Introduction
In a previous blog post, we wrote about some practical steps to make your company data-driven, and instigate data culture change within your teams.
One of the things we stressed there is the value of data skills training for all companies - but also making sure it was done in the right way. In this post, we’ll dig into this in depth. How should you think about doing this internally? How should it align with your hiring plan? And when should you invest in external providers and how?
We’ll also talk about how this process and execution-maze is massively simplified for companies that partner with Enki.
This post will be most useful for fast-growing companies that need to lay the foundation for data skills as they grow, but we think any company looking to build or refine a data culture will find something useful.
Why invest in the first place?
In short, stronger data skills leads to increased efficiency across the whole org. Which makes it one of the biggest levers for the whole company to grow faster.
One thing that makes data skills unique among other technical skills is that they pay immediate dividends in any role. For example, an Operations Manager (OM) who can code might use those skills in a general way, but a OM with strong data skills will be able to ask and answer many questions that would otherwise require product or engineering time.
We’ve studied and worked with some of the top-performing growth companies in the world, like Stripe, Revolut, and Airbnb. What separates those companies from the rest is the quality of their data culture.
A big part of this is the level of data skills across the whole company and a culture of continually improving those skills. They know that broad access to data skills acts as a multiplier for every function in the company and they invest in making that happen.
So what are the investment options?
There are four options that companies often think about when investing in data skills:
1. Doing nothing proactively, but ensuring (assuming?) that their hiring process is testing for the right skills;
2. Investing in hiring a “head of data”, analysts and BI professionals to establish a data pod;
3. Training a large part of the existing team with internal training and data workshops;
4. Promoting or working with 3rd party vendors and platforms.
Option 1 is what companies do early on, and many companies stay there. The problem with only relying on this is that over 95% of the employees you’ll want to hire don’t have the skills you need on the data side. The reality is that data skills are not taught in school and are taught poorly in all but the best companies (if at all). And is your hiring team really testing for them in the right way?
Then, many companies go for Option 2. This is important, but soon companies realize this takes a long time and is very expensive. Once a data pod is set up, this helps to solve many of the urgent data needs of the business, but the data skills to provide real-time vision for the whole business is constrained to just a few people in the company. The majority of larger companies stay here.
Only once a company has resources and scale do they think about investing in Options 3 and 4. Both turn out to be very, very expensive if done in a proper and personalized way. The internal approach is also time-consuming and difficult.
The big mistake that companies make is not thinking about all of the four options above in tandem, and early on in the growth cycle of the company.
As we’ll discuss in detail later on, one solution is to partner with a company like Enki, which offers personalized, software-enabled, on-demand data skills training for a whole company. It replaces a big part of Options 3 and 4 in a cost-effective way, while easing the jobs of the data and hiring teams in Options 1 and 2.
If after reading this post you still can’t decide what the right plan of action is for your team or company – feel free to write to our data advisory team who will help you out.
Pros and Cons of Hiring
Having spoken to hundreds of managers across a wide spectrum of companies, hiring data analysts to solve their data needs is usually the first thing that comes to their mind. If you have the capital, it’s a must. But it needs to be done in the right way, and it’s definitely not a silver bullet!
The pros: Building a dedicated data / BI team is a great solution for the long-term
Hiring data analysts or establishing a full-scale department of analytics is very helpful in the long run. The reason is simple: it takes time, effort, and specific expertise to set up the scalable data infrastructure, business intelligence tool integration, and the right tools for a company.
All of these elements are extremely important and valuable for high-growth companies. As companies scale, they end up with a “pod” of data professionals who become responsible for organizing and maintaining all data infrastructure.
The cons: not a silver bullet! (& a starting investment of $500k+ a year)
The MOST important thing to keep in mind is that just hiring a data pod will NOT be the silver bullet that solves all your data needs. There are two big reasons why:
First, your data pod will be overwhelmed with requests from the rest of the company. Yes, most requests will be basic, but they still take time and focus away from higher-leverage work. As the company grows, the pod will find it harder and harder to focus on anything but the most urgent requests.
Second, even if they kick ass, the fact is that the only expertise to provide the data needed for the whole org to be productive is constrained by a few people. This massively reduces efficiency across the whole company, and therefore the leverage of the whole company to execute in a data-informed way and grow fast.
Also, keep in mind that it’s quite expensive and takes a long time to establish a data pod that’s showing value. A proper data team consists of at least three people – a data engineer, a data analyst, and a project/product manager to translate the business needs and use cases into proper technical specs.
It will take you 3-6 months to hire the pod and another 3-6 months to get everything set up and running. By now, we’re talking about $400K+ in annual payroll and another $50K in software licenses. Not cheap! So you have to be strategic in thinking about how you build out this team.
The fastest-growing companies like Stripe, Revolut, and Airbnb realized early (just as they were really starting to grow) that it’s essential to couple your data pod hiring strategy WITH an org-wide skills training strategy.
So, how might you think about doing the latter?
Pros and Cons of Internal Training and Workshops
Some companies already have many talented engineers on their team that know data pretty well. What they tend to do - is ask an engineer or a data scientist to organize a training workshop. Let’s explore this option.
The good: Maintaining docs and engaging employees with data workshops is good for corporate hygiene
If you happen to have one or more data scientists or data-skilled engineers with an educational background, and lots of desire to share knowledge – this option would work very well for you.
Having an educator on your team will improve the motivation of employees. It will also create new relationships between developers and non-developers. Eventually, you’ll build a library of presentations and documentation that could be shared with new and existing hires. There are many positives here.
The bad: Internal workshops turn expensive software engineers into low-quality teacher assistants
If you happen to have engineers or data scientists like we just described – it’s very likely that a) they are great engineers; b) their time is expensive; and c) they would deliver much more value solving core engineering problems for your business. It takes a lot of time to ensure the right curriculum and process is set up. Designing, organizing, and/or running regular internal workshops for non-developers is rarely the best use of developers’ time and skill.
Another aspect of it is that internal workshops have a tendency to never end. New hires are coming in; some departments want a workshop that is tailored to their needs; learners from previous workshops forget things and keep pinging engineers with questions. And so on.
Think about it this way – each workshop creates an infinite stream of questions from learners which developers then have to process and reply to. Essentially, you are turning an expensive software engineer into a low-quality teacher assistant for a large chunk of their workweek. This might make sense for a tiny number of people who really need your engineers’ hand-holding, but it definitely doesn’t scale.
The ugly: A lecture-format data workshop is the least efficient way to teach data skills to anyone
Quite often, workshops are just PowerPoint presentations with a Q&A session at the end, and with homework (that’s not reviewed properly) if you’re lucky. Presentations are great as a reference source but they don’t provide much detail or nuance to the audience.
To truly deliver value, data workshops need to be well-designed, provide a digital playground for practice, and an instructor willing to answer the questions. Organizing workshops also requires a knowledge base that is maintained to be relevant over time. Let’s just say that it’s very hard, and even the best software companies struggle to do it effectively.
A quick guide to external solutions and providers
Today, there are hundreds of external data training solutions, courses, and providers. From free YouTube videos to expensive bootcamps that require full-time involvement – there are many ways to learn data skills.
In general, we can organize most of the solutions into three buckets:
1. Free online resources (i.e. w3schools, YouTube, etc)
2. Structured online courses (i.e. Udemy, DataCamp, LinkedIn Learning)
3. Bootcamps (i.e. General Assembly, Galvanize)
1. Free online resources – don’t expect to learn more than the high-level from a long YouTube video
You can find videos on any topic on YouTube. They are free, available to everyone, and you can watch them on mobile. Sounds amazing, right? Unfortunately, there are two big issues:
The first: less than 5% of users watch those videos to the end. That’s because videos are similar to lectures you had at school - not engaging and in many cases just plain boring. They are not interactive, not customized, and you end up having something more urgent to get done halfway through.
And second - even if one does get through the videos and is diligent in turning listening into practice - there’s no-one there to help you apply it to their work, or answer their questions when they’re stuck.
Don’t expect to learn and apply anything relatively complex like data skills by watching several YouTube videos.
2. Structured online courses – teach how data works but won’t teach you how to use it in real life
Usually, online courses provide you with a series of videos and some interactive playground which mimics a real data environment. Watching those videos and completing exercises is a good way to understand the basics of data tools yet on a very high level that is not relevant to daily tasks or skills.
In many ways, structured online courses are just YouTube videos with interactive tools built around them. Even engagement rates are more or less the same.
To provide you with an example, structured online courses will teach your employees how to create a new data table and how data works in general. They won’t teach them how to navigate your company’s database schema or solve company-specific problems.
3. Bootcamps – full-time commitment, usually used for a career-changing move
In a time when colleges don’t prepare students for modern-day jobs – bootcamps come to the rescue.
Bootcamps require full-time commitment which is impossible to combine with a regular full-time job. It’s an expensive yet powerful solution for students or business professionals looking for a major career-changing move.
It doesn’t work well for on-demand employee up-skilling that most companies want and seek.
How smart companies up-skill their employees
As you can see from the above, improving your team's data culture and skills can be a complicated, messy, and long process. We experienced it first hand while building three $100M+ companies.
The secret to making your company data-driven is in setting up scalable processes that result in a foundation of data skills across the whole org. Fast-growing companies like MANTL, Outschool, Lending Home, Jyve, Privacy.com, and many others have partnered with Enki to help them with this.
Enki is a scalable way to ensure a whole team is empowered with data skills. Enki provides the benefits of BOTH internal workshops and external providers (Options 3 and 4), but in a simple and cost-effective way.
Enki is simple and cost-effective because:
— There is no need to set up and do all the heavy lifting of internal workshops (Enki does it for each team)
— The cost of Enki’s platform (remote experts enabled by Enki’s software) is a fraction of the cost of your company’s data analyst/scientist or engineers’ time
— The sequence of practical exercises are customized to each employee role
— The best combination of external resources are provided, customized to employee roles
As a bonus - the practical projects focus most on opportunities to maximize the benefit of cost and time savings in the team.
In summary, Enki makes it 10x easier and more cost-effective for companies to navigate and execute on the training process described above.
While making your team data-driven can be a messy and lengthy process, we encourage you to start taking action now - especially if your company is scaling! No matter which solution resonates with you the most, make sure you don’t postpone making the RIGHT investments in your company’s data culture and skills.
And if you still don’t feel confident about what is the right plan of action for your team or company – reach out to our experienced data advisory team who will happily help.
