Tuesday, 4 June 2024

Step by step guide for Powerbi Mobile app deployment

 

App Link: https://app.powerbi.com/Redirect?action=OpenApp&appId=5aaa9abf-91b5-461b-9cb6-4252dbee6c6b&ctid=c036e821-d169-45e7-89a2-40b42dccb9cf

Username: frenchtrendz@mndatasolutions.onmicrosoft.com

Password: Bhaiya80$

 

Download Powerbi app from Google store





Choose your company: in this case ‘MNNBI’


Step 3: choose your report



Contact for  your own mobile app(integrated with Tally):

Apoorv chaturvedi

Email: apoorv@mnnbi.com;support@mndatasolutions.com

Phone: 8802466356

Pre requisite for Mobile app:

·         Tally serial number

·         Odbc should be enabled


Increase your prospects by using our inactive customers report

 There is a need in the current times to increase the number of prospects for the sales resources. Getting new prospects is a very tedious and time consuming job.

However a company is likelier to get business from its past customers than prospecting for new customers.

It is in this regards we have added ‘Inactive customers’ reports to our mobile app.



Why use our solution:

·         The data is stored on the Azure cloud. So it is extremely secure

·         The Tally data has been converted to DAX for extremely fast analysis. One can get mobile apps and dashboards with high complexity with very low refresh and reload times

·         The solution gives the ability to edit the dashboards on the cloud itself

What is the purpose of this report?

·         Identify those customers whose last transaction dates have been more than 90 days from the current fiscal date.

·         Identify the important and the less important customers based on the closing ledger balances. These customers have been segregated into the following categories:

o   ‘A’ category customers are those customers that form the top 25% of the ledger balances. We assumed the ledger balances as a criteria since we wanted to include all the historical transactions.

o   ‘B’ category customers are those that are between 10% and 75% of the total cumulative ledger balances

o   ‘C’ category customers are those that are in the bottom 10%.

The report also gives the rank of the given ledger by the historical balances.

A quick word regarding ABC classification:

·         The ABC classification has been done based on the historical voucher inventory analysis. Any partyledgername associated with the parent vouchertypename ‘sales’ in the historical analysis has been included in the list. This we could include both sundry debtors and sundry creditors in the list.

·         Once the ABC classification was done we identified the entire inventory movement by sales for the given set of items f the given partyledgername as under:




This report run by a slicer would straight away give the important items sold to the given partyledgername on a historical basis.

The same action has been replicated for B and C category partyledgernames.

 

Top A_category partyledgername: this customized report gives the top 10 partyledgernames  for the ‘A’ category(partyledgernames) and the values of the associated stock items as given under:


How to use the given report?

a.)    Identify all the A and B category party ledgernames by using the Turbodata-Powerbi module

b.)    Within each partyledgername identify the top items sold to the given party. Identify the items to be prospected for with each given party.

c.)     A customized mailing solution from Turbodata is also available for the end clients.

 

How to access this report:

a.)    One can access the given report using the following login features:

https://app.powerbi.com/Redirect?action=OpenApp&appId=5aaa9abf-91b5-461b-9cb6-4252dbee6c6b&ctid=c036e821-d169-45e7-89a2-40b42dccb9cf

 

Username: frenchtrendz@mndatasolutions.onmicrosoft.com

 

Password: Bhaiya80$

 

Deployment process: https://mndatasolutionsindia.blogspot.com/2021/04/step-by-step-guide-for-powerbi-mobile.html

 

 

How to go live for your company:

·         Contact Turbodata team for a customized tcp. Turbodata team needs the Tally serial number for the given firm

·         After the tcp has been loaded then the syncing between the client Tally machine and the Azure cloud shall take place through the secure Microsoft gateway for each individual account(details to be given by the M&N BI team)

 

Other features of the product:

a.)    The solution uses $$Loopcollobj functionality from Tally. That implies that all the loaded companies shall be extracted from Tally ERP 9.0 for the purposes of analysis

b.)    The analysis has also been done across all companies

c.)     One can edit the apps and the dashboards using the Azure edit functionalities


Cost of the solution per year : The entire mobile app for the end client costs Rs. 15000/-(fifteen thousand)+GST for the end clients

Price validity: 15th June 2021.

 

Contact details:

Apoorv Chaturvedi

Email: support@turbodatatool.com;support@mndatasolutions.com

Phone: +91-8802466356

website:https://mn-business-intelligence-india.business.site/


Turbofast Powerbi Reports from Tally

 Turbofast Reports: Powerbi

Consolidate and minimize your spreadsheets

M&N Business Intelligence India LLP(www.mnnbi.com)  is a leading analytics provider providing faster, easier and more convenient reporting from various ERPs. We have worked with clients like Medanta, PVR Cinemas, Kurlon etc for helping them optimize their operations through management reports. We help clients reduce their working capital requirements by helping design processes.

In most of the firms, accountants and managers are developing a large number of Excel spreadsheets manually for the top management. This is time consuming, manual and expensive. The large number of spreadsheets are due to the following:

  1.           Complexity of information required
  2.           Heavy load on Tally during processing large amounts of data

At the end of the process the management has to work with large number of complex spreadsheets.

For getting immediate benefit, we have made a product for   end clients

.Please fill out the form to enable us to get in touch with yourself for deployment:

https://forms.gle/dZGZDnqADxZquJrL9

Product download details: https://drive.google.com/drive/folders/13TOzjHxvTU8mEYuELYWmYj8MyLk-5V5f?usp=sharing

The solution is integrated with Tally and the solution is secure, easy, fast and very convenient.  Please contact  us for deploying the same on your machine.

Benefits:

We can help minimize the number of spreadsheets for the top management by the following:

·         Developing complex reports from Tally machine using Microsoft technology(DAX).  This is done with minimal deployment requirements. We have made the base product with a free version of Powerbi. 

·         The complex reports run atleast 50% faster than in Tally using Powerbi.We also provide data audit features with our product.

Our product gives reports with lower refresh and reload times than Tally on your machine or on secure Azure server. This shall help the client with the following:

·         Give managers required information in lower number of pages in an automated manner.

·         Extensive graphical representation can be done with ease and convenience.

        Advanced predictive analysis modules are also possible.

S      Sample for automation of the spreadsheets by our developer: Automation of spreadsheets sample 

            Next steps:

·        

·         The cost of the module is Rs. 7000/-(seven thousand only)+taxes as applicable. We would need the Tally serial number for the permanent license.

We shall be thankful to understand your requirements in depth  for reporting and Excel automation.

 Sample reports:


    1.) ABC Ledger analysis: This gives an immediate view of the important ledgers whose outstandings are due.

Please capture the important ledgers for receivable purposes from this report.

Application: This report is for collections teams


2.) Cost center wise sales report: here we look at the complete visibility starting from sales order, to sales invoice(including delivery notes) to billing outstandings. Use this report for the following:

a.) Understanding the sales order outstandings

b.) Look at the pending receipts

This report has been developed godown wise.

Application: use this report for collections team


3.)  A critical problem for the end users is that they do not have access to batch wise and godown wise inventory. The valuations are required for calculating the gross profit and for calculating the value of work in progress inventory. This report has been developed as a consolidated statement for stock summary, closing stock report.

Application: use this report for production planning team


5.) Payable and receivable snapshot has been developed for giving a one shot look into payments are receivables basis.




6.) Monthly, yearly and quarterly gross profit views for for those managers who want to automate the profit and loss statements.


6.) Stock item batch analysis: this view gives complete visibility to the inventory planning managers for batch wise and godown wise analysis



7.) Receivable and payable turnover: it helps analyze the credit quality of various parties.



8.) Daybook view: it gives views by voucher, voucher inventory and voucher ledger. One can make most reports from these 3 views.





Other views:  

·  

 

Price: Rs. 5000/-(Five thousand only)+GST

(This includes one simple spreadsheet automation)

Payment details: 


Beneficiary Name: M&N Business Intelligence India LLP

Bank Name: Axis Bank

Branch: DLF, Gurgaon[HR], Gurgaon 122002

IFSC Code: UTIB0000131

A/c Type: CHECKING

Account Number: 910020041330740

Inputs required: Tally serial number

 

 How to install Powerbi:

Download Powerbi from the following link:

https://www.microsoft.com/en-us/download/details.aspx?id=58494

 

Contact details:

Apoorv Chaturvedi

Phone: 8802466356

email: apoorv@mnnbi.com;support@mndatasolutions.com

Accounting analysts for improving collections and margins with ERPs

 Get accounting analysts to develop custom data entry forms over Turbodata using python


Need : Sundry debtors mapping at the time of entry into Tally and at the time of sales order processing.


Problem statement: many a times in various ERP systems various parties, that do not have a good payment history, are entered.

Sales to these parties can result in cash losses. 

The owners desire protection in terms of the quality of the parties that are entered into the system.


Why is an accounting analyst required?

a.) Custom ledger and item hierarchies for the end clients

b.) Complex Tally implementations

c.) Complex legacy collection issues

d.) Design of a custom data analysis and data capture system because there is no workflow system for most other ERPs

e.) Management to work closely with the accounting analyst for handling difficult accounts.

The process flow is explained below:



Solution(proposed to CEOs/CFOs)


a.) A thorough analysis of the client Tally ERP data across all the companies using the Turbodata tool(Tally to Powerbi). This tool does not require the installation of any database. The extract shall involve ledger hierarchy flattening, item hierarchy flattening for trial balance, receivable and payable and ledger movement analysis. A sample of the ledger hierarchy flattening is attached herewith:


https://drive.google.com/file/d/15Y1KjYLDJ3CFEeXS6B0pQSFG2HBg9qWj/view?usp=sharing


A similar exercise shall be done for item hierarchy flattening.

Reason for doing the analysis is the following:

  • Tally allows dynamic hierarchies. Understanding each level of hierarchy is critical for understanding the organizational dynamics.
  • The ledger and item hierarchies in each company can be different
  • We need to understand the balance sheets and bill payable and receivable history  of the companies very well. We need to identify if the pain points of various customers.
b.) Once the proof of concept and analysis has been done then the data is to be input into the Turbodata SQL Server datawarehouse. The SQL Server database shall help with the following:

  1. Incremental data load

                2. Maintaining historical data

                3. Customizing the inventory valuations

                4. Speeding up the data access using forms

                5. setting up secure access across various levels of hierarchy

               6. Generating customized python forms for data capture and workflow approval.

Data capture process


a.) A workflow for parties authentication(various hierarchies). In addition to the software systems we propose an online verification system using Turbodata SQL Server along with the python data entry forms. The verification will take place through the Turbodata entry forms at various levels of hierarchy. A sample data entry form is given herewith.



b.) A thorough credit check for new parties: attached bank statements, financial statements, details of credit bureaus

We shall look at quick ratio, Total turnover of the party for the prior fiscal year, profitability, cash profitability ratios.


c.) Possibility for the firm hierarchy to block the party in case the  credit history is not up to the mark. In such a case no entry into the Tally ledger master to take place.

Additionally the management can specify the following:

i.) Decide the hierarchy level from the group flattening at which the given party shall be handled.

For example Tally allows flexible hierarchies for a given party. The management can decide that a party with a large order values 

will be handled  by ledger level4, a lower order value will be handled by ledger level_5. In such a case the key stakeholders for the collection process 

shall be defined.



c.) Solution shall be deployed on client server. The custom hierarchy of the client to be taken into analysis process.

d.) Every party to have a set of stakeholders for ease of collections: this shall be ensured by having the mapping of each party at a particular ledger level 

during the data entry process. Additionally the following steps shall be done to ensure that collections take place in time:


Deploy checklists for the sundry  debtors based on balance sheet analysis from google

Decide the payment terms for the party.

Link the payment terms with the sales order entry process.

Block the sales invoice entry without the sales order entry. The credit check for the party to take place at the sales order entry level.



The above is the ledger entry form
a.) It matches with the ledger entry form with Tally.
b.) Additional fields and control systems shall be added based on the client requirements






Data validation at the sales order entry process entails 2(two) levels of credit checks:

  • Ledger: at the ledger level 
  • Item: this shall entail checking the inventory valuations at the batch, godown level for the required client. This has been done because many end clients require only specific movement types to be incorporated in the inventory valuation, remove the errors due to company splitting and get a more accurate closing stock valuation at the batch godown level


For example in the above form, for a party for which sales order has come the following 6 KPIs have been checked base don the Turbodata datawarehouse.

a.) Closing ledger balance of the party

b.) Total receivable as of date for the party. This entry comes from Billdetails and should match with the closing balance of the party.

c.) Receivable more than 90 days: This indicates the total receivable with due date more than 90 days

d.)  Cashinflow: this metric indicates the total cash inflow from the party

e.) Sales order pending value : this is the total sales order pending value for the party as of this date. This shall include the pre closed orders also,


Based on the same above inputs the hierarchy of the company can approve or disapprove of the order.


A similar form shall be created for the items(batch, godown wise analysis) to ensure that the required profitability level is maintained.



Name: Apoorv Chaturvedi

Phone:+91-8802466356

email: support@turbodatatool.com;support@mndatasolutions.com

Workflows and secured custom reports access(unlimited users)-Turbodata

   Workflows and secured custom reports access(unlimited users)-Turbodata

 

 Who are the stakeholders with whom you need to share data?


Are you a customer who is on Tally server?


Then let us see who could be the possible users of your data?


For most firms the user hierarchy details are in ledger hierarchy.




Say in the internal hierarchy there are 25 people  each across various levels of hierarchy.


say the company has 50 sundry debtors and 20 sundry creditors. SO the total people involved for sharing information securely are atleast 95.




Methods of contact by whatsapp, email are time consuming.






a.) Internal users: The number of internal users could be large. Various reports with row level security could need to be shared with various end users.


b.) External users: primarily under the primary group sundry debtors and sundry creditors. Currently They are informed by whatsapp or email intimations. However


Additional details that can be shared are as possible:


i.) Pending sales orders


ii.) Pending purchase orders


iii.) Pending bills statements: https://mndatasolutionsindia.blogspot.com/2023/11/issue-problem-statement-in-many.html


iv.) Intimation regarding new receipts


v.) Sales order input forms.


vi.) Open stock position for reordering: https://mndatasolutionsindia.blogspot.com/2021/01/sell-more-of-your-inventory-to-existing.html




Additional requirements for internal users:


a.) Workflow approval requirements for sales orders and purchase orders within the internal heirarchy:




Current solution:


a.) Most companies use third party apps for getting one or more tasks done: issues regarding data security and data confidentiality are open in many cases.


b.) Tally server companies also use third party apps




Our proposed solution:


a.) If on server in Tally then use the same to give access to multiple end users without paying additionally for each user by using own server.




b.) Design custom workflows for approval process




c.) Get custom reports in a very quick and customized fashion without paying any licensee fees






Process steps for the same:


a.) Use Turbodata Powerbi module to understand the Tally data using ledger hierarchy and item hierarchy flattening: https://mndatasolutionsindia.blogspot.com/2021/10/turbofast-powerbi-reports-from-tally.html


b.) Load the data into SQL server. For faster incremental data load indicate masters entries for data load. Optimize the reports in SQL server: https://www.youtube.com/watch?v=m8Lg6Og9bcU,https://www.youtube.com/watch?v=PI5DmU5IUSo,https://www.youtube.com/watch?v=B-odSuZkYhU&list=PLZPWImFqOkLsGgXPGf82nHPVaaZtJx4BL&index=1


using views and functions : https://www.youtube.com/watch?v=B-odSuZkYhU&list=PLZPWImFqOkLsGgXPGf82nHPVaaZtJx4BL


c.) Set up row level security at the database level:  


d.) Use Turbodata data capture forms to give custom reports at each level of stakeholders.


Name: Apoorv Chaturvedi


email: support@mndatasolutions.com;support@turbodatatool.com


Phone:+91-8802466356

Cost centre analysis add on-Tally

  Problem statement: Many companies require cost centre wise analysis(specifically with Tally). Their needs are as follows:

  1. Profit analysis per cost centre

                2. Understanding the billing implications for various projects. Material could have been             purchased but billing would not have been done.

Relevant industries: any project related industries for construction.

In order to meet the need for cost center analysis we have a software add on for revenue analysis.

For quick querying we have developed the same over SQL Server. The clients can  also develop workflows over the cost center module.

Why is the SQL Server module required?

a.) The logic as per the required revenue accounts has been developed in the database. We extensively use ledger flattening for the same.\

b.) In order to make the system more secure and customizable we have used python code.

c.) The client can specify the screen to be shown based on ledger and item hierarchy flattening.

Why is this module better than the Tally system:

a.) Customized revenue models have been developed

b.) Better control flows can be specified using item hierarchy and ledger hierarchy flattening. That is a CEO can contact his subordinates with ease using the module.


Explanation of the module:




 In the sample screen below we make a preliminary snapshot of a sample cost center module developed for a company.


a.) Company name is a drop down box: we can work with unlimited companies

b.) Cost center: these are the associated cost centers for the company that has been selected in the combo box above.

Hereafter we explain each of the list boxes below:

1.) Sales: this gives the total sales entries as per the sales accounts for the given cost center.

2.) Purchase: this portion can be further customized base don the ledger flattening by the end client. Various sub groups can be added. this indicates the total purchase(across all voucher types) for the purchase accounts.


(Note: delivery notes, receipt notes, purchase order and sales order have been excluded. The revenue accounts report from Tally has been used to generate the given numbers).

The sales and purchase entries include both ledger and inventory movements(so the journal entries for the sales shall also be counted)


Labour, Transport and other expenses have been extracted from the ledger flattening.


The net profit per project is sales-purchase-indirectexpense-direct expense for a project per cost center.


Other customizations that are possible: 

For example automated emailing and whatsapp feature across various ledger level hierarchies are also possible(the same is been done for a client given below):



Inventory movement analysis: In order to make the module more comprehensive and showcase more details for the project movement we also have added sections to showcase the inventory movement for sales, purchase. In the module the following actions can also be taken:

a.) Conversion of units can be done

b.) Link with bill receivable can be done

c.) Workflows based on project completion can be done.


For deployment of the module please contact:

Name: Apoorv Chaturvedi

email: support@mndatasolutions.com;support@turbodatatool.com

Phone:+91-8802466356

Estimated cost: (To be decided on case by case basis)

Website: https://mn-business-intelligence-india.business.site?whatsapp

Youtube channel: https://www.youtube.com/@mndatasolutions2042

Hire an accounting analyst for gross profit calculations

 Use the accounting analysts to help and calculate the cogs at perpetual  level across unlimited companies, 

unlimited branches and across the complete historical time frame.


Issues with the ERP valuations:

a.) The purchase value is normally calculated with ledger parent 'purchase accounts', excluding order entries. However in the schedule 3 balance 

sheet entries the CFOs and CEOs like to exclude the following entries also:

i.) Delivery notes and receipt notes

ii.) Stock journal entries(this calculation shall also include for dales entries)

iii.) Branch transfer details(a firm could have multiple companies or multiple branches. Stock transfer between them needs to be excluded)

iv.) Purchase of capital goods.

v.) Rate differences(this is included in ledger masters)

vi.) If the company is using cost centers with branches then the solution needs to be configured accordingly.



Similar calculations are to be done for closing stock value with the sales accounts.


Why use our accounting analysts:

a.) The accounting analysts shall come prepared with the software module(with and without database) on site

b.) The reports shall be automated and tamper proof(a critical requirement for many CFOs)

c.) The product will have features row level and filter level contexts and ledger and item heirarchy flattening

d.) The CFO and CEO can dictate how the solution is to be configured.


what else would the analyst do?


a.) Based on the custom calculations the analyst shall help adjust the MRP entries in Tally/other ERP.

b.) Help automate the distribution of MRP details to various internal and external stakeholders in the ERP


Why can't a normal accountant do this job?

a.) Normal accountants are trained only in Excel

b.) Large data limitations

c.) Routine tasks for the accountant

d.) Reliance of the accountants only on the numebrs given in the ERP

e.) Profitability is not a priority for the accountant




Number of companies that have implemented the solution: more than 15


Automation of profit and loss statements: https://www.youtube.com/watch?v=O8_3PMVBUv8

Usage of vlookup and sumproduct in Excel: https://www.youtube.com/watch?v=hpNs1uUYbTg

Sumif(Excel to DAX): https://www.youtube.com/watch?v=_pndahQvpvk

Row level and filter level contexts: https://www.youtube.com/watch?v=_3XlMxB_R0U

Lookup(DAX): https://www.youtube.com/watch?v=eIT4YAITIIg

Time intelligence: https://www.youtube.com/watch?v=6bowkQ4dcxk



Blogspot link: 

Turbodata: https://mndatasolutionsindia.blogspot.com/2021/10/turbofast-powerbi-reports-from-tally.html

Running  sum problem for inventory valuations: https://mndatasolutionsindia.blogspot.com/2021/01/resolving-running-sum-problem-using-dax.html




Weighted average valuation: https://mndatasolutionsindia.blogspot.com/2020/08/weighted-average-valuation-from-tally.html

Batch wise analysis: https://mndatasolutionsindia.blogspot.com/2020/01/turbo-analytics-batch-wise-bill-wise.html


Name: Apoorv chaturvedi

Phone:+91-8802466356

email: support@turbodatatool.com;support@mndatasolutions.com

Initial and Incremental data Load Template by M&N Business Intelligence-SAP Data Services

  INCREMENTAL LOAD/CREATION OF DIMENSION TABLE LOGIC AT SAP DATA SERVICES END CLIENT In this particular document we shall be looking at the ...