Troubleshooting your data

Let’s start a topic where we can share ideas about how review your data, and solve for differences from ERP to Phocas.

1 Like

When a difference in data occurs - what do you do?
Many times the difference will be visible by going to Transaction Mode and looking at the data from the most granular level. This may allow you to spot the smallest detail that is the root of the issue.

How do you search for the difference from your ERP?

I take a few steps when trying to troubleshoot. The first thing I actually do is look into where the raw data in Phocas is pulling from (what dictionaries in the ERP system) to make sure the correct data is going into the Phocas database. If I can’t figure that out, I’ll start running reports in my ERP system - usually at the most granular level - to compare to what I’m seeing Phocas pull into my query. If there are certain types of transactions that are/are not pulling, I know my problem. Otherwise, I check to make sure that all of the dimensions I’m drilling into are correct, and also the date ranges on my query in Phocas versus the date ranges in my ERP system.

If all else fails, I contact the Phocas support team! They have been very good at troubleshooting so far, especially since I’m a new user to the software.


Great suggestions @tim.janes! Cross checking the dates solves a lot of issues. Thanks for the feedback about support, they are great!


I spent months validating our Phocas Sales database which was built from the sales orders database from Pronto. I found it quick and easy to download the Phocas data into a spreadsheet and then the Pronto sales data from the general Ledger into another spreadsheet and do VLookUps in each spreadsheet to check for discrepancies.


Finding data discrepancies is detective work. And so, my dear Watson here are my elementary tips.

1: Use Phocas to do your analysis - it is the best tool in the world to interrogate data.

2: Start big. Do the monthly totals agree ? . one day’s transactions missing ? check the sync and build logs.
likely cause is a sync or build issue. Check your raw data and task schedules. Rule out what you know is “not” the problem

3: Check Dimesion groups. Do the group totals add up? fous on the areas with the biggest discrepancy
Check ### for ungrouped data
Data grouped incorrectly . Are Chilli flakes really cleaning products?
Check Credits
Check Net zero transactions.

4: Now Drill down to the transactional level. Is a transaction missing (or duplicated) or wrong - Are the differences always under , always over or varied?

5: What business rules (calculations) are applied to your data after it is extracted. Is the Rule correct?

6: Call Support. And, ideally, have some examples of some transactions that are “wrong” it will make our troubleshooting easier if support staff know what the correct data should be.

and finally. It is not unusual for our customers to uncover discrepancies in Phocas that may be attributed to some housekeeping required in your ERP. You might find Phocas helps identify issues in your source data.


This was a big one for us, Phocas has helped us identify too many areas in Pronto that require rectification!


I have seen recent examples where the customer had unknowingly categorised Canned Asparagus as Poultry and another customer had Chilli Flakes as Cleaning products. A simple amount of housekeeping in the ERP and all was well. We had a bit of a laugh together too. Who knew validation could be such fun :slight_smile:


Hopefully not too many! Great to hear you found Phocas useful in identifying areas that need attention in the ERP.

For me one of those ERP housekeeping tasks of merging duplicate products or customers will cause historical “…Missing from Data” in Phocas; so I setup alerts that search for those and when the numbers build up enough I drill in to see which past years need to be re-fed.



This has been my biggest struggle with Phocas. I cannot pinpoint where our discrepancies are coming from. Everything is close but nothing matches up perfectly, specifically our sales and inventory DBs. I think it mostly has to do with sales being put in with rebates or at no cost and cost being added later but I’m not positive. I’ve drilled in on the Phocas transaction side but I don’t know enough about the eclipse dictionaries and where our data is pulling from to do the same on that side. If anyone has any suggestions I’d love to hear them

I’m not really an expert but I can relate, and I didn’t want you to wait too long to see a reply. I understand Eclipse Dictionaries a little, and rebates. What I can tell you is that Phocas gets sales data from the Eclipse PSUB file. Certain Eclipse reports that you are running could be getting data from the LEDGER file, or AR file, etc. The Eclipse Business Summary for example gets the data from the GL file. And it is reasonable that the GL and PSUB might not match exactly, even within Eclipse itself. I got a document from our Phocas consultant that explained that specific discrepancy.