Excelsheet-driven validation

Many Startups prefer to talk about their product. ‘We are still looking for a business model’, you sometimes hear them say. As if that business model is somewhere at the bottom of the tree, like an Easter egg, waiting to be found. But finding and developing such a business model is the heart of the matter, and often the reason for innovating after all.

Innovation around the Innovation Factory at Rabobank is mainly aiming at finding new business models for the coming years. New services with which we can earn money, but sometimes also optimization of an existing business model. A random example of that:

A financial planning tool (app, web, AI, VR, whatever) that allows users to plan their financial future. It is assumed that the use of this tool will lead to a better and more compact consultation and therefore cost savings, and to an increase in the number of people who choose to invest their savings.

You can therefore see the business model as a formula, and we work it out to a business case by setting assumed values for the various variables. Below the line we see the turnover, costs and profit projected for the coming years. The same example, but made more concrete:

Let’s take 100,000 people to download the app, 25% of whom will focus on investing, 5% of which will actually invest an amount of D, and that for 4 years, let alone, with an annual automated growth of 5%. And let’s save 30 minutes on the consultation, which means that we have 50% less consultancy costs. Then the final turnover — costs = etc etc

You can already feel it coming: we prefer to work this out in an Excel- (or other kind of) sheet, so that everything calculates automatically, with all parameters isolated in a row, like dials that we can turn:

Simple example of a parameter list. The color indicates the proof level.

But wait a minute: what about the product?

Most innovations start with a problem or a product idea. An app that helps you find second hand stuff. A new kind of jeans. A taxi service. A booking site. Or, as in this example; a financial planning app.

When asked ‘How is your startup doing?’ people often pull out the prototype and hope for happy faces. While at that very moment the original business case calculations might be on some harddisk, unopened for months.

Turning it around

I think it’s interesting to turn it around; to make the Excel sheet with the formulas and variables central to your startup, instead of your product. Designers may find it an unappealing idea, but any investor will nod in agreement.

And in that sheet you validate all the variables that you have estimated, and you do that one by one. Week in, week out. Back to the example:

This week we are going to do experiments to measure G: the minutes we save on a consultation. We are going to see whether we can actually live up to the ‘30’ that we have entered. It’s G week, peeps!

After about twenty consultations in which we measured the savings, we arrive at an average saving of 23 minutes with a normal distribution. We then enter it into the sheet again. The experienced data cruncher can calculate the normal distribution in the formula, so that the total result also has a normal distribution. It’s geeky time!

All parameters, one by one

We make the validated paramter green. And then we look at the next parameter in the sheet that we are going to validate and specify. Which parameter we then take depends on a few criteria:

  • What is the weighting or impact on the business case?
  • What bandwidth do we now use, and what probability does this have?
  • How much effort is it to specify that parameter?

Hypotheses

‘It makes sense, that’s just lean startup; prioritize and validate hypotheses!’ you might say now. And that is certainly the case, only the business case is put in the centre here, and we are automatically forced to translate the experiments into figures. Because Excel simply doesn’t do anything with texts like ‘considerable’, ‘much’ or ‘profitable’.

So automatically we are forced from ‘We believe that serious time savings can be achieved’, to ‘We believe that we can reduce the advice time from 60 minutes (the 0-measurement) to 30 minutes.’

That aligns with what investors say: bad news comes in adjectives, good news comes in numbers.

Customer satisfaction

Hey, wait a minute: customer satisfaction is the goal, right? That’s where success begins, isn’t it? Could very well be, and we can also give that a place in the sheet:

In this way of thinking, K (Customer Satisfaction) is a means to maximize the factor R (Retention) in the sheet, so that the average L (LTV) goes up, and the total C (CAC) can go down, partly due to the higher V ( viral growth).

Customer satisfaction is one of the most important parameters in your business case, that’s for sure. And designing epic products that make people happy is a great profession, don’t get me wrong: Ich bin ein designer. Super important too, but it is not the core of your (corporate) startup.

Different perspective

So instead of looking at the product and documenting in Excel what that product does, we want to set a target value (also a hypothesis!) in Excel and design the product/process and marketing that will get us there. So we are constantly busy feeding and perfecting our Excel sheet.

From an expectation or an assumption, the sheet gradually becomes a reflection of reality, via an approximation or prediction. Or maybe over time it becomes reality itself, but I’ll leave that philosophical question for some other blog.

--

--

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store