Hiring an analyst can be intimidating. First, the term is so widely used, it’s near impossible to determine who spends the bulk of their time making sense of data versus a wide variety of other tasks. Even a glance at PayScale gives you some insight into this issue.
So, how do we find these people? Many people throw jobs out there, hoping the right candidates will find them on one of the career sites or LinkedIn. Sometimes, people contact candidates they think are a good match directly. This approach can be successful, but it takes a lot more time. Both approaches lead to the concerns about the severe lack of qualified candidates. The other challenge is the cost. When PayScale shows you a range of everything from $35K (USD) to where the sky is the limit, costing a job becomes challenging. I used to do this for a living and it’s a real pain. I don’t envy you.
Another challenge is software. There’s a lot out there and different packages offer different things. I’m 100% biased in my flavor of analysis and my focus in life is extremely narrow. If you want broader, there’s a number of experts you can hit. I’m not alone in buying into a particular software. From the outside, we look unreasonable. Why does it matter if we use Tableau versus PowerBI if the skills are generally the same? Some people use many and others are as rigid as concrete on this. Pick your poison.
How we describe jobs also significantly impacts who applies. I see a number that read like this:
Computer science or mathematics degree
X number of years using business intelligence tools like PowerBI, Qlik, Tableau, Cognos, etc
X number of years doing enterprise data warehousing
X number of years writing SQL, Python, etc
Must be a team player and able to work solo in a fast-paced work environment
R, Python, or predictive analytics a plus
They’re usually 3 times this length with other, very technical requirements. Here’s the challenge with some of these:
- Education makes great a tremendous impact on your life and path. Yet, I find more people shifting careers than staying in a straight line. Of the 5 female Tableau Zen Masters, not one holds a degree in mathematics or computer science. Chemical engineering, demography, anthropology, industrial design, political science, and interpreting – yes. These are all analytical fields. The same is true for others of diverse backgrounds and many others – business degrees are another I see often. Typically, these people learn analysis on the job after getting their degrees. Not only do they bring solid technical skills, but they often have an easier time translating business requirements and bringing novel ideas to the field. How many of these people opt out of applying for these jobs due to degree requirements alone that – frankly speaking – often become outdated 5 years (or less) into the work force?
- For tools like Tableau, while they’ve been around 13 years, they’re still in the process of hitting critical mass. I started with Version 6.0 around 2010 and that makes me “mature” in the Tableau world. You could be missing out on tremendous talent due to arbitrary level-setting. Some people get very deep with Tableau very quickly (you’d be surprised how far people can get in 1 year) while others may never get there (consider driving – some people are just A to B drivers) despite being users of the product for years. Consider naming specific milestones (ex: level of detail calculations) or asking for work samples. In the Tableau world, there’s Tableau Public and a wonderful project called Makeover Monday with data sets that can be used on virtually every platform.
- What is the goal of the job? Cross-training became in vogue at one point as a means to control costs. However, when you have an artisan level job (one requiring deep expertise in a subject), you lose efficiency because you end up hiring a generalist (the person that seemingly can do everything). Can you grow some skills? What percentage of the job is comprised of the add-ons? Consider specifying this (ex: 5%) as, without clarifying, all things are assumed at equal weight, and those more proficient in one (ex: data visualization) may not apply to jobs that seem heavy on coding or data warehousing requirements. A lot of people get lost due to add-ins that comprise a small amount of the work and that can often be learned or supported elsewhere. Also, what drove the need for the job? Are your dashboards good, but not great? Focus on storytelling. Do you need server support and scalability? These are different skills than desktop. Do you need evangelizing? Get a champion who can teach users and get others on board. Internal employees may be able to highlight the gaps.
- Are you undervaluing soft skills? I see tons of ads focused on technical skills and not many covering the other parts of the job – user engagement, requirements gathering, evangelizing the platform, etc. Without these items, many BI platforms fail in the organization. They hire great programmers and analysts, but poor champions of the service. You need a balance or the struggle will be real. Champions are the lifeblood of a BI platform and get energized by others. Many organizations try to cover all the bases. While some candidates will apply if they meet most skills, a larger number – particularly women and minorities – opt out at the kitchen sink job description. I know I do, despite being a Tableau Zen Master, Tableau Public featured author, and 3-year consultant with SQL skills.
- I’m still looking for the ad for a slow-paced work environment. You know, the one with the hammocks and wine cart. When the internet hit and various economic bubbles burst, fast-paced became the norm. Every organization needs self-starters who play well with others at a certain career level. What is a real pain point you have? Is it communication, prioritization, or completing projects? Put that here and you’d be amazed at the solution-finders who apply.
- Do your “nice to haves” scare people away or come at a price point you’re not willing to expend? Data sense-makers, especially those with advanced skills, can and do command a high price. They’re the ones who are most likely to find the golden nugget that businesses seek. For this reason, most go into consulting. Do you really need this person as a full-time employee or does a short gig with someone like this get you what you need? Consider getting a consultant for the out-there items and an employee to maintain, engage, and grow your program. Long term, this helps with retention and keeps your employees from losing their minds. Or, consider growing one of your employees. If you’re just launching data visualization, then you need time to grow into predictive analytics. Too much at once will overwhelm and disenfranchise your end users.
- If you’re hiring anything with senior or lead in the title, consider disclosing the pay range early. There’s nothing like finding the perfect candidate only to find you’re $30K off in salary expectations. Yes, the person will say no immediately, but they may refer someone appropriate for the pay grade. Also, states like Massachusetts have banned asking about previous pay and many job coaches recommend avoiding this question. Yes, someone may get a tremendous pay bump – it’s often the reason many people switch jobs and candidates of diverse backgrounds or females are likely to be underpaid. People stay when they feel fairly compensated. Also, be aware your title can make or break who applies. Words like developer, architect, and engineer all have very stringent meanings in BI and favor a more code-centric approach, not visualization (or a tool as code-free as Tableau). Consider something more descriptive, as those within data visualization lack a standard name and more than a few have strong opinions about what they are. These labels may also have connotations about pay grade (see above PayScale images).
Consider screening your ad both internally and externally. Also, look at your job application process. Does it unfairly screen certain populations out? What is critical (background check) versus potentially political (online application / personality assessments / certain practices in the interview)? And yes, I know how staunchly organizations defend these processes. If all your candidates look or sound the same, you may be creating an unintentional wall. Personality assessments, much like IQ assessments, are biased.
Social media has also changed how the pipeline works. Hot talent that’s well connected will often find jobs through the closed circuit, as well as the open one. Keeping in touch, being open and transparent, and being clear when things don’t work out will at least keep a connection to someone who, in the future, can act as a powerful referrer to you. Flexibility, speed, and communication are your friends. Also, it’s critical to find where people are – do they interact on Twitter, forums, or LinkedIn? Also, know that there are groups dedicated to women, diversity, and other groups who may get missed in mainstream channels or in referral programs. If you don’t know where to find people, ask. LinkedIn groups, InMails, and Twitter can be a real boon here.
Lastly, realize the employment culture has changed. Reid Hoffman notes this and offers a model that, even if you choose not to use, you should keep in mind. Fewer people are staying at jobs longer. Instead of focusing on getting a person at a role indefinitely, consider how long term that role can grow and create new roles behind it. I know it’s hard in most hyper-fast companies and few organizations have the time or money to throw behind this level of career planning, but long term, it does save money and retain top-tier talent. Consider future meetings and let your new hire direct you.
Not all of these ideas may work at your organization, but hopefully 1 or 2 will help. Best of luck in hiring!