Join us at The Whalies • April 10th
Get Your Ticket
Talk With Data: 10 Prompts to Use With Moby

Talk With Data: 10 Prompts to Use With Moby

Last Updated:  
May 8, 2024

I have to say, this will be a good one.

But first, if you haven’t read my blog on the fundamentals of chatting with Moby, check it out here.

In this article, I will walk you through 10 awesome prompts you can use with Moby, and how you can use the prompts to obtain actionable insights for your DTC business. 

If you haven’t already signed up for the waitlist. You can do so here.

Let’s get into it!

Prompt #1: Investigating Ad Performance

“Give me my Ad Spend, ROAS, CPA, CTR, CVR, AOV, CPC, and CPM from Meta Ads over the last 30 days.”

As you can see, Moby outputs the results of this query in a table, featuring the metrics over a 30 day period. The key trends or outliers may not be obvious from the data in this format, but with Moby we can easily convert this table into a number of data visualizations including a line chart. This is a great way to track trends in your performance over time and identify key points where performance may have peaked or fallen off.

Looking at the graph, we can see poor ROAS on April 7th as well as really strong ROAS on April 23rd. With this information, you can go investigate the attribution boards for each of those dates to dig deeper and determine what led to those swings in ROAS. 

You can customize your chart to display any combination of metrics from the original table.

Simple line graphs like these are a very powerful way to track trends and monitor performance across your business. 

With the “Add to” button below the chart, you can add this chart to any of your existing boards or you can create a new one. (you can do this with any chart or table!)

Moby also provided some high-level analysis along with the graph. Moby is trained to do this for every output. You can analyze the data deeper by asking follow up questions. Just like ChatGPT, Moby is fed all of the context from your current conversation and provides deeper analysis as you provide more context and data. 

Prompt #2: Forecasting

“Forecast my total revenue for the next 6 months. Use the last 30 Months of Data.”

Hold up, did Moby just create a forecast? That may seem too good to be true. So here’s the truth: our team has been working behind the scenes for over a year to develop a forecasting and planning tool for ecommerce brands. At the same time, we started to build the greater next generation of our product. What you’re seeing above is a powerful tool that analyzes the past 30 months of your data to generate a forecast. 

Key elements of the forecast include:

  • Total Revenue and Historic Revenue forecast for the last 6 months
  • Future total revenue forecast
  • Upper and lower confidence intervals

As part of the analysis below the forecast, Moby will provide the Mean Absolute Percentage Error (MAPE) which is a statistical measure used to assess the accuracy of a forecast model. It expresses the accuracy as a percentage, providing a clear and intuitive measure of prediction error. 

MAPE less than 10% is considered excellent. This indicates that the forecast model is highly accurate, with predictions deviating very little from the actual values.

MAPE 10 - 20% is considered acceptable for a good prediction model. This range indicates that the model's predictions are about 80% to 90% accurate.

MAPE greater than 20% suggests that the model may not be very accurate, and the predictions could be significantly off from the actual values.

Prompt #3: Benchmarks

“What is the median value of various channel-reported key metrics for health and beauty brands with a $100+ average order value (AOV) and a gross merchandise value (GMV) segment of $10M+ specifically for Facebook Ads, grouped by event date for the last 30 days”

That’s right, now you can access ecommerce benchmarks data with Moby. Feel free to customize the above prompt for your industry, AOV, GMV, and Channels of choice. 

For reference, here are the options for benchmark customization:

Industry: All, Art, Baby, Books, Clothing, Electronics, Pet Supplies, Home & Garden, Sporting Goods, Toys & Hobbies, Health & Beauty, Food & Beverages

AOV Segment: All, Less than $100, Greater than $100

GMV Segment: All, <$1M, $1-$10M, $10M+

Channel: Meta, Google, TikTok, Triple Whale Blended

Just like our above example, you can better see these trends as a line graph! With this clear visualization, you will notice that CPMs are on the rise and as a result ROAS has seen some large dips recently.

Prompt #4: Daily Order Revenue & Cumulative Contribution Dollars

“What is the daily order revenue, contribution dollars, and cumulative contribution dollars for the last month by day, considering the total spend on ads, total custom spend on ads, and total cost of goods sold.”

From the prompt, Moby generates a table including contribution dollars and cumulative contribution dollars. Let’s customize the visualization to a combo graph that more easily shows the relationship between cumulative contribution dollars and total order revenue:

Beautiful. Now we can clearly see our total daily order revenue on the right y-axis and the cumulative contribution dollars on the left y-axis. 

But what are contribution dollars?

Contribution dollars = Total Revenue - Variable Costs.

Your contribution dollars should at least cover your fixed costs each month to be in a good position for profitability.

So if you know during a month what your fixed costs will be (Salaries, overhead, bills, etc.), you can set a target cumulative contribution dollars that need to hit.

This chart allows you to visualize that progress and to see how key inputs like total revenue are impacting the your cumulative contribution dollars.  

Prompt #5: Discount Codes

“What are the top 20 discount codes used on Shopify platform for the last 30 days, based on the count of usage and total revenue generated from orders excluding the 'no discount”

Have you ever wanted to visualize the impact of all your discount codes in one place? Now you can. The query above is a great way to pull your top discount codes for a specific date period. Now lets convert this into something more visual:

Now we can see which discount codes are used most often, and how much revenue they are generating.

This chart is a great way to identify any unwanted discount codes that may be circulating around. Or codes tied to specific campaigns that are proving to be effective. 

Prompt #6: Daily Profit & Loss

“Show me total order revenue, total sales, shipping price, taxes, payment gateway costs, refund money, spend, gross product sales, total costs, net profit, gross profit, net margin, cash turnover, and gross revenue for the last 30 days?”

Need to granularly track your profitability on a daily basis? Moby’s got you covered. Let’s start with a prompt that pulls all of our key P&L metrics into one table. 

Now, we can convert this table into a customizable combo chart to visualize which metrics are having the greatest impact on our net profit. In the above chart I’ve added:

  • Net profit: Gold area
  • Spend: Magenta line
  • Gross Revenue: Blue line

We can start to see the relationship between revenue and profit. Adding more expenses would help us create a more complete picture.

But what if you just want a high level view of your P&L? On to the next prompt:

Prompt #7: High Level Profit & Loss

“What are the total sales, total costs, gross profit, total spend, and net profit for the last 30 days.”

With this prompt, Moby creates simple metric tiles to show the totals for each metric for your desired date range. Just like your summary page! Similar to data tables and graphs, you can easily add metrics tiles to any dashboard, or create a new one. 

Prompt #8: Sales Cycle

“What is the average time between the first and second, second and third, and third and fourth order of new customers that purchased in 2023?”

This prompt is a simple and effective way to determine optimal retargeting time for customers after their first and subsequent purchases. In our above example, the time between orders drops for each subsequent purchase, that's a great sign of customer loyalty.

We could also customize this query to analyze the average time between orders for customers who purchase a specific product first, or come from a specific channel, or even where they live. This would enable you to customize retargeting intervals for key segments. 

Just like data tables, we can also customize the visualization of metric tiles. Here, I converted them to a bar chart so that we can better see how the time between orders is decreasing between each additional purchase. 

Prompt #9: LTV

“What is the lifetime value (LTV) and customer count for new customers who made purchases in 2023, using the Linear All model and lifetime attribution window, grouped by channel and ordered by customer count in descending order?”

So you might know your average customer lifetime value, but do you know which channels have the highest LTV as well as which channel drives the highest LTV? With Moby, seeing that data is easy.

This can be a great way to identify promising channels to invest more in. In the above table, I can already see that Attentive has a strong LTV. But let's convert this into a graph so we can visualize more of the data. 

Now, I can clearly see which channels are driving the most customers, and which channels have the strongest LTV. Some of the channels with strong LTV might not be scalable, this chart is a starting place to identify prospective channels.

One example above is Microsoft. It’s not a major customer acquisition channel today, but it has a strong LTV and it’s a platform that has a large enough audience to consider investing in. 

Prompt #10: Intraday Performance

“Show me total order revenue, new customer revenue, returning customer revenue, total spend, return on ad spend (ROAS) for each event hour on for the last 7 days?”

And last, but certainly not least, let’s get granular and look at some intraday data: some of our key metrics broken out by hour. In the prompt above, I asked for the hourly breakdown for the last 7 days. This will aggregate the totals for each metric across those hours, so we can start to identify some key trends. 

With the data visualized as a combo chart, we can identify a few interesting trends.

First, our order revenue peaks around 9:00am and 3:00pm, primarily driven by new customer revenue. However, our ad spend peaks around slightly different times, 12:00pm and 7:00pm. It might be worth focusing more spend right before those peak revenue times to capture more customers at the right moment.

Also, armed with the knowledge of when new vs. returning customer revenue peaks, we can optimize the timing of our social posts as well as email/SMS campaigns for each cohort. 

That’s all for now! Happy prompting. 

The next generation of Triple Whale is intuitive yet powerful. And the possibilities? The team is discovering new ones every single day. 

If you haven’t already signed up for the waitlist. You can do so here.

© Triple Whale Inc.
266 N 5th Street, Columbus OH 43209