How to Automate Payroll in a Large Company
These days, most organizations invest in payroll automation as they grow, but some large companies still handle their payroll in spreadsheets or clunky legacy software.
Automating payroll can help large organizations save time and money, but it’s far more complicated to make the switch when you’ve been running a business for a decade or more.
We’ve helped many large organizations automate their payroll administration systems as part of a digital transformation, and we’ll share our best tips for you on that today.
1. Define the outcomes you need first
Many people make the mistake of letting salespeople tell them what’s possible and what they should expect from a payroll automation solution.
Do not buy enterprise automation software on its features. Make sure you pin down what the solution can deliver and how much of that deliverability will depend on your and your team. That leaves them open to overlooking a key function or requirement that only appears once it’s too far gone in implementation to backtrack.
Some questions to think about:
- What does ‘automated payroll’ look like to your team?
- What’s your dream solution?
- What will it have to integrate with?
- How would you like to use or interact with the solution?
- What security requirements or regulations will it have to comply with?
2. Don’t commit too early
Core systems like those involved in payroll do not frequently change — especially in companies with several hundred or thousand employees. And for good reason.
The complexity of these processes — and the sheer work involved in unpicking and understanding their mechanics — means implementing payroll automation software in a large enterprise can take anywhere between six months to two years.
Most payroll and sales comp analysts have been burnt by oversold software or a hellish implementation at some point in their career. They are rightfully wary of putting themselves through it again.
Don’t feel any pressure to commit to get a good price or get the job out the way. Taking additional time to ensure you choose a solution that will deliver what your team needs for years to come is worth the investment. Measure a thousand times, cut once, as they say.
3. Integrate all relevant employee information
You will need to gather and upload all the basic employee and payroll information into the new system. That will include:
- individual employee pay rates
- bank account information
- employee benefits information
- personal tax forms and related tax codes
- social insurance numbers
- vacation pay information
That information will need to be programmed during implementation by your team or someone from the vendor. Most payroll solutions allow for a simple CSV upload but may still require manual data entry and validation.
Our advice is to check it all three times over. Errors at this stage of implementation will be a nightmare to fix later if they don’t break the automation system completely.
4. Integrate all variable compensation data
If you’re running an organization where everyone is on a fixed salary, you can skip this bit, but that is not the case with most large organizations. There are two types of employees with variable comp data that need to be considered.
a. Sales employees with commission-based income
Sales employees will require more attention because their pay will vary month-to-month. Collect all the rules, exceptions, and details about the compensation plan structure and individual products.
b. Hourly or shift-based employees
Shift-based employees or those on an hourly rate will also have variable compensation because the number of hours they work a week will vary. Information about their hours worked will need to be collected and transferred from whatever time-tracking system you use into the automated payroll system.Most attempts to automate in large organizations fail because they forget to decide the outcomes they want during the procurement processes, and end up buying features they have no idea how to use. #automation #rpa #procurement Click To Tweet
These integrations will need to be fully automated or require minimal human involvement to avoid bottlenecks that slow down the whole process.
Most large enterprises find that their legacy sales compensation software creates an impassable bottleneck that prevents the full automation of the payroll process; variable compensation slows down any attempt to automate the rest of the system.
5. Create business rules and logic
The programming of the business rules and logic will be done during the initial setup. Some aspects will need to be continually monitored and adjusted. Depending on the solution, this will be done by your team or a team from the solution vendor.
Programming business logic is a critical stage in the payroll automation process.
Most attempts at automation break down at this stage for one simple reason: payroll and sales comp analysts are not software programmers. The internal team often knows the underlying structure of the plan and the relationships between its components but lacks the programming expertise to understand how to implement that in the system. Many enterprise implementations — even if they hire external consultants — fall flat at this stage.
Prepare for Ongoing Change
Most enterprises overlook one element of automation: the need to change the rules or make exceptions in the years after it has been set up originally.
Businesses and business strategies change, meaning the underlying rules and logics need to change too. That isn’t a problem for simple fixed-salary or smaller companies, but things get exponentially more difficult as soon as sales commissions come into the equation.
Our best tip for ensuring the successful implementation of an automated payroll system is always to keep agility top of mind. Anyone can buy a toolkit with lots of features; learning how to use those tools to get what you want is a completely different job.
Don’t just ask if the payroll automation software can do what you want. Ask how that outcome will be achieved, and perhaps most importantly, who is responsible for delivering that outcome.