Articles on: AI & MCP

How to Use AI with your COUNT data

AI assistants like Claude and ChatGPT can help you work with your COUNT data using natural language.

Instead of manually searching through reports or transactions, you can ask questions, analyze financial information, and complete accounting tasks faster.

This article covers practical ways accountants and bookkeepers can use AI with COUNT.


Ask questions about your accounting data

Once connected to COUNT, you can ask questions in natural language and let AI retrieve the relevant information for you.

For example, instead of manually filtering reports, try asking:


“List all transactions from last month”

“Show overdue invoices over $1,000”

“Which customers paid the most this quarter?”

“Show uncategorized transactions from this month”


AI can retrieve information from COUNT and organize the results into an easy-to-read format.



Analyze financial activity


AI can help you quickly identify trends, unusual activity, and changes in financial performance.


Try prompts like:


Goal

Example prompt

Review expenses

“What were my largest expenses last month?”

Analyze spending

“Summarize spending by category this quarter”

Monitor cash flow

“Give me a summary of cash flow changes”

Spot unusual activity

“Find transactions that look duplicated”

Compare periods

“Compare expenses this month vs last month”


This can help reduce time spent manually reviewing reports.


Review transactions and accounting records


AI can also help review accounting data and surface information faster.

For example:


“Show transactions over $5,000”

“List transactions missing descriptions”

“Find invoices that are overdue”

“Show all transactions from Vendor X”


This is especially useful when reviewing large volumes of accounting data.


Use AI to take actions in COUNT


Depending on the task, AI can also help perform actions inside your COUNT workspace.


For example, AI may help:

  • Update information
  • Create records
  • Assist with accounting workflows
  • Reduce repetitive manual work


When actions require access or approval, you may be asked to confirm before the task runs.


Best practices for getting better results


AI works best when prompts are specific.


Instead of asking:

❌ “Show expenses”

Try:

✅ “Show expenses over $1,000 from last quarter”


Instead of:

❌ “How is cash flow?”

Try:

✅ “Summarize cash flow changes over the last 3 months”


The more context you provide, the more accurate and useful the results will be.


Example prompts to try

Here are some useful prompts to get started:

Transactions

  • “List all transactions from last month”
  • “Show uncategorized transactions”
  • “Find duplicate transactions”

Expenses

  • “What were my largest expenses this month?”
  • “Summarize spending by category”

Revenue

  • “Which customers generated the most revenue this quarter?”
  • “Compare monthly revenue trends”

Invoices

  • “Show overdue invoices”
  • “List unpaid invoices over $1,000”

Financial review

  • “Summarize cash flow changes this quarter”
  • “What changed most compared to last month?”


Updated on: 25/05/2026

Was this article helpful?

Share your feedback

Cancel

Thank you!