The capabilities of advanced analytics, combined with an organisations’ data assets, can be used to support better, more informed decision making - creating a smarter business able to better deliver on its commercial ambitions.  In this blog, Ascent’s Rich Pugh reveals how focusing specifically on decisions helps businesses identify potential areas for data-led value creation.

4 steps to a great decision.

In my first article I spoke about the significant value that could be unlocked through decision science.  But how do we find a decision that could enable a successful data initiative?

In our experience, there are 4 steps:

1. Identify the role of decisions in your success.

Depending on the culture of your organisation, the role of decision-making in your success might not be clear. Perhaps strategic decisions are easy to identify, but day to day, unconscious and random decisions are less easy to see.

To foster the right environment for a conversation about decision science, we first need to help people to see this connection, and (re)evaluate their role in terms of a series of decisions. This can be done, for example, at the start of decision science workshops.

This allows us to better relate the success of an organisation to the overall ‘rightness’ of its decision making. When we start to see our organisation as a decision-making machine, we can start to have a good conversation about which decisions we should focus on.

2. Find the high-impact decisions.

Once our teams are comfortable identifying decision-making processes, we should start asking questions about impact. Good questions to start this include:

• What are the really important decisions we make that influence our goal?

• How often are these decisions made?

• How do we measure the impact from this decision-making process?

• What is the impact of a perfect decision being made?

• What is the impact of a wrong decision?

• How is the decision currently made?

• How are we doing today? Is there room for improvement?

This conversation should get us to a point where we can identify decisions where potential improvements would have a high impact. For example, we might identify that ‘how we decide how to create an offer for a customer’ is a critical decision with a lot of potential upside.

3. Understand the potential for change.

The next step is to look for constraints (mechanisms that would stop us from achieving improvements to a decision-making process). These typically relate to:

• Operational change: related to logistical change required to enable a new approach (e.g. solution would require a significant change to the phone system) and potential lead to additional cost

• Business rules: agreed rules that could inhibit any optimisation (e.g. agreed minimum and maximum stock levels)

• Culture change: where teams may not be bought into a more data-driven decision approach (e.g. whilst we could create a fantastic pricing engine, the pricing team are quite happy with their current approach and might not be bought into a change).

Using our example of creating offers for customers, perhaps there are rules in place that we have to follow that will restrict change in this area?  Or perhaps your customer service team have established working practices based on experience over time - and any suggestion of a process change might not be well-received.

4. Look for supporting data.

The last step is to look for sources of historical data on which a solution could be based. Decisions where you have rich data on the inputs, actions and outputs will typically stand a far better chance of success within a shorter time frame.

For example, if we’re looking at improving decision making around customer offers, do we have accessible data on historical customer offers created and the outcome this led to?

Finding the right decision to start.

While decisions provide a significant opportunity for data-led value generation, finding the right decision to start with can be a challenge. Following the 4 steps outlined in this post can help to identify high-value decision points - and ensure you have the right data, culture and environment to deliver a successful outcome.