder, CEO
Mari was previously a Senior Innovation Strategist for the city of San Francisco where she focused on procurement and talent, an early employee at Coursera where she led expansion into China and India, and the Valedictorian of her class at Georgetown.
Alicia Chen - Co-founder, CTO
Alicia was previously the first female engineer and head of Desktop Infrastructure at Dropbox, where she led a team of 30 and saw the company grow from 70 to 2000+. She holds a B.S. in CS from Stanford.
Ali Zaidi - VP Growth
Ali was formerly the Head of Growth, Scaled Business at Facebook and an Engagement Manager at McKinsey. He holds an MBA from Wharton.
Siyuan Tu - Product Design Lead
Siyuan was previously a product designer at Netflix and a Principal Experience Designer at Autodesk. She holds an M.S. in Human-Computer Interaction from CMU.
Vanessa Lauf - Head of Platform Data
Vanessa was previously VP of Operations at Kumo.AI, a Sequoia-backed Series B company. Before that, she was Chief of Staff to the CTO and a Data Science Manager at Airbnb. She previously also worked as a Transportation Planner at the SF County Transportation Authority.
Julie Tung - Software Engineer
Julie was previously a Principal Architect at Quip (Salesforce) and a Software Engineer at early Dropbox, Facebook, and Google. She holds an M.S. in CS from Stanford.
Shane Russell - Software Engineer
Shane was previously an Engineering Team Lead at the US Digital Service, a technology unit that builds custom products for different government agencies. He previously also worked as a Senior Software Engineer at One Medical.
Lillian Zhao - Growth Manger
Lillian previously worked at Bain and Silver Lake. She holds a B.A. from Dartmouth in Quantitative Social Science.
When asked about the team’s culture, Mari emphasized the team’s raw ambition and hunger to disrupt an almost century-old institution in public procurement. She also emphasized the team’s ambition to challenge norms in tech: Pavilion was founded by two women, the product and engineering teams are majority female, and everyone hired shares a genuine passion for social impact.
Mari also described the team as “unabashedly earnest” about what they do. For their first ~25 hires, they would ask candidates what their favorite song was during the interview process. Then, they’d re-record that song with personalized, procurement-themed lyrics and send that to the candidate along with their offer.
We’ve seen it ourselves: one of Pavilion’s investors showed us a Christmas video they sent to their users of the entire team dressed up in holiday sweaters singing Christmas songs with procurement-themed lyrics.
People really like working there.
The Long Term
In the short term, Pavilion has many opportunities to expand. They can collect more contracts by establishing additional partnerships with CoOps and governments. They can add additional filters and personalizations to their search. They can make it easier for governments to digitize the thousands of contracts they already have locked in file cabinets.
With this, they can cement their position as the first stop every procurement professional makes when buying anything, a position they can monetize via industry-standard commission fees or something more creative.
With this alone, they’ll have built a major business with enormous social impact: without ridiculous procurement timelines, governments everywhere will be so much more nimble and responsive.
But like we discussed above, there’s no need to stop there.
Long-term, Pavilion’s dominance at the top of the purchasing funnel (search) positions them perfectly to solve other challenges in procurement. They could build payments features, collaboration features, supply-side features, and so much more. A vertically integrated “operating system” for procurement would revolutionize the industry responsible for 10% of our annual GDP.
Public procurement hasn’t meaningfully changed since the 1990s, and it’s about time we brought it up to date.
Conclusion
America’s procurement system might be slow, but it plays a major role in preventing wasteful and frivolous spending.
We wouldn’t, after all, want to end up like Turkmenistan, who recently blew 2.3 billion on a single road and experienced inflation of 294%.
Public procurement is the most important purchasing flow in the country, and Pavilion is the first technology company giving it the attention it deserves.
They’re hiring.
If you’re a founder starting something of your own, we highly recommend applying to Neo Accelerator. The program offers generous terms, world-class mentors, and a unique residential experience. They’re accepting applications through March 15th.
In case you missed our previous pieces, check them out here:
And to make sure you don’t miss any future ones, be sure to subscribe here:
Finally, if you’re a founder or investor with a company you think we should cover please reach out to us at ericzhou@stanford.edu and uhanif@stanford.edu - we’d love to hear about it :)
Insightful post! Thank you for sharing!