Innovation

 iVise: flexible, agile technology consultancy services, tailored to your business needs

Be innovative. We are told this often, but how do we achieve it?

Innovation thrives when people have ideas.  However, ideas are really just a hypothesis that first needs to be tested and that hypothesis comes in the shape of a question. So if ideas are just questions that need and answer then it stand to reason that the faster you can answer the question the more opportunity you get for innovation. However, people don’t always have the tools that can uncover the answer to their questions.

 

We are on the search for new value

Innovation

We know that not all ideas lead to value….. but some do.  The more ideas that pose questions the higher the chance of finding ones that create valuable initiatives.  However, If people can’t get answers quickly they eventually stop asking for them.  At this point innovation withers and dies.

To prevent this, you need the ability to find and analyse data in very short cycles.  Almost any question can be answered quickly by agile analytics, however, innovation is not just about a good idea, more importantly it is the transformation of an idea into value.  Without this it is just wasting time. 

We can spend precious time brainstorming good (or bad) ideas, but without testing them, they are just concepts without any evidence to prove that they would work. So, the question to ask is not “what’s your idea?” but “how have you tested it?” or “how do you intend to test it?”

If you don’t ask these questions, ideas will be born with arguments around the table and the biggest voice wins, not the best idea.  The key is to drive a culture in which ideas are supported with facts not feelings.

question mark 1872634 1280

So how do we test ideas?

With a targeted analysis of data we can support or kill an idea. Fact based answers allow decisions to be made and action to be taken with confidence and in this way, new ideas become valuable innovation.   Sometimes this does lead to yet another set of questions that need to be answered with new data and in this process, the journey to innovation is started.

Decision latency and the cost of answering business questions slowly

The longer you take to respond to innovative ideas, the less value you create.

In essence, there are three types of latency -

  1. Capture latency - How long, it takes to collect data to analyse a question
  2. Analysis latency - How long does it take to create information and insight from this data.
  3. Decision latency - How long to act on this insight – really two parts, deciding and then acting on the decision.

As shown in the diagram above, by compressing the time required to deliver information, we substantially increase the value of decisions.  When you can ask questions (or pose a hypothesis) and get answer in short timeframes it creates a culture of thinking and experimentation. 

How your current information platform and practices may discourage innovation?

The traditional Enterprise Data Warehouse system is designed for reporting, not iterative, single use analytics.   This is why an EDW has shortcomings when used for innovation.  The EDW suffers from the following:

  • High dependency on constrained IT resources. How often does a person with an innovative idea ask a question to which IS responds that the report would take too long, cost too much and is not a priority. In this case they resort to using MS Excel, Access or AWS themselves or even worse just forget about it.

  • Requirements and value are often unclear with innovation. It is hard to know what you need until you see it and realise that’s not quite it.  And even if you get a valid answer, that answer may only present another question. Data warehouses are built in a traditional waterfall method with very clear requirements they are not suited to short sharp iterative cycles.
  • You can’t keep everything in an EDW. BI platforms are expensive repositories of data, and need to run at speed. This results in aggregated data to a level of granularity that is assumed adequate and it only integrates systems considered material, its also purged of aged or irrelevant data.  If someone asks a question that requires access to data outside this scope the cost to change is normally too prohibitive for just one question.  Bringing third party data into a BI platform is generally not considered cost effect in most cases.

Right now, people with innovative ideas will need data that is not easily accessible.  These people need a more agile process where they can build ad-hoc analysis in short iterative sprints that create fast insight that allows feedback.  The iVise Agile analytics tool set and processes aim at reducing decision latency by collecting data analysing it and presenting insight in the short time possible.  To do this we have borrowed concepts from Agile project management and the scientific method .

The basic idea is simple, to do the least amount of work, and get the most amount of information.

The diagram below outlines part of the iVise method where we pose a hypothesis and use that to define scope and guide our data discovery for analysis.  From the fresh insight new actions can be agreed and experiments (or processes) created.  Then action is taken and measured for us to review to quickly decide if an impact has been made, or not.  If the action does not live up to the hypothesis, the initiative can be shut down quickly and very little cost.  However, if the action uncovers value or savings it can be expanded and funded as a new initiative with clear understanding of the ROI.

What are the six steps?

 

Here are the six steps:

  • Hypothesis proposed. Define the question. Experiments needs to have a clear question (hypothesis). This question is at the heart of the experiment.  The sole purpose of the experiment is to prove or disprove this question (and if it can’t then it has no purpose) A good experiment will tell you something, even if it’s something negative. If you already know the outcome, it is not an experiment, it’s a KPI.
  • Define the scope and investigate data. What data doe you have to answer the question and where is located.  List your assumptions. What kind of assumptions do you have? What are the things you are unsure about or don’t know? List them all. Identify the most critical assumptions. We have lots of assumptions on any idea or solution, but it would be difficult to test them all at once. Focus on testing just the critical ones.
  • Analyse and uncover insight.  Use this insight to help design the experiments.
  • Design and run the experiment. Keep it simple. Design your experiment so that you can start tomorrow. The idea is to collect as much as information with as little effort as possible. Forget surveys and market research
  • Action taken and measured. Collect data. Record everything: data you collect and record, will guide you further.
  • Review results and decide on next steps. Assess the impact of the experiment against its goals. What did you learn? What do you need to change? Change your idea/solution based on what you learned. Do you need to repeat your experiment? Do you need a new experiment? Will you move forward with the solution or do you need more data? Decide how you are going to move on.

These analysis projects are carried out using an ‘Agile’ time boxed approach to ensure an outcome is delivered within the approved time and cost constraints. It is not intended to build a complete production ready solution, but to instead prove that there is valuable insight to be gained by running and evaluating, targeted experiments. These learnings will be used to more accurately estimate the effort required to deliver business ready components.

How do we get started?

Step one is to run an ideation workshop

This workshop is used to surface up as many ideas as possible and dig deeper into each one to understand the value it could create and how we can understand them better.  We then look into what data we have available and the complexity involved in using it to test the idea.

Once we have list of candidate ideas the next step is to rank them based on value created vs complexity in testing them.

Step two

This gives us a shortlist. We then take the idea at the top of the list and scope it up for the first test.  It is designed just like a science experiment where we pose a hypothesis and go about proving or disproving its value.

Step three

Review the results to either launch an initiative, a business plan, or shut it down and pivot to another idea.  At this point you have a fact based decision to either keep moving forward or kill the idea.  The goal is to succeed fast or fail fast.  Both are good outcomes.

Step four

Rinse and repeat