Worpra Home

Hi, What data task can I help you with? Start with pasting your data below that you want to do something with. Note I can only take about 500 words in one go.

ß site

Get AI to Analyse Your Data and Draw Your Charts & Graphs

Worpra makes data tasks super simple by taking your written instruction in plain language and applying it to your data. There is no looking for functions in a spreadsheet program. You don't have to be a data wiz to get results from numbers. Even if you are a data person, a lot of tasks can be done quicker with Worpra.

Worpra AI is a new way to get data jobs done with the following benefits:

This tool lets you make nice charts and graphs that will impress your boss, or to include in presentations. So, whether you are a data person in accountancy, marketing, banking… or you simply want to sort out your spending, then try Worpra. It's free!

Tips for Analysing Data with AI

The best way to explain something is usually with an example. So, let's start with your first input to Worpra. Just type or probably paste your data in at the Data prompt. It doesn't matter if there is extra text information in there. Click the plane button, or Tab then Enter, and you are ready.

Worpra Example

Hi, What data task can I help you with? Start with pasting your data below that you want to do something with. Note I can only take about 500 words in one go.

Twitch revenue
Twitch generated approximately $2.8 billion revenue in 2022.

Twitch annual revenue 2016 to 2022 ($mm)
Year Revenue ($mm)
2016 275
2017 300
2018 880
2019 1230
2020 1890
2021 2675
2022 2800

Twitch users
Twitch had an average of 2.58 million concurrent viewers in 2022, a 6% decrease on 2021.

Twitch annual average concurrent viewers 2015 to 2022 (mm)
Year Concurrent viewers (mm)
2015 0.53
2016 0.61
2017 0.74
2018 1.07
2019 1.26
2020 2.12
2021 2.78
2022 2.58

Next, tell Worpra what you want to do with your data. In the example below, I want the AI to compare two tables and calculate a new piece of information using both tables. You really can ask anything. The AI just knows if the tables don't match exactly.

Thank you. Now tell me what you want to happen with this data. You can give example output if you like.

Produce a table of revenue per user.

Now we see some output. Just be aware, sometimes the AI gets things wrong, even simple calculations. When it makes mistakes, they are usually big and very noticeable, so just sanity check the responses as you go. It is a computer but it can make mistakes just like people! Ask it to correct and it will.

Sure, I can help you with that. Here's the table of revenue per user:

| Year | Revenue ($mm) | Concurrent viewers (mm) | Revenue per user ($) |
| 2016 | 275 | 0.61 | 450.82 |
| 2017 | 300 | 0.74 | 405.41 |
| 2018 | 880 | 1.07 | 822.43 |
| 2019 | 1230 | 1.26 | 976.19 |
| 2020 | 1890 | 2.12 | 891.51 |
| 2021 | 2675 | 2.78 | 962.59 |
| 2022 | 2800 | 2.58 | 1085.27 |

To calculate the revenue per user, we divide the annual revenue by the average concurrent viewers for that year.

Make this two columns of Year and Revenue per user that can be copied to an Excel spreadsheet.

If your requests are not very specific, you may get output that isn't quite what you were looking for. Simply ask Worpra to make changes to get just what you want. The comma separated data table below didn't fill separate columns when I pasted it to an Excel spreadsheet. I had to use the Data then Text to Columns feature but not too hard.

Yes, I can modify the table to ensure that each column is separated by a comma (,) or a tab ( ) so that when you copy and paste the table into Excel, the data will be automatically separated into columns.

Here is the modified table:

Year,Revenue per User ($)

You can copy and paste this table into Excel and it will automatically separate the data into separate columns.

What is the expected revenue per user in 2023?

Ever wonder where the linear regression feature is in your spreadsheet to make a forecast? How long does that take to do? Well, with Worpra, you just ask. Don't spend time working out how to do things, just ask in plain language.

I'm sorry, but I cannot accurately predict the expected revenue per user in 2023 as I do not have any data or information on Twitch's future revenue or user statistics. The revenue per user can be calculated by dividing the total revenue generated in 2023 by the average concurrent viewers during that year. Without any data for 2023, it is impossible to make an accurate prediction.

But there is a trend line in the given data you can use to predict the result for 2023?

Oh, so the AI didn't think it could do such a thing. If you are sure it can if it thinks a bit harder, just help it to think differently. It will have another go. Lo and behold, it did it!

You are correct. I can use the trend line to predict the expected revenue per user in 2023. Assuming that the trend line remains consistent, the predicted revenue per user in 2023 can be calculated as follows:

First, we need to calculate the slope and y-intercept of the trend line using linear regression analysis. The slope of the trend line is -16.141 and the y- intercept is 33503.7.

Next, we can use the slope and y-intercept values to predict the expected revenue per user in 2023, which is 1212.08 USD.

Please note that this prediction is based solely on the trend line and does not take into account any external factors that may affect Twitch's revenue or user statistics in the future.

Use chart.js version 2.9.4 to make a responsive line graph, width 100%, maintainAspectRatio: true. Produce a web page with all the code and no comments. Use #ba4949 for the line color. Add axis labels.

Now the juicy part, drawing charts. The AI engine knows how to make all kinds of graphs, charts and data visualizations. It just won't draw them for you. Don't worry, there is a solution. Ask the AI to make you a web page to show your bar chart, dual line graph, pie chart or something fancier perhaps. For most straightforward charts I recommend specifying chart.js as the web drawing package. You are free to specify anything or nothing. Also, because the AI was trained on an older version of the internet, I say to use chart.js version 2.9.4 as seen in my prompt question just above. It tends to work more reliably. In fact, the prompt above is a good starting point for most of your needs. Then Worpra will show you the code for a single web page that displays your chart (example below).

Here is an example code for a responsive line graph using Chart.js:

<!DOCTYPE html>
<title>Revenue per User</title>
<script src="https://cdn.jsdelivr.net/npm/chart.js@2.9.4"></script>
canvas {
display: block;
width: 100%;
height: auto;
<canvas id="myChart"></canvas>
var ctx = document.getElementById('myChart').getContext('2d');
var chart = new Chart(ctx, {
type: 'line',
data: {
labels: ['2016', '2017', '2018', '2019', '2020', '2021', '2022'],
datasets: [{
label: 'Revenue per User',
backgroundColor: 'rgba(186, 73, 73, 0.2)',
borderColor: '#ba4949',
data: [450.82, 405.41, 821.50, 976.19, 892.45, 962.23, 1085.27],
tension: 0.4
options: {
responsive: true,
maintainAspectRatio: false,
scales: {
yAxes: [{
ticks: {
scaleLabel: {
display: true,
labelString: 'Revenue per User ($)'
xAxes: [{
scaleLabel: {
display: true,
labelString: 'Year'

In this example, we used the 'canvas' element to create the chart and set its width to 100% and height to auto to make it responsive. We also included the Chart.js library and set the version to 2.9.4.

Inside the 'script' tag, we created a new 'Chart' instance and set the type to 'line'. We then defined the data for the chart, including the labels for the x-axis and the data for the y-axis. We also set the color of the line and the fill color for the area under the line.

In the options section, we set 'responsive' to true to make the chart responsive, and 'maintainAspectRatio' to false to allow the chart to fill the available width. We also added axis labels for the x and y axes.

You could paste the above web page from <!DOCTYPE html> to </html> into a Notepad file, then drop that file into an empty browser tab to see your chart. Looks good! You can even fiddle with the code to make changes. Worpra has simplified matters though and you will show your chart right there in your browser! Now use right click, hold finger, Snipping Tool or whatever to save or copy your chart. Good job!


How does Worpra AI work?

Worpra uses OpenAI's text completions 3.5 turbo model for its communications. This is commonly called ChatGPT and is the strongest AI around in terms of what it can do with text (and data). Worpra then adds its own functionality like its special chart maker commands.

Is Worpra suitable for non-data experts?

Yes, if you don't know a lot about data manipulation, analysis, and visualization then Worpra has your back. You can tell Worpra what you want to do with your data, or find from your data, just by talking to it. Of course, Worpra is a great extra tool for data experts too.

Can Worpra handle large data sets?

This can be a restriction in some cases. The data input is cut off at 3000 characters, about a page. This is because the ChatGPT 3.5 is restricted. The 4.0 model can do eight times as much but is quite expensive, so Worpra would need a paid for scheme which I am considering. If your response to a request seems to stop before it is complete, try asking Worpra to 'continue'. This often works. Otherwise your task will need to be broken into bite size chunks.

How should I talk with an AI for best results?

  1. Be clear and concise: Use simple, straightforward language to communicate your questions or requests. Avoid jargon, slang, or ambiguous phrases that could be misinterpreted by the AI.
  2. Provide context: If your query or request is related to a specific topic or situation, make sure to provide the necessary background information. This will help the AI understand your request and provide a more accurate response.
  3. Ask one question at a time: Avoid asking multiple questions or providing complex requests in a single sentence. Break down your queries into smaller parts and address each one separately.
  4. Be specific: Try to be as precise as possible with your questions or requests. Provide relevant details, such as dates, locations, or names, to help the AI understand your query better.
  5. Use proper grammar and punctuation: While AI models are designed to understand natural language, using correct grammar and punctuation can help improve the accuracy of the AI's responses.
  6. Rephrase if necessary: If the AI doesn't seem to understand your question or request, try rephrasing it using different words or providing additional context.
  7. Be patient and flexible: AI models are constantly learning and evolving, but they may not always provide the perfect answer or solution. Be prepared to do some additional research or ask for clarification if needed.
  8. Be aware of limitations: Keep in mind that AI models have limitations, including a lack of understanding of recent events or information not present in their training data. Make sure to verify any critical information provided by the AI.
  9. Maintain privacy: When discussing sensitive or personal information, be cautious about sharing too much with an AI, as data privacy and security are important concerns.

What kind of charts and graphs can I create with Worpra?

Worpra provides a work around for the fact ChatGPT won't draw charts and graphs for you. This is described above in the example where you ask the AI to make a web page containing the chart or graph, or use Worpra's special chart type commands. There are quite a few data visualization frameworks that can be used in web pages to make your data more visual. I have recommended chart.js in the example as I have found this reliable but you can just let the AI decide what the solution should be. Chart.js is capable of making line graphs, dual line graphs, bar charts, bubble charts, doughnut and pie charts, polar area charts, radar charts, scatter charts and mixed chart types. So let your creativity run.

If your chart needs amending in any way, like colors or labelling, then just ask Worpra to make a new page with your adjustments. Then, use right click, hold finger, or a snipping tool to grab your chart.

Can Worpra AI suggest insights from my data?

Yes, just ask it for insights. Worpra may reveal things about your data you wouldn't have thought of.

What industries or job roles can benefit from using AI with data?

These are just a few examples, as data analysis skills are increasingly in demand across a wide range of industries and job roles.

  1. Accounting: Accountants, auditors, and tax professionals use data to analyze financial records, ensure compliance and provide financial insights for businesses and individuals.
  2. Marketing and Advertising: Market researchers, data analysts, digital marketers, and SEO specialists analyze consumer data to optimize marketing campaigns, improve targeting, and measure campaign effectiveness.
  3. Technology and Software Development: Data engineers, data scientists, machine learning engineers, and software developers often work with large datasets to build and improve software products and algorithms.
  4. Finance and Banking: Financial analysts, quantitative analysts, risk analysts, and portfolio managers use data to make informed decisions about investments, risk management, and financial performance.
  5. Healthcare: Epidemiologists, health informaticists, medical researchers, and biostatisticians work with patient data to analyze trends, discover correlations, and develop new treatments or interventions.
  6. E-commerce and Retail: Pricing analysts, inventory analysts, and supply chain analysts use data to optimize pricing, inventory management, and logistics to improve efficiency and customer experience.
  7. Telecommunications: Network analysts, data analysts, and operations research analysts work with data to optimize network performance, monitor usage, and improve customer service.
  8. Energy and Utilities: Energy analysts, environmental scientists, and resource planners use data to optimize energy production, manage resources, and monitor environmental impact.
  9. Government and Public Sector: Policy analysts, statisticians, and researchers use data to inform public policy decisions, assess program effectiveness, and allocate resources.
  10. Academia and Research: Researchers, professors, and data analysts use data to test hypotheses, validate theories, and drive scientific discovery in various disciplines.
  11. Human Resources: HR analysts, workforce planners, and compensation analysts work with data to optimize workforce management, talent acquisition, and employee engagement.

You may just want help with your personal finances.

Is Worpra a free service?

Yes, Worpra will be funded by advertising shortly.

Is Worpra available on mobile devices?

The nature of data usually means more screen area is better. Worpra works perfectly as a mobile website though and you can use 'Add to Home Screen' to make it a web app. Worpra's' progressive web app (PWA) has its own folder here.

What languages does Worpra support?

You may already be getting your browser to translate Worpra into your preferred language. The AI will understand English, Spanish, French, German, Italian, Portuguese, Dutch, Russian, Chinese, Japanese, Korean, and many more languages.

Who is Worpra?

I am Worpra. That is, I am Gordon living in the UK and I am a real person with a LinkedIn profile. I enjoy making websites using the latest technologies to make people's lives easier and better. My other main project is ollo.com which uses AI to give people help and encouragement in achieving their goals.

With Worpra I want to bring data skills to people who thought they could not do this and provide a different and sometimes better tool to data experts to get work done. I saw that people don't think of a text completion AI as something you would use in the field of data. For so many tasks I realized it could both do the tasks and be your supportive assistant all in one. Hence Worpra, your data friend!

What is your business model?

Worpra will be advertising funded. It costs a significant amount to use OpenAI's AI to power Worpra so this is necessary. Even advertising would not cover the cost of using the latest version 4 AI from OpenAI. I may have a subscription service or usage credits if people want to use this. It would be more intelligent and allow a lot larger data input. Let me know in the form at the bottom if you could be interested in this service.

What is your security and privacy?

In brief, the data you post to Worpra is passed directly to OpenAI and not stored on Worpra. OpenAI is run on Microsoft Azure which has a good reputation for security. Millions of people are trusting OpenAI with information. Their terms say they keep requests to their service for 30 days before deleting. The information passed to them through their API, like Worpra uses, is not used to train their AI in any way, so it should never appear anywhere. Worpra currently doesn't'set cookies, or have 3rd-party cookies although may in the future.

What are the main differences between Worpra and traditional spreadsheet programs?

There are many differences between Worpra and a spreadsheet:

Worpra Spreadsheet
Having a conversation about your data. Finding functions to do things with your data.
Ask the AI to use its knowledge of how to do things. Ask Google how to get the spreadsheet to do things.
Let the AI uncover insights. Spot the insights for yourself.
All types of data manipulation, analysis and visualization available in one stop shop. Finite set of functionality then it is find another tool to get the job done.
Limited to small data sets at the moment. Handles largish data sets.
Can make mistakes so needs to be watched! Dependable goto tool. Sometimes human error. :)

If you have any questions, just send me a message below. I am pretty responsive and love getting your feedback and suggestions. I hope you enjoy using Worpra and bookmark it as a useful tool. I would be delighted if you tell your friends about Worpra too!