Report Fault Finding
My Report Is Wrong!
We occasionally get a request to check a report because there is a suspicion, or a strong feeling, that it is wrong.
This article goes into Report Fault Finding and outlines what we do to check a report. Above all else, checking your data is valuable as it means you value the data and also staff will see that. Motivating staff to use the system and keep using it can significantly improve results.
Golden Rule – Please ask us for assistance after doing some checks
Yes, we have found bugs in reports in the past, but we get these fixed as soon as we can.
We have a high degree of confidence that our reports are true and correct. But it never hurts to check and if necessary, get a second opinion. So, if possible, do some checks of your own then call us here at Elite-ID.
The following suggestions are taken from our real life activities in looking for report system concerns.
Fault Finding Checklist
- Make sure all the data has been uploaded.
- For classic wand systems, data may not have been sent to WebEye.
- For AtSite, if users are not connected to the Internet, data may not have been uploaded.
- Check the Time Profile – if it is used. Are the times correct? Any overlapping times (super bad)?
- For analytical reports like Location Analysis, Coverage and the Location Analysis reports – this is vital.
- Compare Reports. Probably your most powerful tool.
- Use Location Attendance to display the data.
- Use a time range to expand out the scale.
- Try one location for a period.
- Get a gut-feel for the data.
- Location Attendance is your best data visualization tool.
- Use Center Visits report to inspect the raw data precisely.
- Return Time -vs- Gap Analysis can show significant differences.
- Isolate locations.
- Inspct them one at a time.
- Use Center Visits to check precisely.
- compare days of the week. Are problems related to who wa susing the system?
Check Time Profiles
Time profiles are described here.
Check start and finish times. If your report uses a time profile, this is the first thing we check. Are the time starts and finishes correct?
Check Targets. Are they correct? Are they correct by day and for each start and finish set? If missing, some reports won’t work (Location Analysis).
Use Location Attendance to check your time starts and finishes. Is there data where it should be? Which is correct? Check the staff roster against Location Attendance.
Check for overlapping times. If the 2 time ranges have any overlapping times between the ranges, this will make Return Time and Gap Analysis etc calculations very wrong.
Check Filter Time. This reduces the effects of multiple staff attending an area at the same time, and will change return time and gap analysis calculations. Filter time is explained here.
Many of our reports have other reports with similar calculations. You can compare them.
For example, Location Analysis has a return time calculation, as does the Risk Management report. You can do comparisons. Same for Gap Analysis. You can compare Location Attendance graph data to other reports as a quick check.
Using Location Attendance To Check Data
If you suspect a report is wrong, particularly if the report does calculation, then use the Location Attendance report to get a feel for the data.
Use a Time Range. This expands out the display so you can see more detail. For example, set from 9:00 to 17:30 to show trade time for the start of a week. This expands the data and makes the data much more understandable.
Try displaying one location for a date range. If you are concerned about a location over time, Location Attendance will show a date range for just one location. This can be a big help in showing trends. Use with a time range and expand the data etc.
Use Center Visits to check data
This is your ultimate tool. Center Visits shows you the exact logged data without any embellishment or calculation other that the times since last attendance. Use this report to show you precisely what is happening.
It Can Be a Huge Amount of Data. A Center Visits report for one location for a month may be 50 pages long! So, reduce the time range, reduce the date range, do the report for just one location. Zoom on on a location, zoom in on a date and time, you can see exactly what has happened.
Return Time -vs- Gap Analysis
These 2 calculation methods often show significantly different performance values. Your return time might be very good, your gap analysis may be poor. Do not be surprised if this happens. There is a good reason.
Return time is a simple calculation. It divides the number of attendances in a date and time range into the number of minutes in that range. It can produce surprisingly good results for what in reality is a poor performance as a result of this simplicity. The calculation is described here.
Gap Analysis is a more powerful calculation. It looks at the time between attendances then uses that to say what percentage of the time the attendances were compliant with your target. It is explained here. Basically, if every attendance is within your target time so there are no gaps greater than your target time, then your performance will be 100% compliant. If you have one or more gaps greater than your target time then your performance will reduce.
For Example. You have have a return time of 8 minutes. Your target is 20 minutes. Your Gap Analysis shows 70%. Do not be surprised.
So Which Is Correct? They both are. The simplicity of the Return Time calculation means it is a simple average. The Gap Analysis looks at times between each attendance and produces a more precise result. An example – one location we checked had an 8 minute return time which looked really good, yet its Gap Analysis Coverage value was 64%. There was a gap of over 100 minutes. Because there were so many attendances at some times, the return time looked great. In reality, the gaps were significant and the performance was not so good.
Isolate Locations – check one at a time
Checking one location is always easier. Make it easy on yourself. Check just one location in isolation. After you have checked the location, check another and from that, after a few repeats, the report will either be shown correct or not.
There are onbly so many ways that data can deviate from your expectation.
- Has all the data been uploaded? All wands? All AtSite users?
- Are beacons or PipTags sensibly placed to capture stadd data?
- Were all the wands or AtSite devices charged and actually used?
- heck the time profile start and finish and target values.
- Look at the Location Attendance graphical report data.
- Drill down to one location. Expand out time ranges. Do a detailed check of one location.
- Look at the Center Visits data and compare to your report.