AI for business in the Philippines: a starter guide
Short answer
AI for business means using AI software on a defined business task, such as answering customer questions, drafting content, sorting documents or summarising reports. The best first use is one task that is repeated often, has clear inputs and has a person who checks the output. This guide sorts uses by business function and size, lists what a business needs before starting, points to the cost pages on this site and covers the Data Privacy Act. Firm-level adoption data for the Philippines is old, so the guide says what is and is not published.
On this page
Key facts
- Common first uses
- Drafting replies and posts, answering routine customer questions, sorting and summarising documents, reading invoices
- What each needs
- One defined task, sample inputs from your own records, a named owner and a way to check output
- Who is involved
- The owner or manager who picks the task, the staff who do it today, and a developer only for custom connections
- Main costs
- Tool subscriptions, usage charges from the AI provider, setup effort and staff review time
- Main rule
- Data Privacy Act of 2012 (RA 10173) applies when customer or staff personal data goes into an AI tool
- Latest firm-level adoption figure found
- 3.02 percent of establishments, 2021 survey (PIDS)
What AI for business means in practice
AI tools do two kinds of work. They produce language, images or summaries from a prompt, and they classify or extract information from text and documents. A business uses both on tasks it already does. A shop owner asks an AI tool to draft a reply to a customer. An office manager has invoices read into a spreadsheet. A clinic sorts incoming messages by topic.
Two terms are easy to mix up. Using an AI assistant by hand, one request at a time, is one thing. Connecting an AI model to a workflow so it runs on every incoming message is AI automation. This guide covers both. The second needs more setup and more care, and the sections on requirements and rules apply to both.
Generative AI can produce wrong answers that read as correct. That risk is called hallucination, and it is the reason every use below includes a check by a person.
What the Philippine data says
Published data on AI use by Philippine businesses is limited and not recent. The page on AI in the Philippines lists each figure with its source.
- The Philippine Institute for Development Studies (PIDS) analysed its 2021 Survey of Innovation Activities of Establishments and reported an AI adoption rate of 3.02 percent for all establishments, 2.79 percent for micro enterprises and 3.59 percent for small enterprises (PIDS Discussion Paper 2024-35, December 2024).
- The same paper states a separate figure of 14.9 percent of firms using AI and machine learning technologies. The paper does not reconcile the two figures and is marked preliminary and unreviewed. This guide shows both and treats both as 2021 readings.
- The Department of Trade and Industry (DTI) publishes MSME statistics. Its 2024 page lists 1,241,476 business establishments, of which 1,236,908 are micro, small and medium enterprises (99.63 percent). Micro enterprises are 90.66 percent of the total, small enterprises 8.60 percent and medium enterprises 0.37 percent (DTI, 2024 Philippine MSME Statistics). The figures appear in infographics on the page, and an infographic on the same page gives 99.66 percent for the MSME share, which does not match 99.63 percent. This guide uses the breakdown counts.
In plain terms, nearly all Philippine businesses are small, and few were using AI when the last survey was run. A small team that starts with one task is in the same position as most of the market.
Uses by business function
The tables below are illustrations of common patterns. They are not client results and not a ranking.
Customer service
| Task | What the AI does | What a person still does |
|---|---|---|
| Routine questions | Answers hours, services and location questions from approved text | Handles complaints, refunds and anything not in the approved text |
| Message sorting | Tags a message as booking, price question, complaint or other | Replies to complaints and unclear messages |
| Reply drafts | Writes a draft in English, Filipino or a mix | Edits and sends |
Swipe the table sideways to see all columns.
Sales and marketing
| Task | What the AI does | What a person still does |
|---|---|---|
| Content drafts | Drafts captions, product descriptions and email copy | Checks facts, prices and tone, then publishes |
| Lead follow-up | Writes a first reply and logs the lead | Speaks to the lead and sets the price |
| Review summaries | Groups customer reviews by theme | Decides what to change |
Swipe the table sideways to see all columns.
Operations
| Task | What the AI does | What a person still does |
|---|---|---|
| Document sorting | Reads emails and files and routes them to the right person | Handles exceptions |
| Order updates | Looks up an order status and replies | Handles lost or damaged items |
| Meeting notes | Summarises a recording or transcript into action items | Confirms the action items are correct |
Swipe the table sideways to see all columns.
Finance and admin
| Task | What the AI does | What a person still does |
|---|---|---|
| Invoice and receipt intake | Reads supplier, date and amount into a draft entry | Approves entries and checks amounts |
| Payment reminders | Drafts a polite reminder from invoice data | Decides who is contacted |
| Report summaries | Writes a plain summary of a spreadsheet | Checks the numbers against the source |
Swipe the table sideways to see all columns.
HR
| Task | What the AI does | What a person still does |
|---|---|---|
| Job posts and policies | Drafts and rewrites for clarity | Approves wording and checks it against labour rules |
| Staff questions | Answers leave and payroll-calendar questions from approved policy text | Handles anything personal or disputed |
Swipe the table sideways to see all columns.
Hiring, pay and discipline decisions affect people directly. Keep those with a person. Staff records are also personal data, covered in the rules section.
Uses by business size
Solo and micro businesses. The owner does most tasks. An AI assistant used by hand fits best: drafting replies, writing product text, summarising a document. No connection to other software is needed, so there is little setup. Micro enterprises are the largest group in the DTI count above.
Small teams. A small team has repeated tasks shared across people, such as inquiries from Messenger or email, quotation follow-up and invoice entry. This is where connecting AI to a workflow starts to pay for the setup effort. The guide on workflow automation for small business maps the ten workflows small businesses often start with.
Mid-sized businesses. More people, more systems and more data mean more rules about who may use which tool and where data goes. A written policy for AI use, an approved tool list and a named owner for each automated workflow become necessary. A mid-sized business may also need custom connections to its accounting, CRM or HR systems.
What a business needs before starting
A defined task. Write the task as one sentence with a clear input and a clear output, for example “read each supplier invoice and enter supplier, date and amount into the ledger sheet”. If the sentence needs an “and then, depending on” the task is two tasks.
Clean enough data. AI reads what it is given. Price lists with three versions, customer records with duplicate names and policies kept in staff memory all produce weak output. Gather the current price list, the current policy text and 20 to 50 real examples of the task, with the right answer for each.
Someone responsible. One named person owns the task, reads the results and decides when to change it. Without an owner, a workflow keeps running after the business changes around it.
A way to review output. Decide what a correct result looks like, then check a sample of results against it. For a new workflow, check every result until the results have been right for a stretch of real use, then move to a sample.
Costs to expect
Costs fall in four parts, and different parties bill them.
| Cost | What drives it | Where to check |
|---|---|---|
| Subscriptions | The tool, the plan and the number of users or runs | The vendor’s own pricing page. See n8n pricing in PHP and HighLevel pricing for the Philippines |
| Usage | How much text, how many images or documents the AI model processes | The AI provider’s current rate card |
| Setup | How many systems are connected and how many exceptions the task has | A written scope before work starts |
| Staff time | Review of outputs, fixes and changes to the workflow | Your own records of time spent on the task today |
Swipe the table sideways to see all columns.
This page prints no prices. Prices change, so this site prints a figure only from the vendor’s own page, with the date it was read. Compare any quote against what the task costs your team in staff time today.
Rules and responsible use
This section is general information, not legal advice. Ask Philippine counsel about your specific use.
Data Privacy Act
The Data Privacy Act of 2012 (Republic Act No. 10173) governs how personal information is collected, used and kept. Section 11 states that processing is subject to the principles of transparency, legitimate purpose and proportionality, and that personal information must be collected for specified and legitimate purposes. Customer names, phone numbers, messages, invoices and staff records are personal data.
The National Privacy Commission (NPC) issued NPC Advisory No. 2024-04 on 19 December 2024, on how the Act and its rules apply to AI systems that process personal data (NPC Advisory No. 2024-04). Read it before sending customer or staff data to an AI service. Habits that follow from the principles:
- Send the AI tool only the data the task needs.
- Check where the provider stores and processes data and what its terms say about using it.
- Keep a record of what the workflow did, so a mistake can be traced.
Customer disclosure
No official source read for this guide states a general rule that a business must tell customers when an AI tool drafted a reply. Telling people when they are dealing with an automated system, and giving them a route to a person, is good practice and fits the transparency principle in the Act. The platform you use for messaging may set its own rules on automated messages, so read its terms.
Accuracy review
A person reviews what the AI produces before it affects a customer, a payment or a record. Check prices, dates, names and any claim about a product, a service or a person. Keep a sample of outputs and the correct answers, and compare new results to them whenever the tool, the model or the instructions change.
A first project, step by step
Choose
Pick one task from the function tables that your team repeats often. Write it as one sentence with a clear input and output, and name the owner.
Test
Run 20 to 50 real examples from your own records through the tool by hand, with a person sending the results. Compare each result with the right answer.
Measure
Count how many results were right as they stood, how many needed edits and how many were wrong. Note the staff time spent on review.
Decide
Keep the task as an assisted step, connect it to a workflow, or stop. Move to automatic sending only for case types where results stayed right, and keep review for the rest.
Add a second task only after the first has an owner and a routine for review.
The counts in the Measure step belong to your business. This site does not publish a general accuracy figure, because none applies across tasks and tools.
What is not published
- A current official national rate of AI use by Philippine firms after the 2021 survey year was not found.
- The PIDS documents do not explain the gap between the 3.02 percent and 14.9 percent figures.
- No Philippine-specific figure for business spending on AI tools was found from a primary source.
- The DTI MSME page does not state the source of its 2024 counts in the page text that was read.
- No official requirement to disclose AI use to customers was found in the sources read.
Method
Figures come from the publisher’s own page or document, shown as printed with their period. The uses and tables above are illustrations assembled from the task patterns in the sources and vendor documentation. They are not case studies and no result is claimed. The editorial team has not run these uses for a client. See the editorial policy for how pages are reviewed and corrected.
Frequently asked questions
How can a small business in the Philippines use AI?
Start with one repeated task that has a clear input, such as drafting replies to common questions, sorting inquiries or reading invoices. Use an AI assistant by hand first, check the results against your own examples, and connect it to a workflow only after the results are consistently right.
Which business tasks suit AI first?
Tasks that repeat often, start from a clear input and have a correct answer you can check. Customer questions, content drafts, document sorting and invoice entry fit this pattern. Decisions about hiring, pay and money stay with a person.
What does a business need before using AI?
One defined task, sample inputs from its own records, current price and policy text, a named owner and a way to review output.
Which rules apply when AI handles customer data?
The Data Privacy Act of 2012 (RA 10173) applies to personal data, and NPC Advisory No. 2024-04 covers AI systems that process it. Send only the data the task needs and check where the provider stores it. Ask Philippine counsel about your case.
What is AI in business?
Software that reads, writes, sorts or predicts, applied to a business task. A person defines the task and checks the output.
Sources
All sources were read on 12 Oct 2026.
- Philippine Institute for Development Studies, Discussion Paper Series No. 2024-35, Readiness for AI Adoption of Philippine Business and Industry, December 2024.
- Department of Trade and Industry, 2024 Philippine MSME Statistics.
- Republic of the Philippines, Republic Act No. 10173, Data Privacy Act of 2012, text as hosted by the Arellano Law Foundation.
- National Privacy Commission, NPC Advisory No. 2024-04, Guidelines on the Application of Republic Act No. 10173 or the Data Privacy Act of 2012, its Implementing Rules and Regulations, and the Issuances of the Commission to Artificial Intelligence Systems Processing Personal Data, 19 December 2024.