why startups love Ascend cover image

Why Startups Love Ascend

Successful startups share some attributes that are commonly understood and widely published. Bill Gross shares his finding in this TechTalk, but, in general, the success criteria are:

  1. The timing and fit of the product release must be ideal for the market
  2. The team must be skilled, work well together, and execute
  3. The idea must be solid
  4. The business model must work
  5. The funding has to be sufficient to support the expected growth

Anything that distracts a startup team from managing these key attributes could become a contributing factor to its failure. Founders know that focus is key. A data strategy that is embedded in its products and constantly informs early key decisions will help maintain focus and avoid problems that lead many startups astray.

This blog illustrates how startups can eliminate distractions, make more data-driven decisions, and stay focused by building a data strategy that considers buying instead of building the right data infrastructure.

Data Distractions Undermine Startups

Most startups either deliver products in the digital space or require digital technology to compete in their market and support their customers. They need access to large volumes of good-quality data from many sources. Quite often, startups need this data even before the first release of their actual product. 

Access to the data can be technically difficult, and usually requires special processing to become useful to the product and valuable to the startups’ customers. For example, streaming product data might need to be combined with user experience logs and data from external APIs to create a training dataset for an A.I. model. Creating scalable and reusable pipelines can take lots of time and require specialized skills.

After launching with a minimal viable product, shortcuts and technical debt often grow into painful difficulties as real-world customers and market challenges come to bear. To address these problems, many startups have to reinvest in their data infrastructure by moving people and money from key product areas. Yet, the last thing a startup should do is to retool the data architecture in the middle of a massive growth spurt while onboarding new customers for its core products. 

Some typical pain points that growing startups see with their data environment are:

  • Hiring: it is difficult, expensive, and time consuming to quickly hire data engineering talent.
  • Architecture: The data architecture does not support unexpected scaling requirements.
  • Complexity: The competitive landscape creates unforeseen data challenges.
  • Speed: Business insights are too complex to produce in a timely manner.
  • Cost: Data processing is inefficient and cloud and staffing costs are higher than expected.

How can startup teams overcome these pain points and avoid wasting resources on work that doesn’t directly deliver product value?

Maintain Focus with a Data Strategy

A successful way to avoid data-related pain and related engineering distractions is to develop a data strategy that is directly aligned with the business and product strategy. Creating a solid data strategy can be a challenging process with multiple stakeholders, but it rewards the team with a playbook of how the data needs will be met. It helps answer questions like:

  1. What methods and techniques will be available to manage growing costs related to acquiring, managing, and processing data, as the product and the business scale up? How many people need to be hired?
  2. How easy will it be to integrate new data sources as the product scales, and as the team’s needs for more data evolve? 
  3. What skills and tools will be needed to draw increasingly complex insights from the data?  How easily will they plug into the overall data strategy?
  4. Anticipating the scenarios you don’t see, can’t plan for, or are unlikely to occur, how will they be handled?

Solid data strategies get the team off on the right path with confidence that the direction will sustain the startup from inception through exit. As Riley Newman, the first data scientist at Airbnb, describes, the trick is to “manage scale in a way that brings together the magic of those early days with the growing needs of the present”.

The strategy addresses how to approach people, process, and technologies in ways that are aligned with the overall startup’s mission. It clarifies the role of data in the startup as a key differentiator, defines mission critical data assets,  and supports business decision making at every step. The level of detail could be just a few key directional statements, or be as detailed as how to manage day-to-day data governance and compliance issues.

Consider Buying Instead of Building

The data strategy will ultimately drive a list of key decisions that set the future direction for the company.  Some of these decisions will be difficult and expensive to change down the road. Examples of these critical decisions include which cloud providers to use, what reference architectures to be guided by, which technology platforms and solutions to procure, what technologies to implement in house, and which data skills to hire and encourage.

One key type of decision that is often the subject of executive bias and contention among the startup team are build vs. buy decisions. For example, a more obvious decision would be whether to write your own HR system or find a vendor to support the HR functions you need. Unless the startup’s innovative product is actually a new type of HR software, it would be ridiculous to divert money and valuable time to writing a system from scratch when buying one is simple, easy, and typically less expensive. 

Among the most strategic decisions a startup makes are which technologies in the data strategy to build vs. buy. On one hand, starting small with building a custom solution to postpone major decisions and investments can seem like a good idea. Startups by nature must be fast and lean, so it feels right to get off to the races. However, skipping the data strategy process and making a poor build vs. buy decision can leave a growing startup struggling with a data architecture that doesn’t scale just when it matters the most – their product is getting traction and customers are piling on. Having to pause to retool at such a critical juncture can be a terminal event.

On the other hand, buying a mature platform fast forwards development, so a team can deliver business value immediately and eliminate the need to hire the highly specialized skills to support and maintain a custom data platform. 

So, what if startups could buy a data system just like it can buy an HR system, that scales and grows along with its data needs?

Winning the Startup Race with Data Automation

Just a few years ago, a startup setting up its data system had few options, most of which involved hiring talented data engineering types and assembling native cloud services and open source code into a bespoke platform with lots of custom software. Building software requires taking on two key challenges that devour precious capital and time. First, hiring great people who are on board with the product vision and are experts in native cloud services and other software. Second, maintaining and growing custom software features while keeping up with the pace of change of cloud providers. 

Today, startups have more options, including buying tools that are dedicated to ingesting, transforming, orchestrating, sharing, and governing many aspects of data. By far the most advanced of these options is Ascend, a data automation platform that spans all these data functions, and acts as a superpower for startups that adopt it as a central piece of their data strategy. 

Startups leveraging Ascend move much faster and more efficiently than their peers. Ascend users benefit from:

  • No need to write custom software that integrates various tools to compose a data platform
  • No need to hire specialized skill sets, since SQL and Python are first class languages
  • Retain flexibility to use any cloud (GCP, Azure, AWS) and any data engines (BigQuery, Snowflake, Databricks, Redshift, etc.) to mix and match to workloads 
  • Demand pricing keeps cost inline with the growth of the business
  • Technology proven by much larger companies using the platform at vast scale
  • Realistically start delivering data products in just a few days 

Startup teams that choose Ascend instantly have a fully featured platform at their fingertips that flexes in scale and complexity as the startup learns and grows about what data is impactful to its business and how to harness that data to shape its own future. The platform supports startups through the massive scaling that it inevitably encounters, and meets the need for access to sophisticated data features as they become relevant to the business.

 

Additional Reading and Resources

Ready to see the platform in action? Learn more about why startups choose Ascend to accelerate their growth trajectory.