Using an API to Export Your Own Numbers
Data Access · 10 min read ·
Pull listing data into your own sheet instead of copying by hand. How APIs work in plain terms, how to use one politely and keep exports accurate.
Copying numbers by hand is how most dashboards begin and how many end. It is slow, it breeds errors and it makes you dread the weekly update. If the numbers you need live on a platform, there is often a better way: ask the platform for them.
That is what an API is for. This guide explains the idea in plain words, shows what a sensible export routine looks like and offers habits that keep your data accurate and your use polite. The platform's developers page describes exactly what is available here; this guide is about the general approach.
What an API is
An API, short for application programming interface, is a defined way for programs to talk to each other. A web API lets one program send a request over the internet and get back data in a predictable format.
You can think of it as a waiter. You do not walk into the kitchen. You ask the waiter for a particular dish using the menu, and the waiter brings it. The menu is the API's documentation, the order is the request, and the dish is the response.
Most web APIs return data in a lightweight text format called JSON, which organises information as named values and lists. It is readable by people and easy for programs to handle. Spreadsheet tools and scripting languages can both read it.
Why use one
- Accuracy. No retyping, so no transcription mistakes.
- Speed. A refresh takes seconds.
- Consistency. The same fields every time.
- History. You can store each fetch and build your own record.
- Automation. A scheduled fetch can fill your dashboard without you.
For a launch, this means your Data tab can refresh itself with listing information, instead of you copying figures after each check.
What you can fetch
Depending on the platform, an API may provide information such as:
- Listing details: name, category, status.
- Public metrics: scores, rank, change and the dates they were checked.
- Link information: the DR shown for a site, its trend and when it was last checked.
- Categories and their members.
- Lists such as recent launches.
Each platform documents its own fields. Read the documentation, and fetch only what you need.
Nexaro Launch offers a free API and an MCP server for listing data. The developers page explains what each provides, how to get access and any limits.
What an MCP server adds
An MCP server exposes tools and data in a standard way that AI assistants can use. The Model Context Protocol is an open standard for connecting assistants to the data and tools they need. In practice, it means you can ask an assistant a question such as "which of my listings changed this week?", and the assistant can use the server to look it up.
It does not replace the API. It is a different door for a different kind of user. If you use an assistant in your workflow, the developers page explains how to connect it. As with any tool that can read your data, review what it can access and keep your credentials private.
A sensible first export
Start small.
- Read the documentation. Find out what is available, the format, any authentication and any limits.
- Get access. If a key or token is required, follow the instructions and store it safely.
- Choose one useful question. "What is the current DR and its change for my listings?" is a good start.
- Make one request with a simple tool: a spreadsheet function that fetches data from an address, a small script or an automation service.
- Look at the response. Find the fields you need, check their names and units.
- Put the values into your Data tab, with the date of the fetch.
- Check against the website. Open the page and compare one or two values by eye.
If the numbers match, you have a working export.
Store snapshots, not only the latest
When you fetch, add a new row with the date of the fetch. Do not overwrite the previous one. The value of an export is the history you build. A week's data tells you little; a year's tells you a lot.
Include:
- The date and time of the fetch.
- The values.
- The date the platform says the value was checked or last changed, if provided.
That second date matters. A value fetched today may be a week old, and you should know which.
Use it politely
APIs are shared resources. Using them considerately keeps them available.
Respect rate limits. Platforms often limit how many requests you can make in a period. Stay well under. If you receive a message that you have made too many requests, slow down.
Cache. If the data only changes weekly, fetch weekly. Fetching every minute wastes resources and helps nobody.
Fetch only what you need. Ask for the fields or records you want, not everything.
Identify yourself properly if the platform asks, with a clear name or key.
Handle errors gracefully. If a request fails, wait and try again later, not in a rapid loop.
Read the terms. Check what you may do with the data. Do not republish in ways the terms forbid, and credit the source where required.
Keep keys private
If the API uses a key or token, treat it like a password.
- Never paste it into a public document or a shared spreadsheet.
- Do not commit it to a code repository.
- Store it in a secure place such as a password manager or the secrets feature of your tool.
- Use a separate key for each use, so you can revoke one without affecting others.
- Revoke any key you think has leaked, and create a new one.
If you share a spreadsheet that calls an API, check that the key is not visible to others.
Check the data, always
A pipeline that runs by itself can also fail silently.
- Look for gaps. If a day's row is missing, find out why.
- Look for impossible values. A DR of 200, or a negative number of visits.
- Look for stale values. The same figure for weeks may mean the fetch is failing.
- Compare with the website now and then.
- Add a "last successful fetch" cell on your dashboard so you can see at a glance.
Never present an automated number to others without having sometimes checked it by hand.
Handle changes
APIs change. Fields are renamed, limits are altered and versions are retired. Protect yourself.
- Subscribe to announcements if the platform offers them.
- Note the version you are using.
- Keep a manual fallback. You should be able to update your dashboard by hand if the fetch breaks.
- Test after any change you make to your setup.
Respect people's data
Listing data is public information about products. If any data you fetch relates to individuals, handle it with care. Data protection rules apply to personal data wherever it sits. The Information Commissioner's Office publishes guidance on UK data protection for organisations. Collect only what you need, do not combine data in ways that identify private individuals without good reason and delete what you no longer need.
For most launch dashboards, aggregate numbers are enough, and personal data is best left out.
When a script helps
If you are comfortable with a little code, a short script can fetch, tidy and store data on a schedule. Keep it simple.
- One script, one job.
- Clear names.
- Logs that say what happened.
- Comments explaining why.
- Version control so changes are recorded.
If you are not, an automation tool or a spreadsheet function may be enough. The best system is the one you understand well enough to repair.
A weekly pattern
- Monday morning: the scheduled fetch runs, adding a row.
- After it runs: you glance at the "last successful fetch" cell and the new row.
- Once a month: you compare one value with the website.
- Once a quarter: you review whether you still need each field and the schedule.
That is all. Most of the value comes from the consistency.
What an export is not
An export is a copy of what the platform shows. It is not a fuller truth. Paid visibility, community votes and verified facts are kept separate and labelled, and the data you fetch should keep those labels. When you present numbers, keep the distinctions: do not blend a count of votes with a measure of paid exposure. Votes can never be bought here, and your dashboard should not suggest otherwise.
On this site
The developers page describes the free API and MCP server, how to get access and the limits, and the acceptable-use page sets out the rules for automated access. The launches and leaderboards pages show what the numbers look like to visitors, and the submit page starts a listing.
The short version
An API lets your sheet ask the platform for numbers instead of you copying them. Start with one question and one request, store dated snapshots, keep the platform's own check dates, respect rate limits, cache sensibly, guard your keys, verify the results by eye, keep a manual fallback and keep paid, community and verified labels apart. A small, steady export is worth more than a clever one that nobody trusts.
Questions and answers
- What is an API?
- A defined way for one program to ask another for data or actions, so you can read information without copying it by hand.
- Do I need to be a developer to use one?
- Not always. Many spreadsheet and automation tools can call an API with a simple setup, though some scripting makes it easier.
- Is there a free API here?
- Yes. This platform offers a free API and an MCP server for listing data. The developers page describes what is available and how to use it.
- What is an MCP server?
- A server that exposes tools and data in a standard way that AI assistants can use, so an assistant can look things up for you.
- How do I use an API responsibly?
- Read the terms, respect rate limits, cache results, keep keys private and only collect what you need.