TallyPrime and Pipedrive integration

What moves between them

Sales and payment vouchers flow from TallyPrime into Pipedrive as deals and organizations. Every 5-15 minutes, ml-connector polls TallyPrime's Day Book for new or updated vouchers, matches customer and vendor names to Pipedrive organizations (creating them if needed), and upserts deals with the invoice amount, date, and payment status. Customer payment receipts in TallyPrime update the deal stage in Pipedrive (Won for fully paid, Open for pending). Pipedrive deal updates do not flow back to TallyPrime; the integration is unidirectional (Tally to Pipedrive only).

How ml-connector handles it

ml-connector runs a background agent that polls TallyPrime by sending HTTP POST requests with Export Data envelope XML, including date range filters and company name (which is case-sensitive). Because TallyPrime is local and single-user, all requests are sequential to avoid lock contention. Each poll batch compares returned voucher IDs against a stored high-water mark, so the next poll only retrieves newer records. For each new voucher, ml-connector looks up the customer or vendor name in Pipedrive organizations, creates a new organization if the name is not found, and inserts a deal with the voucher amount, date, and type (Sales or Purchase). Payment vouchers update deal stages: a Payment Receipt matched to an earlier Sales voucher moves that deal to Won. ml-connector stores the Tally company name, port, and optional password encrypted, and the Pipedrive OAuth2 token is refreshed automatically when a call returns an auth error. When TallyPrime is offline or the agent is not running on the LAN, the sync is simply deferred to the next poll interval.

A real-world example

A small business in India uses TallyPrime to manage customers, sales invoices, and payments, and employs a sales team using Pipedrive to track pipeline and close deals. Before the integration, the business manually exported customer and sales data from Tally each week, imported it into Pipedrive using CSV upload, and the two systems drifted as sales reps updated deal stages without updating Tally. With TallyPrime and Pipedrive connected, each new sales voucher in Tally automatically appears in Pipedrive as a deal, and when a payment is received in Tally, the corresponding Pipedrive deal is marked Won. The sales team always sees the current pipeline in Pipedrive, and the finance team always sees the current deals in TallyPrime without manual sync.

What you can do

• •Poll TallyPrime for new sales and payment vouchers on a configurable interval and sync them into Pipedrive as deals.
• •Map TallyPrime customers and vendors to Pipedrive organizations by name, creating new organizations automatically.
• •Update Pipedrive deal stages based on payment vouchers in TallyPrime (Won for paid, Open for pending).
• •Securely store TallyPrime company name, port, and password, and refresh Pipedrive OAuth2 tokens automatically when they expire.
• •Detect when TallyPrime is offline or the local agent is unavailable and retry the sync on the next scheduled interval.

Questions

Does data flow both ways between TallyPrime and Pipedrive?

No. Sales vouchers, customers, and payments flow from TallyPrime into Pipedrive. Changes made in Pipedrive (deal name, stage, custom fields) do not sync back to TallyPrime. The integration is unidirectional (Tally to Pipedrive only) by design, since Pipedrive is CRM-only and has no GL accounts or invoicing data.

Why must TallyPrime be running on a local machine, and what does ml-connector need to access it?

TallyPrime is a desktop application that publishes its API only over HTTP on a local LAN port (9000 by default). ml-connector must be deployed on a machine within that LAN or have network access to the Tally host. The company name, port (usually 9000), and optional company password are stored encrypted in ml-connector and used to authenticate each polling request.

What happens if TallyPrime is offline or the local agent stops running?

ml-connector will fail to connect on the next poll interval and defer the sync until the next scheduled attempt. There is no backlog or retry queue. When TallyPrime comes back online, the next poll will retrieve any vouchers created while the system was down, since ml-connector tracks the high-water mark of the last successfully synced voucher ID.