In our first article of our pay review series, we focused on market research. Because salary benchmarks, in particular, are a key component of conducting a compensation review. But after you’ve collected all that data, you actually need to do something with it… so let’s focus there next.

What is ‘data wrangling’, anyway?

When you hear the word ‘wrangle’, you might think of cowboys, flying ropes, herds of seemingly uncontrollable livestock – and that’s not a bad image to focus on. Instead of cattle, think of swathes and swathes of raw data. Numbers, facts, and figures that need ordering, cleaning up, and making sense of.

When we talk about data wrangling, we’re talking about making sense of all that information. And much like the metaphor, it can be hard, tiring, frustrating work. Things can go wrong, mistakes can be made, it can take far too long – but at the end of the day, it has to be done.

3 key steps to making sense of your data

Data wrangling – or simply, making sense of your data – involves a couple of steps. Once you’ve compiled all your data in the market research stage of your salary benchmarking exercise, then you might expect to follow these steps (unless you use our platform):

1. Data cleaning

  • Removing duplicate data, general errors, and incomplete records
  • Correcting formatting errors and standardising things like currency

2. Data transformation

  • Normalising variables like role, seniority, or location
  • Find and fill gaps in missing data, by guessing or deleting

3. Data validation

  • Double-check for outlying numbers or anomalies to validate it
  • Cross-reference a few sources to make sure the data is reliable

Why does accurate data matter for pay reviews?

If this is all sounding like a lot of hard work, that’s understandable. You haven’t even started the actual pay review itself, and the preparation has been nothing short of extensive… but don’t be tempted to skip this part of the process, because it’s more important than you might realise:

  • ‍Pay decisions need to be validated: if your data makes sense, it’s easier to justify pay decisions to leaders and line managers. With messy data, you run the risk of making the wrong calls – and under or over-paying staff, which could be disastrous.
  • ‍Benchmarks need to be accurate: if you compare yourself to incorrect benchmark data, you risk either disengaging and losing your current staff or turning off prospective hires – simply by not staying aligned with market trends.
  • ‍Costs need to be managed properly: sometimes it’s a balancing act, because budgets don’t allow you to exactly match the market when deciding compensation. But having all the information helps you make these complicated calls.
  • ‍Pay structures need to be equitable: with all the right data on hand, you can spot pay gaps across a range of demographics – from gender to ethnicity and age. This matters to current and future employees, and can have legal implications too.

Why not to do the ‘data wrangling’ yourself

So we know that once you have all this fascinating, complex, insight-ridden data piled up on your desk, you need to make sense of it. But we’re not saying that you have to do it yourself. In fact, we’d actively encourage you not to for three very good reasons…

1. It’s hard work

If you’re reading this, you’re probably a HR professional or some other kind of people leader. And if that’s the case, we know you probably have a wide range of skills… because inevitably, you wear a lot of hats. Stategiser, administrator, payroll manager, counsellor, event planner – a lot of those may apply. But that doesn’t mean ‘data analytics expert’ is one of them.

Data analytics is a highly specialised field, and even for an expert it would be hard work. And when you’ve got 101 things on your plate besides pay reviews, it may be left to leave it out.

2. It could go wrong

Again, even for an expert, wrangling data can go horribly wrong. Multiple spreadsheets and download files, hundreds or even thousands of data points, complicated formulas and inaccurate figures – handling all of this yourself puts you at a real risk of human error.

And even small mistakes can have a huge knock-on effect. Suddenly, you’re basing pay decisions – remember, decisions that affect your employees’ real day-to-day lives – on inaccurate information. That’s not going to end well.

3. It takes too long

Not only is cleaning and validating data difficult, it also takes a lot of time. There’s no getting around that if you’re handling the process manually. And that delays your pay review itself. Leaders get impatient, line managers feel pressured, and employees begin to get disgruntled that they still haven’t heard any news about the pay rise they were expecting.

For some organisations, compensation reviews are annual – which may mean a longer process feels justifiable. But then you run the risk of basing your decisions on outdated market benchmarks.

Compensation IQ handles the data, you make the decisions

Much like doing your own market research on salary benchmarks, it’s not an efficient use of your time to clean, transform, and validate data. Instead, you can leave all that to Compensation IQ – letting our product do the heavy lifting, so you’re left with clear answers around pay.

As well as providing you with up-to-date, industry, location, and company-size relevant benchmarks around compensation, Compensation IQ also organises and maps all that information to your own HR data. What you get is ready-to-use insights.

All you need to do is connect your HR platform. That’s it. After all, the product does the rest – summarising everything cleanly and clearly in dashboards, report packs, and data files so you can make justified decisions about pay.

Basically, it comes down to this: Compensation IQ helps you make easier, faster, better decisions about pay. And you can try it for free for 7 days.

Get a free trial, and simplify your salary decisions