According to a Gartner study, about 40% of enterprise data is either inaccurate, incomplete, or unavailable.
This poor data quality translates into an average of $14 million per year in a ripple effect of financial loss, missed opportunities and high-risk decision-making. Also, it can impact your budgeting, financial planning and data management in a negative way.
That’s why we’re diving into different types of bad data, their outcomes and what you can do to prevent bad data from sneaking into your financial planning and analysis.
Duplicate data
When two or more identical records appear, this is considered duplicate data. What’s the big
deal? Surely it’s a case of just finding that error and removing it as quickly as possible, right? If only it were that simple.
When duplicate information appears for thousands of records, it’s not going to be an easy fix to delete a couple of errors in a spreadsheet by hand. And when 80% of spreadsheets typically contain errors, you’ve got your work cut out for you.
There’s more than just a time cost to duplicate data. The SiriusDecisions 1-10-100 Rule by W.Edwards Deming states that it costs $1 to verify a record as it’s entered, $10 to fix and $100 if nothing happens. Now multiply that rule by tens of thousands of duplicate data.
Plus, you’re going to have misrepresentation of inventory counts and multiple lines of credit for the same organization. Not to mention unnecessary billing that’s going to confuse your finance and sales teams and potentially cause tension between them. The last thing you want to happen is for a ‘he said, she said’ argument to break out across your teams that will bring cash flow to a grinding halt.
Conflicting data
This kind of bad data involves the same records showing differing attributes. When this happens, you have ineffective budgeting across teams, not calculating the right amount of investment for projects and supplier & employee payments not being received because of inaccurate accounting.
A cautionary tale of conflicting data happened with Uber. Back in 2017, the taxi firm had to pay out tens of millions of dollars to New York drivers because of a data error. Since 2014, Uber had been taking a 25% cut from the net fare of a ride (i.e. the price a customer paid minus the sales tax and other fees).
But Uber realized it had been taking from the gross share of a ride (i.e. sales tax and worker compensation fees). This simple accounting error severely damaged Uber’s brand reputation and employee trust.
Don't let poor data quality skew your financial analysis.
Incomplete data
Here we’re talking about missing attributes in records (i.e. customer contact email addresses or the last time they were contacted). There’s several issues that crop up with this. A risk of payroll not being processed and employees growing resentful. An inability to rank high/low-risk customers. Or extending credit to high-risk customers by mistake. These can all happen.
One place where incomplete information may crop up is in your CRM platform, especially if the responsibility for that data is shared among multiple people. Whether it’s a salesperson forgetting to input data directly after a sales call or finance not bothering to check if data has been updated, having too many (or no) cooks in the kitchen isn’t a good idea.
Returning to the 1-10-100 rule, it definitely applies to your CRM, as venture capitalist Tomasz Tunguz discovered. “If I’m lazy and don’t correct the error, later on that day, one of my colleagues might search our CRM for the company and comes across the erroneous record which he suspects is inaccurate. First, he will check his notes, then he will call me to verify and then he will change the record. The rigamarole has undermined his trust of my data and the 10 minutes he spent correcting my data entry are wasted. Worst of all is if I contact a startup to inquire about an upcoming fund raise with incorrect data. As a result, I could miss an opportunity to partner with a great company because of incorrect timing or lose credibility with the startup’s executive team. The cost to the firm could be in the tens of millions of dollars.”
Also, incomplete data will certainly have a detrimental effect on your marketing budget. Nearly a third of the average marketing team’s time is wasted on bad data, leading to 20 cents of every dollar draining away on marketing campaigns.
Invalid data
Data validity is about information being relevant to the business metrics that are being described. Invalid data is anything that doesn’t conform to standardization (i.e. a 10-digit phone number being the standard metric and then someone decides to add 11 digits). The latter is invalid.
The consequences of invalid data include skewed analysis, inaccurate financial figures and ethical violations. For example, a member of the sales team may put in the wrong product name and this erroneous information is then sent to the warehouse, adding to the total sales numbers for the day. The POS system is down, so it makes it difficult to get the sales numbers for the day, leading to incorrect sales figures for the month.
At the end of the chain, the Sales Director decides to change an input and manipulate the total amount of sales for the month. The board sees success on paper, yet the invalid data will only create more problems later on.
Unsynchronized data
Data synchronization is when information is constantly shared between one or more systems accurately. It’s of the utmost importance because it makes sure data is accurate. Plus, different teams can see multiple attributes at the same time and make the right business decisions.
Unsynchronized data creates team friction, a misalignment in goals and increases the risk of losing money in the long run. For instance, you might record a new sales deal in your CRM. But if it just sits there and doesn’t get integrated with your ERP system or other technologies, then how is finance meant to approve the deal or find out the credit history of the potential customer?
Now that you’ve seen the consequences of bad data, here are the surface-level benefits of cleaning your data and integrating it with your accounting systems.
- Streamlined data flow
- Less need for manual entry
- Fewer instances of data duplication
- More accurate budgets & forecasts
- Less wasted spend
- Higher project ROIs
4 tips to stop bad data from creeping into financial planning
Create a strong data management policy
Once you’ve recognized there’s bad data present and you’re ready to make the most of the above benefits, there’s a lot you can do. First, take the time to build a data management system that’s tailored to account for all the problems we’ve listed.
Here’s some of the things you can do:
- Set benchmarks for data validity: Determine specific metrics that will be used to measure data validity. This covers completeness rate, which is the percentage of expected data that’s in a dataset. There’s accuracy rate (i.e. the percentage of data that’s correct) and there’s timeliness rate, which is the amount of time between an event happening and attributes being included in the dataset.
- Improve data collection techniques: Make it easy for customers to share their data with you in several ways. Be upfront about how you plan to use their information, offer incentives (i.e. product discounts), keep data collection forms short and demonstrate that their data will be secure.
- Update existing data: Go through your CRM, ERP and other systems with a fine-tooth comb. Assess all information that you have and see what needs to be changed and plan for how you’re going to update it. For example, data-matching software can match email addresses against databases.
Clean your data
Scrubbing bad data is a given. But there’s no point chasing after the first shiny new data cleansing tool you see if you don’t understand how it works or how it fits your requirements.
Here are some points you’ll want to think about:
- Purpose: Do you need a tool that will provide total data management or are you only cleaning specific types of data? Are you looking to have multiple views of your data or do you need a single source?
- Ease of use: How long are you going to spend training teams on how to use the tool? How long do you think it’ll take to train employees so everyone can achieve the best outcome?
- Advanced features: Basic data cleaning solutions will point out missing data and errors. But some offer advanced data quality checking features (i.e. finding odd symbols, examining fields for incorrect numbers and automatic attribute standardization).
- Connectivity: Determine how easy it is for the data cleaning tool to connect with your existing technology stack. Can it connect to your CRM, email platform, MS Access, CSV/XML files? Can it do so easily or does your IT/technology team have to build new processes to enable this?
- Team effort: Remember that you aren’t just choosing a ‘tool.’ You’re investing in your team and are going to make their lives easier. So, anything new that’s brought in has to be the right fit for them as well. Consult multiple opinions and compare features and benefits from different tools.
Build a specialist team to train others
In some organizations, a Chief Data Officer (CDO) would be appointed to oversee a full data management strategy. But all this does is pile pressure onto the shoulders of one person and does nothing to solve the silo problem, which is a key reason why you hired the CDO in the first place.
It’s far more effective to hire a specialist team or task force with complementary skills. Imagine training a finance CDO, a sales CDO and a marketing CDO. Each of them lives and breathes the culture and language of their departments. They can inspire and upskill their colleagues while collaborating with their fellow CDOs to further improve processes.