Build Early And For Tomorrow
Introduction
Over our five years in business, we noticed two interesting trends.
The first is that a lot of the startups that we have helped have second-time founders. These are people who have typically built up and sold their business and now, armed with that experience, understand the power of data and how it can be leveraged to gain a competitive edge.
The second is that while you might be familiar with the likes of Numan, Treatwell, MUBI, and Plend today, they were not big names when we first worked with them! By utilising data from the beginning, they were able to make informed decisions, optimise their operations, and scale effectively. Their growth stories exemplify the transformative power of data.
For us, this is a big indication of how powerful data can be and the competitive advantage it gives you. Data-driven decision-making is not just a buzzword; it’s a proven strategy that can propel startups to success. So why isn’t every startup simply starting with data from day 1?
The Cost Barrier
For most startups, finances are precious and the upfront costs of setting up data infrastructure simply aren’t palatable. Building a robust data infrastructure requires significant initial investment in technology, tools, and expertise. Let’s delve deeper into each of these elements:
Technology Costs: A data stack typically includes various components such as data storage, EL (Extract, Load) tools, Visualisation tools, and possibly AI/ML components. Here are some typical cost considerations for each component:
Cloud Data Warehouses – Pricing is usually based on data storage (per TB per month) and data processing (per query).
Traditional Data Warehouses – Typically involves significant upfront costs for hardware and software licences, with ongoing maintenance and support costs.
EL Tools – Pricing is often centered around the volume of data that you are extracting and the frequency. Where you are creating a custom script, there will also be associated costs and resources needed for maintenance.
Visualisation Tools – Costs are usually subscription-based per user per month, with additional costs for premium features or higher user tiers.
AI/ML Components – Costs can include model training hours, inference costs, and storage of training data.
People Cost: When you are first building out your data stack, and you have low data volumes, you might be surprised at how low the tooling can actually be. The cost that is almost unavoidable however is in getting the right people to set all of this infrastructure up.
According to Job site Indeed these were the average salaries in London for 2023 for the following roles:
Data Architect – £101,237
Data Engineer – £85,484
Data Analyst – £60,246
We can see here that access to quality data people is therefore a major challenge for small or medium sized businesses. So how can they address this?
The Skill Barrier
Even for established businesses, the cost of hiring data people is high, and with that high cost also comes a high risk of hiring the wrong person, especially where that skill isn’t present in the business or they might not fully understand what is needed, and so how can you mitigate that risk?
Use Excel: Traditional reporting still has a place in most businesses. Whilst obviously excel has some limitations in terms of manual input leading to errors, and a latency wherever people are having to update spreadsheets – excel is still a great tool for reporting that nearly everyone knows how to use.
Use Off-The-Shelf: There are many ETL tools that allow you to get started with data quickly and without heavy technical knowledge. These platforms are aimed at business users and can help you centralise and visualise your key metrics from popular data sources. Whilst often these tools don’t scale well – we would advise that they are a great way of getting value from your data in your early stages.
Don’t Hire A Unicorn: A mistake that we often see growing businesses make is that they try to hire one amazing person to handle their entire pipeline. The problem with this is that building a data stack requires a team with diverse skills and knowledge. Data engineers are needed to design and maintain the infrastructure, data analysts to interpret and visualise the data, and data scientists to develop advanced models and algorithms. A multidisciplinary team brings a well-rounded perspective and expertise, ensuring a robust and effective data infrastructure.
What we often see happening is that businesses will try and hire a unicorn, someone with a bit of all-round experience. They get the stack up and running, but often when data volumes start to scale, what they’ve built has not been optimised for this and so the cost increases significantly and the stack often has to be completely redesigned.
The Complexity Barrier
Each different element of your data stack will have different considerations in order to ensure it is set up for optimal performance, and this leads to a high level of complexity in set up. There are often business requirements that, on the surface, seem simple, but in practice may have huge ramifications later on. A great example of this would be ‘Real-time’ analytics. Nearly every stakeholder would tell you they want or need this, but might be unaware of the significant complexity and costs they would be adding to your analytics infrastructure.
It is only really when someone has extensively used a particular tool that they will know it inside and out and all of the considerations in optimising performance. What this most often means is that data professionals will recommend a stack that they are most comfortable with – but in some cases this may not align closely to the needs of the business.
Perhaps the biggest element which adds complexity is scalability. Data is such a fast-moving world with so many different tools that it can be difficult to understand the implications as to whether that stack will still be useful in a few years time. Embedding the best practices around documentation, quality and governance to ensure this healthy growth requires a senior data person who has done it before, but here we come back again to our cost barrier in attracting this talent.
When To Make The Leap?
As businesses grow and expand, their data strategy plays a critical role in driving success and unlocking growth opportunities.
If your company is generating $500k – $1m annually: At this stage, companies are just starting their journey, and the primary focus is on quick and cost-efficient solutions. Off-The-Shelf tools provide a convenient option for getting started with data management.
If your company is generating $2m annually: As businesses scale and revenue increases, the limitations of Off-The-Shelf tools become more apparent. At this stage, businesses often require more customised reporting capabilities to gain deeper insights into their operations and optimise their strategies. In order to support truly bespoke and centralised reporting, you implement your own analytics stack.
If your company is generating $5m annually: At this point, you are essentially losing competitive edge by not employing bespoke analytics.
Is There Another Way?
If you want to get started with analytics, and want the benefit of a seasoned team with decades of experience in setting up and scaling analytics, 173tech can help. Our data launcher service is aimed to help setup or migrate you to a modern analytics stack. We take you through every stage of the process from helping to select and set up the right tools, to modelling and visualising your core data source(s) in just a matter of months you can have a data stack that will last the next 5+ years. Many of our clients will work in tandem to then hire an in-house analyst to take what we’ve built and start to use it to answer those burning business questions.
More details here
More Articles
Our eBooks
Our Services
Contact Us