November 12, 2020

The Mindset of a Data Leader

Image for post

Source: AWS Summit Sydney Presentation

Data is exploding, but the current data management workflow is broken across organizations large and small. Data continues to grow at 10x every 5 years and the collection of data tools continues to proliferate. For organizations evaluating data tools, the sophisticated ones grapple with whether a solution will be able to deliver value quickly as it gets started, with the potential to scale 100x and serve across many departments for the next 10–15 years.

How do you, as a data entrepreneur cut through the noise and land your first 20 data customers? We’ve talked to 50+ data leaders and practitioners and here’s some of our tips.

The Value Proposition: What are the biggest pain points for data leaders?

  1. Data Quality & Reliability — how to alert & address model and data drift over time
  2. Data & Analytics Logging — how to make analytics more of a productivity tool, and improve cross-team consistency of logging and event emission
  3. Access Control & Permission Management — how to manage access and internal data/privacy controls in a seamless way, an aspirational “Plaid” for access control
  4. Data Lineage & Metadata Management — how to trace metadata throughout organization and maintain data observability
  5. Data Governance & Compliance — how to better address and implement data governance policies to ensure data is in compliance with evolving data and privacy standards
  6. Standardized Metrics — how to develop “source of truth” metrics or standards for data, analytics and event schema throughout the organization. Opportunity for tools that standardize communication and improve cross-team collaboration.

The Go-To-Market: How do you sell your product to data leaders?

  1. Know your Budget & Champion

Most of the budget for data tools today are embedded in engineering (where all serving costs are.) There are multiple funding models depending on the size and type of company, but usually you need an advocate on the data team.

The data market is early. Don’t be tied up with long PoCs. Work on building a strong community of early evangelists first. Land fast, deliver value early, and have a strong customer success team to expand.

In some cases, consider trial licenses (low-friction for SMBs) to initially prove your value and up-sell budget overtime instead of trying to go top-down from the VP of Engineering budget.

Try to get a champion in the organization. The most effective champions are leaders of a particular team using your product who become an internal advocate.

2. Pricing — think about the business need and ROI

Always comes back to what’s the business need you’re addressing and why your product can address it most efficiently and effectively (and cheaply!)

Be upfront about how you expect to price over time, even if subject to change. We don’t recommend pricing per seat as it provides an opportunity to customers to sometimes cut corners (e.g. cheat the number of seats.)

Consider trying usage-based pricing (base fee + usage/volume fee atop) or tiered pricing (basic, pro & enterprise levels based on feature set.)

3. Making it Work — How it fits into existing data stack & ecosystem

What are the necessary integrations your data tool needs to integrate with to work and what are the data stack profiles you are vs. are not compatible with?

Be upfront about your product’s current capabilities and your vision over time. What are the crucial partnerships to forge?

The Evaluation: How do data leaders evaluate new products?

1. Addresses key business need

Product provides concrete value to an overarching business objective in terms of ROI. Used by or impacting large number of people or systems within organization (especially those tied to revenue-generating function) are usually more compelling.

2. Most cost-efficient solution

Price is right and a “deal” compared to alternatives or the value the team gets out of the product. Data leaders are willing to pay a higher price if the alternative requires a lot of resources or sweat to build internally.

3. Easily usable

Anyone who needs to use in the organization can do so — easily accessible and integrated into existing workflow or data stack. If a complex solution, find ways to illustrate usability for key users or customers.

4. Proves value early

Start with a trial to prove value of low-friction product with smaller subset of data before full commitment.

5. Don’t say “plug-and-play”

Don’t ever say it’s plug and play. Customers won’t believe you. You lose credibility fast. Be upfront about the integration required. Good customers expect and understand it. It makes your tool more sticky as well.

Many thanks for our DataOps Slack Network for your thoughts and contributions to this article! If you’re passionate about DataOps or building something cool in the data space and want to chat more about how to sell your product, feel free to reach out to us at paul@canvas.vc, grace@canvas.vc or tmacey@boundlessnotions.com to learn more, or check out Tobias Macey’s Data Engineering Podcast!

Thanks!
Your submission has been received!
Oops! Something went wrong while submitting the form.