Having worked with thousands of data analysts, you pick up a few things to help you go from good to great!
A common mistake people make is how they use date filters for their business reporting cycles, especially when comparing time based trends.
For example, if you are a business that sells to other businesses and are closed on the weekends, comparing one month’s sales to the next month’s is very misleading. Or take Feb with 28 days vs the following month of March with 31; those three days could easily impact your results by 9.7%!
That is a BIG difference!
To make things even more interesting some months you have more weekdays then others, so again tracking your performance Month Over Month can be very misleading!
So what counts?
A commonly used measure of time in business reporting is the 30 day cycle. While the reporting periods in each cycle can be broken up in discrete chunks, a 30 day time frame, consisting of just over 4 weeks makes for poor analysis. And that all boils down to the fact that your 30 day schedule may begin on a Monday, but your last reporting day will be a Tuesday, with your next reporting cycle beginning on a Wednesday. So, in fact, the periodicity is not following a natural cycle.
Let's compare that to a 24 hour, 7 day, or 28 day reporting cycle. This consistent time frame will, for example, always begin on a Monday and end on a Sunday. The data or trendline now lends itself to comparison - consistently every reporting period.
Towards an objective reporting cycle
The idea behind consistent time frames is that you break each monitoring period into discrete, equal chunks. Consistent time frames offer a more objective way to monitor and compare your metrics. Not only that, but each day in each cycle is parallel to previous cycles. Spotting patterns becomes much easier compared to the 30 day cycle, since you are essentially comparing apples to apples, Mondays to Mondays.
Is symmetrical reporting for everyone?
The answer is simple: no. Setting up time frames for your organizational KPI dashboard is about knowing the best way to monitor your metrics in order to increase understanding and, in turn, improve your decision making. Your KPIs need to speak to your organization's goals and improve your ability to achieve those goals. Rather than advocating blind adoption of consistent reporting periods, this article is about adding another tool (or question) to your KPI tool belt.
13 months, each 28 days does seem to add up quite nicely. Food for thought!
So now that we understand the unique power of 28 day cycle reporting, I thought it would be interesting to follow that up with a look at the period of your time frames. Just like a sine wave, these types of time frames are characterized by their repeating cycle - allowing you to compare the ebb and flow of your trend more accurately, chronologically speaking.
A quick guide to common time frames in business reporting
If you evaluate KPIs and business metrics using SMART criteria then you already know that the last item concerns your reporting time frame. When do you plan on achieving the goal set out in your KPI? Over what time frame will you measure progress towards this goal?
As you may well know, applying different time frames to your data can yield dramatically different results. For example, if you monitor leads on monthly basis, comparing your current month's performance (in which the data set is still populating) to your previous month's performance is difficult. It may be more appropriate for you to compare the current 30 day period to the previous 30 day period.
There are three important models for tracking metrics: historic, X-to-Date, and moving/rolling metrics.
Historic time frames
Historic time frames have a fixed start and end date and represent a complete data set. For example, last year, last quarter, last month, or yesterday. Historic metrics are fixed and can only be compared to other historic metrics. It's also worth noting that historic metrics tend to look at longer time periods, such as weeks, months, quarters or years.
Common time frames
- Last Fiscal Year
- Last Year
- Last Quarter
- Last Month
X-to-Date time frames
X-to-date time frames have a fixed start date with a moving or rolling end date. For example, year-to-date, quarter-to-date, or month-to-date. This type of measurement is common, particularly in sales, marketing and financial reporting. X-to-date time frames are more tactical as they provide information on your current performance.
Common time frames
- Year-to-date (YTD)
- Month-to-date (MTD)
Moving or rolling time frames
Moving or rolling time frames have a fluid start and end date. For example, the past 24hrs or past 30 days. This type of time frame is generally captured in real time to provide a current view of performance. It's important that when comparing two rolling time frames that you take care to ensure the reporting period is similar, eg, the same number of weekends and weekdays.
Common time frames
- Past 90 days (p90d)
- Past 30 day (p30d)
- Past 24 hours (p24h)
Originally Published: Nov 17, 2011
Originally published April 16, 2019, updated Jun, 18 2019