Venture capital investing is really the only job I’ve had (outside of a brief stint at Jamba Juice in smoothie construction). After several years in the industry, I’m grateful to have progressed and begun earning the privilege of serving founders as a board member.

Thus far, I’ve observed and participated on a number of boards at technology companies at varying points in their journey, from seed-stage to pre-IPO on a few different continents. Being a perpetual student, I have asked for plenty of advice from veterans on how to serve founders as best as I can. My hope is to do the should-haves as best I can, while thinking about could-haves — as to be the most helpful I can be to founders. I always try and push myself to operate from first principles and think creatively which I’ll touch on below.

There are plenty of great write-ups on startup board management and the core responsibilities (i.e. managing compensation, operating plan agreement, executive hiring, firing, etc.) — Mark Suster’s introductory post comes to mind — but from my perspective, board governance lies on a continuum and I believe the important principles are:

• Humility. Asking in service what is needed, rather than presuming you know.
• Being data-driven. Making observations and suggestions supported by data on an ongoing basis, not judgments from low-effort pattern recognition.
• Continuity. Understanding that the real work happens post meeting, and that the conclusions reached inside the boardroom should have continuity outside.

Below are some newer ideas that encapsulate these principles which try to bolster a healthy and productive board dynamic. These may not suit all boards, but the early results are promising, and I wanted to share them in order to start a dialogue:

1. OKRs for founders / companies, as well as for board members

I have found it useful to ask for quarterly Objectives and Key Results (OKRs) for founders as well as the company as a whole. When this exists visually as a slide in board decks, upon review I have observed some board members communicate in a hierarchical tone particularly when the company or individual misses its targets. If board members do act this way, founders may offer a rose-tinted explanatory response to shortcomings which doesn’t address core issues and therefore isn’t the most productive for change. One helpful way I’ve avoided this is to level the hierarchy and instead create goodwill by asking founders to submit OKRs to me, which I’m held accountable for.

As a board member you really don’t want to be there for the 1000+ decisions a CEO makes in a given year. Rather you want to be a sounding board and thought partner for the 5–7 strategic questions that matter most. Early on in the founder — investor relationship, board member OKRs can unearth these while at the same time highlight founder expectations, biases they may have, and / or war stories they’d like to avoid in the future. Some examples of my seed-stage and Series A companies OKRs include (but are not limited to):

• Providing feedback via email to the board deck at least three days prior to the board meeting. This ensures that the CEO will start working on the deck one week before the meeting, which is great :-)
• Introducing two qualified Product Manager candidates per quarter
• Conducting an annual review of the customer data / marketing stack with management, ensuring data-driven culture exists for product and user experience from the top-down

In general, I’ve found asking “how can I help” is one of the most important questions to posit before or after I invest. It is simplistic but being open and more egalitarian just invites transparency.

2. Request gender-based HR data in the board deck

I’ve long been interested in diversity and its impact on organizations in the technology industry. Research shows that homogenous teams are not optimal, but in order to change this in companies effectively it must be measured first. In board decks I request the CEO to provide data on gender split by function, as well as capture employee turnover by quarter. Most CEOs I speak to in Silicon Valley and elsewhere don’t have any idea of where they stand on these vectors. If diversity and related metrics aren’t central for the board, it likely won’t be something the company focuses on. It’s important that GPs and LPs know that high performance and employee retention may be the result of many things, in addition to an inclusive workforce and healthy culture. This is a no brainer and reasonably simple to do.

3. Set time for post-meeting feedback

I’ve started doing 5–10 minute calls with founders the day after the board meeting to get feedback, that often exists but gets lost in meeting fatigue. At most meetings I find there isn’t a structured space to capture the following:

• What was the least productive part of the meeting?
• Was there an area we over-spent or under-spent time?
• Was there an area you didn’t think the board would want to discuss as much as it did? Why do you think it was discussed so much?
• What is an area you thought was worth covering more that wasn’t? Why do you think it was overlooked?

Founders take on a massive task in building an early stage company. As board members we have an obligation to continue pushing our own boundaries in being helpful. To keep building on this base, I would love to hear about other discrete, new ideas (from founders too) that have been productive in these settings.

If you like what you read be sure to share on social or give the author a round of applause!

www.costanoavc.com

Spinning up an online store is easier than ever before. Tools like AWS, Shopify, and Shippo have enabled retailers to quickly scale sales online and be data-driven in every approach to online retailing.

But despite being a much older and much larger sales channel, spinning up physical retail remains incredibly difficult, and most attempts fail to utilize much data in their approach.

My family’s business is in offline retail supply chain. Since I was young I’ve seen brands across the spectrum struggle through the entire workflow of opening up retail stores. Physical retail is complex — from sourcing real estate and capital planning to outfitting stores and visual merchandising — and requires a set of competencies totally different from those needed in online retail. Not to mention the challenge after store opening of managing staff, inventory, and measuring key store KPIs like foot traffic, halo effect sales, and consumer NPS.

We are in the midst of a rebirth in retail, however. Next generation brands like Warby Parker and Bonobos are transforming from digitally native brands to true omnichannel retailers with physical stores central to their brand story and customer acquisition efforts. Physical retail offers a host of benefits to these brands. Average order values are significantly higher. Returns are lower. And offline exposure actually drives further sales online, with recent analysis from ICSC showing one physical store drives a 37% increase in overall local web traffic for brands.

The benefits of physical retail are clear, but the new competencies they require also demand a new solution.

Your Retail Stores, Delivered by Leap

The industry needs a turn-key solution that enables digitally native brands to expand retail footprints at a startup’s pace. We’re thrilled to support that need in leading Leap’s $3M seed round alongside our good friends at Equal Ventures, Brand Foundry, and select angels including Brian Sugar, who founded PopSugar and Everlane.

If you’re building a consumer brand and haven’t yet invested in retail, or perhaps are looking to expand your retail strategy further — talk to Leap. The company’s first customer, Koio, a leading Italian footwear brand, was able to deliver a premium retail experience in Chicago with Leap end-to-end in weeks (with more stores to come). Leap makes opening those stores truly turnkey: no leases, no build- out, no staffing. Leap makes it as easy to turn on retail stores as turning AWS on to develop software.

What struck us about Leap was the team’s unique skills bringing digital brands to life in physical retail. The Leap team has collectively opened thousands of successful retail stores. Leveraging experience from Lululemon, Bonobos, CustomInk, Brand Innovators Labs and Signal, and more, they have a clear vision of what it means to successfully translate an online brand into a retail store that brings the brand experience to life in the physical world and amplifies it. They’ve thought deeply about the process of setting up retail stores and identified opportunities up and down the value chain to use tech to scale.

Leap’s business model creates alignment with brands around success of stores, as measured by sales performance and quality of customer experience. Unlike a “pop-up” strategy, Leap aims to grow with their customers over time as a long-term partner in helping them launch one store to hundreds (or more) eventually — while maintaining the highest quality of brand integrity and consistency.

Jared and Amish, founders of Leap, have developed an excellent relationship, with this being their fourth venture together as partners. We’re thrilled to be working with them. Onward!

Farooq and John

www.costanoavc.com

Investing in Leap to Unlock Scale in Retail was originally published in Costanoa Ventures on Medium, where people are continuing the conversation by highlighting and responding to this story.

We are entering the age of autonomy. From robots to drones to vehicles, it is clear that autonomous robotic systems will have a significant impact on our generation and those which follow.

Our team has spent much of the past year exploring the current state of autonomous systems and seen awe-inspiring demos from companies like Boston Dynamics, Skydio, and Zoox paint a picture of what is possible.

But we’ve also seen that best-in-class autonomy today is still inaccessible to all but the most technically adept, extremely well-resourced organizations. Although they tackle different markets, it’s worth noting that the companies above have all taken a vertically integrated approach to building their products. They build almost all software and hardware themselves through a slow, complex supply chain and manage third party integrations directly.

Indeed, it is common for vertically-integrated companies to dominate nascent markets in the early days. These companies can better control quality and build software purpose-built to their new needs. Apple and the iPhone is a classic example. But as markets mature, horizontal software and hardware purpose-built for these new industries emerge — like Android, which is now a more popular OS by volume and hardware form factor reach.

As drones have expanded beyond their hobbyist origins to become critical inputs of daily business operations, enterprises need an auditable, open-source, world-class toolset for their drone applications. They need high durability, long battery life and hardware agnosticism — as well as cyber-secure, tested integrations to solve end user needs ‘out of the box’.

We’ve surveyed the opportunities in autonomy and looked for markets where a horizontal, operating system-level platform opportunity should exist. The commercial and military drone segment is quietly exploding — forecasted to be over $85b+ in aggregate sales by 2020. We believe there needs to be an open-source operating system to hasten progress in this segment. Commercial, enterprise functionality will be necessary for this open-source operating system because basic services should not be controlled by vertically-integrated, monopolistic providers in walled gardens. The autonomous robotics world deserves a Red Hat for its Linux to help drone manufacturers and service providers accomplish more.

An Open, Inclusive Approach

In this light, I’m very proud to announce our recent Seed investment in Auterion, the leading open-source operating system enabling commercial and military drones and other autonomous systems. Auterion, co-founded by Lorenz Meier and Kevin Sartori, supports the core development in PX4, a hardware agnostic, open-source drone software (created by Lorenz during his PhD at ETH Zurich 10 years ago — a ‘labor of love’ as he describes it). The popularity and technical community of contributors to PX4 is world class, with a team of over 3,500 contributors globally and 20% of the drone market already using the system.

Using Auterion, manufacturers and providers can focus on their customer use case. This may be improving GPS location specificity, optimizing video-link, developing advanced computer-vision based object avoidance, or variable payload management. At the same time, those enterprises can rely on technical innovation from best-in-class partners throughout the software stack, including — but not limited to — 3D Robotics, Airmap, Intel, NXP, Sony, and STMicroelectronics. Several fast-growing enterprise drone providers have pursued and differentiated their offerings with this approach. Yuneec, PrecisionHawk and Zipline are great examples — and they all use PX4 and / or Auterion for a commercial license.

We weren’t just inspired by an opportunity to help build a platform business which will help safely and compliantly deliver the promise of autonomy for military and commercial organizations alike. We are also eager to support Lorenz and Kevin, two technical, highly strategic thinkers out of ETH Zurich who have loved their craft since they were young. I’m thrilled to be doing so alongside some of our favorite European investors, Lakestar and Mosaic Ventures, to help the talented Auterion team build an enduring business.

www.costanoavc.com

Furthering Innovation In Autonomy With Auterion was originally published in Costanoa Ventures on Medium, where people are continuing the conversation by highlighting and responding to this story.

Since the early applications in the 1960’s, Machine Intelligence has had many names attached to it, including data science, adaptive models, and symbolic reasoning. But whatever the nomenclature du jour over the decades, this much is true nowadays: ML is one of the most effective tool sets that a company can use to build and maintain a competitive moat.

We’ve never been more excited for the state of this technology, which has come to be for several reasons:

Academic Research Wins

• Multi-agent Reinforcement Learning: Researchers from DeepMind, UMontreal, and Cambridge (amongst others!) are building successful applications using multi-agent reinforcement learning frameworks, which should provide opportunity to learn from more complex, real world or simulated (increasingly real world-like) environments
• Transfer Learning: We are beginning to see positive results in transfer learning in narrow cases which we believe is an incremental step toward pervasive enterprise intelligence, applicable in even highly regulated industries such as Healthcare and Insurance
• Multi-modal, Sparse Data Utility: Groups are reducing time-to-value and automating pre-processing of multi-modal training data including numerics, image and text into related or single models

Data And Infrastructure For Now

• More Than Moore’s Law: Moore’s Law, which states roughly that the number of transistors that can go on an integrated circuit doubles every couple of years, is more than half a century old. Amid concerns it may be nearing the end of its natural life. Many technical teams are pushing limits to improve hardware architecture from GPU’s to TPU’s and other ASIC chips built for AI-specific use cases. Relatedly, running federated models directly on chip / device is helpful and particularly important for low latency decision making and data security
• Data Generation Democracy: Training data, which is getting created at an exponential rate, can now be used in large scale simulations using tools such as Improbable, as well as generated using a gambit of open source or fit-for-purpose providers, such as Parallel Domain
• People Playing Nicely: Software tools such as Dataiku, Periscope, Alation and others are enabling better and faster software engineering and data science workflows (pre-processing data, evaluating and deploying models, monitoring, governance, etc.)

Commercial Readiness

• The Data Company: C-level execs see success directly linked to data-driven experimentation. Every company isn’t just a technology company anymore — now they must be ‘data’ companies where optimized data flow and dataset normalization become major inputs to operational improvement
• HR Demand: The two most widely posted jobs on Linkedin are ML Engineer and Data Scientist, respectively. More importantly, both roles aren’t any longer siloed to specific departments, like marketing or pricing. Rather, they are now considered key to the success of nearly all departments and software deployments
• Automation Culture: Investments are being made in automation and RPA across the enterprise stack, and employed frequently by even non-technical users

Costanoa Ventures has historically had particular affinity toward computer vision applications. In the past five years, the technical capability has progressed significantly — particularly in image recognition, which now performs at super human-level in certain domains. If you feed enough labeled and formatted image data through well-designed deep net systems, these systems exhibit powerful image recognition capabilities. Perception, however, is far from being solved. Issues remain including the binding problem, applicability of transfer learning on new datasets, the understanding of contextual information, and iteratively reducing false positives, just to name a few. However, it is important to note that these advances are already creating value today in everything from inspection tasks (PreNav, Tractable), to retail automation (FocalSystems*) to continuous aquaculture monitoring (Aquabyte*). We’re particularly interested in connecting with talented technologists working on achieving contextual understanding from video, which should enable better search experiences for consumers and users in the enterprise.

Some argue that the ability to predict is the essence of intelligence. In the context of the enterprise, prediction is powerful in that it fundamentally enables better resource allocation — from people to pricing to product iterations. Rapid institutional learning drives more competitive enterprises, which is near and dear to every startup founder and investor’s heart. We’d love to help build companies in other subfields of applied machine intelligence as well as be participants in the infrastructure shifts that serve it.

Please join us for a tech talk to discuss Uber’s ML-as-a-service platform on May 17.

As perpetual students of technology, we at Costanoa leverage and learn from a deep network of technologists in academia as well as commercial organizations large and small, who are pioneering on both the infrastructure and application level. We’re excited to announce that we’ve invited Jeremy Hermann to our offices on May 17th to speak about Michelangelo: Uber’s Machine Learning Platform. Jeremy will discuss Uber’s ML-as-a-service platform and how they designed it to cover the end-to-end ML workflow: manage data, train, evaluate, and deploy models, make predictions, monitor predictions and support traditional ML models, time series forecasting, and deep learning. He will also talk about the challenges and solutions that drove their approach as well as what they continue to learn. Register: http://bit.ly/2qoQCdg.

*Note: Costanoa is an investor in Focal Systems, Alation and Aquabyte.

www.costanoavc.com

Exploring Enterprise ML was originally published in Costanoa Ventures on Medium, where people are continuing the conversation by highlighting and responding to this story.