Your boss walks over to your desk. Everyone has been busy preparing for the new budget request for next month’s advertising spend.

Hey, hope you have some time today to help?
Just pulling together the final presentation
We could do with a forecast - say $9k spend on FB and $6k on TV?

Per day For the next month
I’ve sent you the model, so you should have everything you need

Scenario planning is a powerful tool for predicting the future and making decisions about it. It involves adjusting marketing inputs to see what kind of outputs we’re likely to get, based on realistic assumptions about the world.

The process of scenario planning starts with understanding the elements of the model. These can include the time horizon, the industry, the competition, the market size, the customer base, the product offerings, and the pricing. Once these elements are established, a “best case” scenario and a “worst case” scenario can be created.

The next step is to analyze the data and make assumptions based on the data to build a believable scenario. This process involves running the model to see what the results are.

The goal of this skill is to help you understand the basics of scenario planning and to learn how to build a believable scenario. This will involve learning how to use the different elements of the model, the different types of scenarios, and the process for building a believable scenario. It will also involve understanding how to use the model to make informed decisions about the future.

Once you’ve mastered the basics of scenario planning, you can use it to your advantage in a variety of situations. For example, you can use it to make decisions about marketing budgets, product launches, pricing strategies, and more. You can also use it to analyze the impacts of different scenarios and compare them to each other to determine the best course of action.

Ultimately, scenario planning is a valuable skill for anyone looking to make informed decisions about the future. It allows you to make realistic assumptions about the world and calculate the likely outcomes of different scenarios. With proper practice and understanding, you can use scenario planning to make more informed decisions and better prepare for the future.

Scenario Planning

0:01 Hey, let's talk about what if scenarios. So what if scenarios are basically you know, forecast that you can do when you've built a model.
0:10 So in this case, we're talking about a marketing mix model, which I'll show you in a second. But let me show you what I'm talking about in terms of what if scenarios.
0:18 So say for example, you want a forecast, an next 30 days of revenue you have some model about how the world works, and that's what we have up here.
0:27 Explain that in a second. We, we got these values from a marketing mix model, which I'll also explain. But what if scenario is where you plug in different inputs and then you get the outputs and it tells you if your model's correct, it tells you how much, you know, revenue or how many subscribers, et cetera you expect to get.
0:46 So for example, if we spend, you know, $7,000 a day on Facebook, 8,000 on tv, and then you know, this is how many Sundays, Thursdays, Fridays, and Saturdays on average there are per week 0.14.
1:02 And then this is just a constant value the baseline of sales. Then you plug all these things in and you get the revenue figure at the top up here.
1:12 And we could change this, right? So we could say, actually, we only wanna spend $3,000 on Facebook per day. So what would our revenue be?
1:20 And it's gonna drop pretty significantly, right? So the figure we got here, 7,042 precise figure. We got this from the average of what we were spending in the last seven days you know, in our data over here.
1:35 So let me take a look at this. And those last seven days this is for Facebook ads. You can see you know, on average we're spending about seven K.
1:45 So this is our data. We have Facebook cast revenue date and then this is what we use to build the model.
1:50 So the marketing mix model is here has a pretty good fit. And you know, this isn't a marketing mix model in course, but just so you kind of understand how to use the outputs of the marketing mix model you know, we can see that it's pretty accurate.
2:03 Like the moral in for past data has been predicting pretty close to the blue revenue line there. The red is predicted these are the errors, and then these are the different contributions of the different channels.
2:15 And you can kind of see over here, it's pretty spiky with the day of week trend.  But then you know, have like a baseline cost as well, or baseline revenue as well.
2:26 And then the rest is, you know, from Facebook, et cetera. Cool. So that is the model. So when you build a marketing weeks model one of the things it does for you is pull out these coefficient.
2:38 So the coefficient is the value for that parameter. And it says, in this case, every dollar you spend on Facebook brings you $4 20 in return.
2:49 It says same for tv. But this time it's lower. It's four 40 cents for TV for every dollar spent. So it's actually unprofitable to spend on tv in this case.
3:00 It says on Sunday we lose $2,000 $2,450 on average. So Sundays are our worst days. But then say like, Saturdays are our best days, we make an extra 5,900 on Saturdays.
3:13 I guess the weekend is, is big for us. Constant is what we would get if it wasn't Sunday, Thursday, Friday, Saturday.
3:21 And we weren't spending anything on ads. So we can assume 11,000. Now, this is a model, A model is a simplification of real life.
3:29 Obviously, you know, these things change over time. Like, you know, Facebook performance might go up or down. TV performance might go up or down.
3:37 This might be a model incorrectly. But I would say that a model, a multi mix model is a way better type of model to use for this than what people normally do, which is just kind of make assumptions and extrapolate them over time.
3:52 So this model is like pretty significant, like all the p-values here you know, for, for what where they are. You know, they're, they're all statistically significant.
4:04 So I think this is like a pretty good approximation of where we're gonna get and it's much better than just like random guessing.
4:09 Alright now this prediction here, this is how we worked out the error rate. This is figuring out the contribution for each shells.

What did the marketing mix model reveal about the performance of TV ads?

What does the constant value in the model represent?

What is the main purpose of scenario planning?

What revenue do you get if you allocate $9k to Facebook and $6k to TV per day?

It’s amazing just how many companies forecast future performance by extrapolating in Excel. If you want to avoid unrealistic targets or to really know how much to spend to hit it, you need a model.

Vexpower Learning Certificate of Completion

Congratulations envious-robot-49 💪