Building your Data Architecture vs. your Skincare Routine
Much like flawless skin, insightful data is not a given.
While some are just seemingly blessed with luck of the draw DNA, the vast majority of us must put planning and processes into place in order to achieve the glowing results of our dreams.
We’ve all walked into a department store beauty section and needed a minute to steady ourselves amidst the dizzying display of lotions and potions promising to fix facades and give meaning to our lives. With so many options and opinions, it can be hard to know where to even begin with your skin.
Building your business’ information architecture can feel similarly discombobulating and with such high expectations on the results, you may be tempted to grab a handful of solutions and get the hell out of dodge.
But never fear! Because whether you’re building your skincare routine or carefully crafting your data architecture, there are tried and tested tips you can follow to get you from here to the place you’ve always wanted to be.
So let’s think of your data architecture as consisting of three main steps: The thinking, the doing and the maintenance. Allow us to break it down for you:
To start off your journey, you’ll need to make a plan before you can stick to it. Which means mapping out where you are now vs. where you want to go. So sit down, take a deep breath and think about what you want. Once you have clean and clear access to your data…what do you actually want to do with it?
When it comes to your face, the likelihood is that you came to the game with a pre-existing laundry list of symptoms to sort out on your journey to achieving skin Nirvana. Redness, dryness and discolouration to name a few. And while your data needs may appear murkier than your complexion – we promise that enlightenment is in your future.
To help you get that clarity start with Tip #1: Involve the end users of your data in the planning process.
You wouldn’t expect a dermatologist to prescribe a skincare regimen without having both spoken to you about your goals and seen your skin in the flesh. That would be ridiculous wouldn’t it. We’re glad you see that! But believe it or not, one of the most common tragedies to befall businesses when building their data architecture, is allowing leadership teams who have no involvement in using the end data, to call all the shots.
Take this approach and all that blood, sweat and tears will be for nothing when you overrun on time and budget; ending up with a solution that nobody actually uses. Because despite the pretty packaging and your shiny enthusiasm for a much needed change – one solution does not suit all.
So speak to the end users of your data and decide what basics you need in your arsenal: A thick layer of analytics to track your chosen metrics and a spritz of reporting for greater visibility on your overall business performance. Then finally, what data models are going to provide you with the necessary insight to take your business from downright dull to dewy.
Now that we know where to begin and who to trust, let’s delve into Tip #2: Don’t over complicate your build.
So you want it all? An all singing, all dancing solution that promises a future of untold data delights. Listen, we get it. We really really get it. The temptation to throw one thousand new products at the problem and hope that one one of them sticks is understandable. We all want in on the latest and greatest super-ingredients for success, but do we really need to use all of them? Probably not.
If you did this with your skin, you would have a pretty poor sense of what was working for you and what actually wasn’t. This is a great way to spend a lot of money with very little measurable ROI to show for it.
Instead consult with your data engineering and analytics teams to decide which tools are suitable for your business’ unique landscape and volume of data. Choosing the right tools will help everyone to maintain faith in what your business is trying to tell you and it will be far easier to get sign off to turn your insights into action.
So now that you’ve thought long and hard about your needs and assembled a council of wise folk to steer you well, it’s time to begin. First, decide which sources of data are valuable to your business. These could be your google analytics, your shopify account and your ERP system amongst others. The basic premise of your data architecture is to afford you a deeper view of your business by funneling all of your disparate data sources together into one easy breezy beautiful data hub.
This is the only way to truly get your story straight.
Then, much like with your precious visage, you will need to get your data cleaned, cleared and under control before you can hope for any bigger transformations. It’s important to treat the skin you’re in, so you’ll need to cleanse and standardise before you can make accurate assumptions about what is working for you and what isn’t living up to the hype.
Which brings us to Tip #3: Stick to the steps.
Just as with your carefully assembled skincare routine, your information architecture must also be deployed in the correct order. A modular approach if you will. For example, those in the know wouldn’t dream of applying a thick product before a thin product as the former would reduce the effectiveness of the latter. This would be like throwing money down the drain. So if you thought you could save some time by skimping on your cleansing and jumping straight into some fancy data modelling – think again. This approach would have you continuing to report on siloed information instead of gaining a greater understanding of your business by cross referencing multiple data sets.
So you’re following all the rules. What now? How about a reminder that patience is a virtue with Tip #4: Give it Time.
There is no such thing as an instant fix. Whether you’re looking at your skin or your data, real results are only seen through consistency. So if you’re prone to binning products that don’t yield transformations overnight, you are likely to be missing out on some radiant results in the long run. Remember that your information architecture needs time to actually accumulate historical data in order to make any meaningful connections and generate those informed insights you so desperately want and need.
Don’t sit around and pick holes in the process. Instead, keep sitting pretty for a minimum of 6 months after your data models have been put into place in order to get a real read on where you’re at. Occasionally after this amount of time you might even uncover gaps in your original knowledge that will require you to reassess your system and possibly rejig, recleanse and rebuild. If waiting that long seems like cruel and unusual punishment – just remember that sometimes beauty is pain.
To maintain or not to maintain? This should not be the question. Why do we seem to think that the moment we’re finally looking fresh-faced and blemish free, that it’s a sign from the universe to sit back, relax and let our cards fall where they may? For those of you who believe that your work is done at this stage – we have some terrible news. Your skin is subject to change. It will age, it will be affected by your lifestyle, environment and daily activities. Sometimes it will bend or break it ways you never foresaw. The way you care for your skin will evolve throughout your life. The way you care for your data is no different.
With that in mind our last piece of advice is of the utmost importance. Tip #5: Build trust in the data.
Do you remember that one time when work was hectic and your zoom socialising got a bit out of control? You were tired, you were busy and you started slacking on your skincare. Suddenly you were adding breakouts and breakdowns to your weekly to do list and cursing your face for not staying in its place. But the truth is that your skin is a living organ and needs to be monitored and maintained to function at optimal capacity. So too does your data.
Even the most meticulously planned information architecture can crumble if not given proper attention – and in reality if your business doesn’t trust the data, there is likely to be little use of the analytics and reporting you so painstakingly put in place. Inevitably things will go wrong, but if you stay primed and ready to address any issues that arise, then you won’t have to waste time and money finagling resources to save you from becoming just another data school dropout.
Remember that if your team is in doubt, they’ll leave it out. So be sure to dedicate regular TLC to keeping your data architecture alive and thriving. In reality your journey with data shouldnt be a million miles away from the care and consideration you already give to your skin – either way you deserve a solution that is bespoke to your specific needs and challenges. Because at the end of the day, you’re worth it.