Career Advice: Data Analyst. (Western Australia)

• 

  couldnthaveknown on 11/09/2021 - 02:50

  +20

  Hey mate, I think you're making a great career move. I say this having been CFO of various tech companies, and most recently, commercial finance lead for Australia for a fintech company where I led a global big data analytics project. I'll answer your questions first and add a few final thoughts

  Is it too late to start this career?
  Nope - not too late to start any career. For what it's worth, over the last 15 years I've changed careers many times. From law, to investment banking, to tech company CFO/big data fields. The key is selling yourself - if you have a convincing reason for the career change, it's a plus. It's better in my opinion to change because you've found a passion for something, rather than continuing to work in an industry that you've always worked in.

  What are the job opportunities for a beginner data analyst in Western Australia?
  Many! There is such growth in data analytics across all industries, whether that's Woolworths, mining, or tech companies. Woolworths might ask what products are most commonly purchased when a customer purchases bananas, and it's up to the data analyst to work that out so Woolworths can advertise those products. Miners might ask which of their machine operators are the most efficient with the least serious mistakes over a period of time. A tech company might ask, how long exactly should a salesperson spend on a lead of $Xm before its unlikely the sale will close. These, amongst many others, are all within the domain of the data analyst. Banks and insurers might ask data analysts to model fraud, etc. Anecdotally, I have heard that most of the investments in these big companies are in big data, not traditional areas such as accounting & finance or customer service.

  How hard it is to enter the job market at my age by the time I up-skill myself?
  I don't think age is a big disadvantage if you have the skill. The key coding languages for a data analysts are SQL, R, Python, and at the more advanced level, Java and C++

  If you don't know any of these languages, or just rely on excel, it will be difficult. If you know common languages like SQL, R and Python, your chances will be good. If you learn Java or C++ on top of that, I would estimate your chance of landing a job is pretty high especially if you require little upskilling to begin producing insights.

  Where should I start? Any recommendation on an online course.
  Datacamp will teach you the basics with some of these languages - SQL and R. I don't think they have courses on Python and Java etc yet but I suggest by starting with free Youtube videos, then practice a lot!

  My maths is average. Is this a problem?
  No I don't think so. Maths is an application of logic, but being a good data analyst is more about understanding data, how to translate data to insights, and database design. Don't worry if you read "database design" and aren't sure what it means - that is quite advanced and in most cases, an entry level role in data analytics won't have database design aspects, you just use what data the company has, however it's structured. Turning data into insights is the key. This I think breaks down into 2 key components. First, is understanding what kind of insights matter to the business. This commercial understanding is actually pretty rare. Most data analysts expect to be told exactly what data is required, without knowing how this translates to commercial success. If you can bridge this gap, it is a massive advantage. Second, is how to write the code to draw that data out. In my work, I might write code to view data in 10 different ways, and from that, I can produce 1 or 2 meaningful insights. The other times I view data either produces an inconclusive result or something that's not actionable. What you might notice, is none of that is particularly mathematically intensive. There aren't integrations and differentiations. You just need to understand what's commercial for a business, and how data can be used to drive those commercial levers.

  My overall comments are that this is an excellent area. Data analytics will only grow as companies produce and collect more and more data. A skilled data analyst can expect to be in strong demand for the foreseeable future. If you put in the work to learn those programming languages, and importantly, learn the commercial aspects of an industry such as banking, you will find success. The starting salary for a junior data analyst will probably be around $70-80k (sydney benchmark), with that rising to $150k with 5 years experience.

  Here's a real life example. Due to data protection regulations, tech companies are not allowed to identify the unique user of a phone that uses their app by name. If you were a data analyst, how would you accomplish this task? Could you use a combination of that phone's screen resolution, the operating system that phone uses (android, ios), the other apps installed etc, so you could identify that unique user without explicitly using their identity? Or here's a different example - many online retailers send out email advertisements. Can you help me work out what hour of the day, and what day of the week, that customers are most likely to open marketing emails? Surely, 6pm at night when people are not working is a better time than 9am when people are just starting work. But a data analyst should be able to prove that, my hunch might be wrong after all.

  Hope that's helpful mate and best of luck on your journey!

  • 

    Ryanek on 11/09/2021 - 13:27

    Holy reply Batman!

    • 

      couldnthaveknown on 11/09/2021 - 15:59

      +3

      Haha yes indeed, I'm something of a workaholic so career related questions is one area I can make a meaningful contribution to. However, I'm always very careful to present the full picture, otherwise it's very easy for well meaning advice to send someone down the wrong path. I try and illustrate multiple reasons for my recommendation, rather than give an outright recommendation, so OP can make up his/her mind on the basis of those reasons. In some cases, I've benefitted from similar advice. In other cases, I wish I would have, rather than going down the wrong path and having to course correct later, so this is my way of paying it forward!

  • 

    digitalfarmer on 11/09/2021 - 17:49

    Appreciate your reply. Thank you

    • 

      couldnthaveknown on 11/09/2021 - 18:58

      A reply so nice you said thanks twice? Haha I'm joking - you're welcome and best of luck! Remember to

      1) Network like crazy - having good technical skills only gets you to the starting line (i.e. score you an interview). Having the right soft skills such as communication and interpersonal relationships, gets you to the finish line. Lots of people won't reply or won't want to speak with you - and that's ok. Don't take those rejections to heart. Be the person that warmly reaches out to a lot of people, letting none of the rejections diminish your optimism - that's very powerful.

      2) Have an elevator pitch - once you've learnt the basics of SQL etc, you should have a 15 second spiel about yourself. When I'm asked, I say "I'm a growth focused CFO with AI expertise" or similar.

      3) Have a good resume - lots of websites have lots of different 'tips', some of which are rubbish. Based on my experience, I found this company's blogs very on point (and based my own resume off this - I have no affiliation with this company) https://arielle.com.au/how-to-write-a-resume/

  • 

    AndyRates on 11/09/2021 - 23:24

    Hey, do employers for data analyst, require a data course from GA or uni or you can learn from Youtube, Udemy and that would be enough to prove your ready to make a contribution in the role?

    • 

      couldnthaveknown on 12/09/2021 - 01:10

      Hey mate!

      The short answer is no, you don't need a uni course or a data course for that matter. You'll be asked how many years experience in X language you have, and be asked to do a small exercise to demonstrate it. Even if it's verbal, the way you describe how you'd do the analysis would be enough for an experienced hiring manager to determine whether you 'get it'. The "contribution in the role" part however is different to the technical skill of writing code to operate a database. To do this, you must have a good understanding of the commercial side of your business - in my experience this is much rarer. There are lots of data analysts who can code. But those who also understand the commercial side of the business can generate meaning from that data, turning it from a 'nice to know' / 'so what' into a recommendation that is immediately actionable.

      The long answer is a bit more nuanced. If you in your mid 30s or older, it is unreasonable for employers to ask you have a data analytics uni degree. Back when you went to uni, there was no such thing as a data analytics uni degree! They'd basically be asking whether you studied this new age expertise when everyone was still using Nokias. It's like asking an older mechanic whether they learnt to work on EVs. You can truly say you learnt data analytics 'on the job' during your 10-15 year career.

      If you're younger, and went to uni in the last 10 years, that question becomes more relevant. Because you would have had the opportunity to study data analytics, but chose not to. So naturally the question is why not. That's a fair enough question, but a genuine desire for a career change is a good reply - and in this case, having done a few courses yourself (Datacamp, or Udemy as you mentioned) and maybe competed in a few competitions etc, shows a genuine desire for change. If you haven't done any of those things, it's hard to show you want a change as you don't know what your destination looks like! Another great way is to have moonlighted in the data analytics department of your workplace now, if that's possible, or you've attended a number of big data tech conferences (lots are free, like Tech23 amongst many others) because you're especially interested in a topic (such as natural language processing, database design, algorithmic processing). As I said above, if you go, network!

      The CTO of a tech company I worked at once said to me it's important to have a work portfolio, much like an artist, that you can show/tell people. In building up your portfolio you can clearly demonstrate your ability to drive commercial outcomes with data analytics. For instance, you might have created a 3 tier insights framework designed to drive commercial outcomes with that company's clients, with the first focused on automation of reporting, second being automated analysis of that company's performance vs. the industry average, and third tier being custom data analysis for that company's specific strategies, which is a premium service offered only to top tier clients. If you gave a reply like that with examples, then regardless of whether you went to school to learn data analytics, I would think you'd have a pretty good shot at getting the job. Remember, knowing the coding languages only gets you to the starting line - I'd rate my coding skill as a 6 or 7 out of 10. All the young kids are better than me. But using that data to create commercial, actionable insights is the key.

  • 

    CharityCase on 18/09/2021 - 20:20

    +3

    Sir, I don’t know how I end up on this thread while randomly browsing Ozabargain on a Friday night, your post got me into registering on Datacamp and finished 3/4 of the Introduction to SQL in just two nights.

    I just want to say thank you, I’ve been working in the airport for many years then covid got in the way, and I was made redundant after Jobkeeper finished, now I’m working on a new entry level job in an unrelated field (starting over man…need to pay bills) and I feel like I got lost in my career, can see myself in five years doing the same shit job over and over (I’m in my 30s). After reading your post I feel like I need to upskill myself, and data analytics seem to be a right direction, we all live in big data, don’t we? Obviously I don’t expect to land a data analyst job soon after completing those online courses, but my question would be, at what skill level (like SQL) can actually open a door for being an entry level data analyst? What career paths/jobs (other than data analyst) that I can choose once I obtained “above average data analysis” skills?

    Thank you in advance and sorry for my English (not my first language)

    • 

      couldnthaveknown on 20/09/2021 - 01:33

      Hey mate, very welcome! I'm glad to see your reply and very happy to have made a difference in your career.

      So first things first, there is absolutely nothing wrong with changing careers. I've done it 3/4 times - law, accounting, investment banking, tech. Most people choose what they study at uni when they're still in their teens, maybe 18. The chances anyone will make the right decision to study in an area they'd still be passionate about at 45, or 55, is honestly low. After all, when we choose what we study, we have no information about what it's actually like to work in that area for 40 years. But, most people don't change careers because honestly, it's pretty hard, and of course there are financial reasons. So, they say "I have a good job" or "this career isn't bad" or "could always be worse" and become accustomed to the steady salary. You never hear those people say "this is a great job" or "I love this". Ironically, the fact that their job and pay cheque is "good" is exactly what prevents them from doing what they love, and become "great" (see the book Good to Great, it talks about this). I also think this is a key cause of mid life crises, and people trying to get their children to live out their dreams. It's because they feel unfulfilled. But you know, you get one shot at this life, and it's a short one at that - so give it a go.

      Ok your second question. The coding skill level required to land a data analyst job is not high. Since you're learning SQL, I'll say this - if you can do each of the below functions, it will be enough: SELECT *, CASE WHEN, GROUP BY, LEFT JOIN, UNION, WHERE, AND, etc

      That's not an exhaustive list, but the point is - those are all basic SQL functions. You don't have to be a master of JSON string extracts. In practice you'd most likely be asking other people for their code and modifying it. Honestly, I'd rate myself a 6 out of 10 on coding, but that's been enough for me to lead a global big-data project. The reason is because operating the data is the simple part. The key is how the data relates to a commercial or profitable outcome for the company. The data is just a tool to create this outcome.

      Maybe we can use the airport for some examples. As a data analyst, you might be helping find ways to help the airport grow revenue. A good data analyst would need to understand the commercial side of the airport business, be able to develop an idea, and test it with data. Here's an example

      Airport retail is a very unique situation, because your customers are 'captive', meaning they have no where to go as they wait for takeoff. Many are bored while they wait as well, so retail offerings get far more attention than usual. This is the commercial side.

      As a data analyst, you might look up all the websites travelers accessed while using the airport's wifi and isolate the websites where they spent the most time, and look for a theme. For instance, you might find that the most visited website by travelers is Reddit, but this has no commercial relevance so you'd ignore it. But, you might find that 3 of the top 10 websites are jewelry related. Maybe 3 popular jewelry brands, or jewelry review blogs or similar. This is the data side.

      So, you theorise that there is a retail opportunity as a lot of travelers might be businessmen who are be looking for a gift for their partner after being away for a business trip. This is the actionable insight, which comes from a combination of understanding the commercial side of the business, and operating the data to support that. You might use the data side again to find supporting pieces of evidence, for instance, do existing retailers within the airport sell a higher % of 'gift' items. A clothing retailer might sell more bracelets than a non-airport store. This might indicate there are more gift purchases as not many travelers truly need a bracelet for their trip, as opposed to say business shirts which might be needs driven.

      Based on all of the above, you might put together a presentation to show management your theory, supported by data, followed by a recommendation that a pop-up jewelry kiosk be set up to test its commercial potential. You might even recommend that instead of the normal rental agreement, a profit-sharing agreement could be put in place instead so the airport shares some of the sales upside if your theory is indeed right. You might even ask the marketing department for help on how to best advertise this to travelers (put up banners that say 'happy wife happy life - buy some jewelry for your wife' at the terminals? haha) and put this into the 'recommended actions' part of your presentation.

      I hope the above illustrates what makes a great data analyst, compared to a good one. A good data analyst can write code to analyse those websites, and analyse retail patterns, but would only do so when they are given explicit instructions by a commercial director. If I'm honest, working with people like that is really exhausting because they want you to tell them exactly what to do, whereas I would like them to understand what I'm trying to do and proactively follow the path. A great data analyst however, would be able to come up with an end-to-end proposal. You don't need to understand the commercials perfectly - the commercial director will nut out the exact details of the profit sharing agreement. You just need to know the broad strokes of how airport retail works, and use the data to create those insights as well as communication skills to get that across.

      I hope the above example shows that yes, coding is one aspect of it, but if that's all you know, you're only across 1 step in a 4 or 5 step process. Once you've got a reasonable amount of coding skills, I'd suggest learning the others. It's more valuable to be a brown belt in all 5 things, instead of a black belt in 1 thing and a white belt in the rest. Make that your selling point in the job search - you might find that's quite a rare skill set.

      Good luck my friend - reach out anytime

      • 

        datalearner on 18/10/2023 - 05:27

        Hi mate! I really appreciate for your valuable insights. I'm currently a third-year student majoring in data science, and your insights and the practical use case example you provided have enhanced my understanding of the data analyst role. If possible, I would greatly appreciate the opportunity to connect with you. Thanks again!

• 

  digitalfarmer on 11/09/2021 - 22:51

  😀Thanks

• 

  RQNinja on 23/09/2021 - 12:39

  following this post so well written and i will be turning 30 soon. feeling hopeless and undervalued in life working for a 9-5 job bowing down too much

• 

  digitalfarmer on 26/01/2022 - 01:52

  Merged from Data Analytic or iOS App programming

  Hi Guys,

  I am 41 years old and working odd jobs since moving to Australia and looking for a career change with no prior experience. The field of interest is Data Analysis or iOS app programming. Whatever I pick I have to start from scratch. Looking for advice.

  1) Which career field is better for long-term sustainability. (Looking to work as a freelance/work from home).
  2) What (Data Analysis or iOS app programming) will be quick to learn and able to get a junior-level job.
  3) Am I too old to start a new career. How difficult it will be to get into the industry as a junior.
  4) Any recommendation on where to start (Data Analysis or iOS app programming) online course/resources preferred.
  5) What would you pick if you have to start your career from the beginning with no prior knowledge in any industry or qualification and English is your second language.

  Am I missing anything any recommendations or advice?

  Thanks in advance.

  • 

    rektrading on 26/01/2022 - 15:34

    The field of interest is Data Analysis or iOS app programming.

    Both of these branches have been saturated by young teens since 2010.

    If you're passionate about learning new things then have look at an industry that can't find enough talented devs and/or DA/TA.

    https://moralis.io/
    https://m.youtube.com/c/MoralisWeb3

    What would you pick if you have to start your career from the beginning with no prior knowledge in any industry or qualification and English is your second language.

    None of these things is important as long as you can show your skills with code and/or DA/TA.

  • 

    Jolakot on 27/01/2022 - 15:07

    I work as a full-stack engineer, traditional route of doing a 3-year compsci degree before moving into a grad program, so I've seen a lot of the industry

    "Data Analytics vs iOS App Programming" is the wrong question to ask, because as a 41 year old with English as a second language, with no experience or qualifications, you have no hope at all competing against young Indians/Ukrainians with compsci degrees for freelance/remote work, who could undercut you below what you could even afford to take with Australian costs of living

    Even for on-shore work you'd be competing against young Australians who grew up immersed in data and code, especially uni students doing casual contract work, who could undercut you with both lower living costs and by being able to complete a contract in half the time it would take you while starting out

    So just forget picking a general branch, it will lead you nowhere, so many people waste money on shiny courses or bootcamps that guarantee jobs only to waste money. Tech employers throw away job applications from older adults with little experience by the wheelbarrow load, the industry is incredibly ageist and racist towards people from West Asian backgrounds

    You've made the same mistake a lot of people make when trying to shift into IT, by discarding what you already have and expecting to start from scratch. Because the secret-sauce here is domain knowledge, a tech employer would likely throw away any resume you give them, but most jobs in tech are with regular companies

    You're 41, what have you been doing to support yourself? Even if you've just been working odd jobs you'd still have picked up a ton of domain knowledge, and that's what will differentiate you from a recent compsci grad

    Look at jobs you've done previously and think about the IT side of them. If you worked casual labor hire then think about who maintains all the scheduling systems, who crunches the data to get useful insights for shareholders, who deals with keeping all the laptops updated, who works with software vendors to configure the products used, who maintains payroll and accounting software, etc

    Because there's a huge shortage of people who are willing to do the unglamorous IT work, especially maintaining boring niche legacy systems. For example MedicalDirector software is used by hundreds of medical clinics, and the amount of people in Australia who are certified to maintain/upgrade/fix it is tiny, especially for legacy versions

    Each industry has their own boring legacy products that few people can/want to touch but that are absolutely critical, so find them in an industry you know and work out how you can get certified for it by the vendor. It's a dead-end career, but it pays damn well and at 41 it won't really matter