Placer.ai Series C - Why Placer Raised $100M

We, the founding team, always loved data - ideating around it, engineering with it, understanding the world better with it. 

But what captivated us most was imagining data products that can be used by tens of thousands of businesses across the world.

Among all the ideas and visions we bounced around before starting the company, one stood out for its simplicity and potential impact - building a ‘Physical Market Intelligence Platform’ to provide everyone in the offline world (a.k.a the ‘real world’) with aggregate insights for decision-making. Or in layman’s terms, “a dashboard to get instant insights for any place to understand its audience, surroundings, and competition”.

In 2016, the Placer founding team gathered in a basement and spent a weekend sketching out a plan to turn this idea into a massive world-class data company. 

Why did we get so excited?

1. We loved using insight tools like SimilarWeb and App Annie that were made for the digital world.
2. A massive market - 80-90% of spend is offline and is not going anywhere, anytime soon. We did not believe in the ‘retail apocalypse’ narrative.
3. An industry ‘flying blind’ - this immense offline world has suffered from a lack of information critical to its decision-making.
4. Data is especially critical for the physical world. The famous Facebook motto “move fast and break things” (which we practice at Placer) does not work well in the physical world. Brick & mortar decisions are costly and irreversible. It also takes a LONG time to understand you’ve made a mistake.
5. Market Research is aggregated data - no need for any personal identifiable information (PII). This means we could build a privacy-first company, without PII data challenges.
6. It’s a hard problem - which presents the opportunity to build something special. And in hindsight it’s been 10x harder than we thought!

Whiteboarding without customers or tech debt is fun!!!

The more paper we stuck to that basement wall, the bigger the vision became! Everything is possible with the stroke of a pen… 

But very quickly, we hit some glaring challenges:

• The platform had to be about answering key business questions. But to generate the BEST reports that do so, there are 100s of relevant datasets that we MUST aggregate. 
• The retail ecosystem is DIVERSE - retailers, CRE, CPG, travel, hotels, billboards are all unique worlds in and of themselves. Can we build a platform that reflects this? 
• And…growing up in a “digital bubble” - the founding team knew VERY LITTLE about the retail world, its major players and how they work.

The best way to approach a big challenge is breaking it down into smaller ones. So we worked hard to define Phase 1 - focusing on building a product that (1) was centered around the mobile location analytics dataset and (2) generated reports tailored for CRE and retail. 

5 years and 5 funding rounds later, we’re FINALLY feeling “pretty good” about Phase 1: we launched a world-class mobile analytics product that’s used by over 1,000 customers, and thousands more are using our free products. 

But it’s also been “frustrating”  - we were always strapped for cash and resources. We’re yet to integrate most of the datasets we need; key reports for certain verticals remain in the product pipeline; and in terms of usability and workflow features, we still have a lot to do in order to create a truly comprehensive platform (vs “read only” status insights tool).

That’s why the $100M Series C funding we just announced is so momentous for me and the rest of the Placer team. It finally removes the shackles and equips us with the tools and materials we need for Phase 2 - rapidly building the full Placer.ai Market Intelligence Platform.

So let’s dive into what that means…

How does it work?

A Physical Market Intelligence Platform is a big data puzzle. Piecing it together - in a nutshell - consists of four phases:

1. The Ingredients - identifying and assembling the data.
2. Ingestion - processing and aggregating that information.
3. Delivery - making it presentable and accessible.
4. Customizations - every vertical is seemingly interested in very similar data, but with a different lens. This requires nuanced packaging around information density, terminology, order of reports, and 3rd party data-sets.

Ingredients

A vast amount of interconnected data is required to create a truly accurate and complete picture of what’s going on at a location. This data falls into two broad categories:

• Point of interest (POI) data offering information on places such as a grocery store, retail centers and wider areas.
• Geospatial data such as impactful events in the area, traffic data and future development projects.

Now consider all things you see going on in the world and imagine how POI and geospatial data can capture and quantify them…

Here’s a snippet:

We track dozens of data categories and thousands of datasets and vendors in order to identify new data that can help answer our customers’ questions.

• Our product team draws on our customers’ feedback and wider market research to identify and triage the datasets we need to answer the questions. 
• Our BD team lines up commercial partnerships with the data providers.
• Our data analysts and scientists carry out a lengthy quality assessment process, which includes testing the data’s relevance, accuracy, data trust compliance, coverage, compatibility, recency, accessibility and alternatives.

This is 50% of our work and is a huge data challenge - but also great fun!

Through partnerships and our App Marketplace, we’ve recently integrated online reviews, credit card data, demographics, vehicle traffic volume, crime figures and planned construction into our platform. And we have lots more datasets in our pipeline: retail sales, property sales, financial data, leasing comparisons and climate data to name just a few.

Ingestion

If the data are the ingredients, then ingestion is the cooking. This includes complex data science processes:

• Anonymization - eliminating personal identifiable information
• Normalization - adapting the data’s various fields to fit Placer’s data model
• Cleansing - ensuring that the data is as accurate and complete as possible
• Enrichment - adding existing data layers to the ingested data, or extrapolating information from it
• Tagging - associating the data with relevant POIs, industry categories, and so on to create meaningful insights.

Tagging data to POIs is a massive task. Placer’s POI database contains millions of entities: a commercial real estate asset in a customer’s portfolio; stores of a retailer’s chain or that hold a CPG brand’s products; a billboard used for out-of-home advertising; a downtown area being regenerated by a municipality or business improvement district. We geofence each one so data can be tagged to it.

But a much greater complexity than the volume of data-POI matching is the fact that our data structure is mutable - it changes. Stores, restaurants, strip malls and other POIs open, close, merge and move. Our physical environment is constantly changing. One of our platform’s standout attributes is that it always reflects historical change.

In practice, this means that, for each POI change, we not only adjust our data tagging but also re-tag 5 years of historical data to ensure any historical comparisons are “like with like”. This is a huge investment of resources on the part of our data science, devops and engineering teams - exponentially increasing our data management burden.

Delivery

To complete the cooking metaphor, after selecting ingredients (datasets) and cooking them (data ingestion), we then lay out a buffet-style feast of solutions for our users:

Basic Reports and Insights

The most basic level of the platform is converting the data into real-world constructs that can be understood by industry professionals: tables, charts, maps and other graphics displaying cross shopping, trade areas (below), cannibalization, risk analysis, visit frequency and so on.  

Solutions 

A key tenet of the Market Intelligence Platform is the approach that insights like those are often not the answer to the questions that our customers are looking for. Rather, they are just part of the explanation behind the answer. That means providing a comprehensive suite of Solutions SUPPORTED by insights, not just a library of uncontextualized insights. 

An excellent example of this is Void Analysis. A key question for retail real estate is “who is my ideal tenant?” While our platform offered important insights (such as retailers’ average monthly foot traffic and cannibalization) for reaching an answer, landlords were doing a lot of legwork. The Void Analysis tool we released late last year enables CRE professionals to instantly analyze thousands of potential tenants through automatically generated reports that include ranking according to our unique Relative Fit Score. This significantly improves the speed and scope of a search for new tenants.

We are now working on the many additional solutions like Void Analysis in our development pipeline - sales forecasting, site selection for retail chains, market selection, market change reports, product optimization for CPG to name a few.

Let’s build it!

To our current understanding, we’re just “5%” of the way to our Market Intelligence Platform vision. The remaining 95% will be built by scaling POI coverage, datasets, answering more questions and developing the other core components of the platform.

So our focus now is on ramping up the velocity of this development. And to do that, we need even more of the world’s best talent across the company. 

So, during 2022, we will use our new capital to double the size of our engineering team and significantly expand the data at our disposal. In parallel, we will also channel more resources to supporting our customers and contributing to industry understanding through our analytical research department and educational content. 

Placer.ai is committed to transforming the way real-world businesses make decisions. And we don’t want to waste any time going about it.