Not another success story
How to handle IT projects (part 2)
In the previous issue of Sollers Insights, we discussed the success factors of an IT project: joint action between business and development, involvement of a reliable implementation partner, cooperation at all levels of the organisation, and a rational approach to customisation. Today, we would like to share our thoughts around the need to set clear goals, performing inceptions, fast failures, and MVPs.
Clear goals
Even in agile projects, we need to know at the beginning what we want to achieve and how we are going to get there. It is not necessary to answer all the questions right away, but it is crucial to decide which ones are important to us. Let us take an example. We want to go on a three-week trip around Europe with the whole family. Before we start the preparations, we need to consider which countries we would like to visit, what we would want to see, set our budget, and agree on when we would like to go. With these answers in mind, we can start preparing. If, during these preparations, we find that any of our assumptions need modification – we must reassess whether the project still makes sense. We wanted to go in the summer, but it turned out that we had to postpone our holiday. Can we go in the autumn? If we have school-age children, it may be necessary to move the whole thing to the following year.
Inception
Inception is, or should be (as it does not always happen), the initial phase of an IT project. During inception, the customer learns about the product they’d like to implement, and their implementation partner learns about the planned project and business requirements. Inception provides an understanding of the key challenges; it is the time for strategic decision making and high-level planning of the project scope and schedule.
Is it possible to have a successful IT project without inception? Likely, yes. But going back to our example of a European trip, is it better to leave as we are, or should we plan something before we set off? Before deciding what is important to us, or how we would like to spend our time? Before performing a safety inspection of our car? Having prioritised, we can prepare accordingly. (Inception is also the time to make sure we have everything we need to complete our project). If we are on a tight budget, it may be worth looking at how much hotels and food alone will cost us to assess if we can afford the trip at all.
Both approaches – with and without a plan – have their advantages. No plan means no restrictions, but it also carries the risk of having to sleep in the car for several nights... Inception gives us the comfort and security of knowing that the most important aspects of our trip have been 'taken care of' and that there will be no unpleasant surprises along the way.
Fail fast
Can 'failure' be a positive thing? Well, yes – if we apply agile principles.
The 'fast failure' approach is based on the idea that a product or project should be constantly and continuously scrutinised. If it turns out that it is not meeting key objectives or that key assumptions need to be changed, then it should be stopped and rethought. The underlying assumption of fast failure is that the project needs to be continually revised. In this approach, you test assumptions and ideas as quickly and cheaply as possible, before investing too much time and money to change them. By failing fast, you can learn what works and what doesn't, and iterate on your solution until you find the best fit for your customers and users.
Creating a 'fail-safe culture' in an organisation is an activity that provides benefits in both the short and long term. It is probably safe to say that there is no such thing as negative feedback in IT projects where this approach is used. All feedback can be used to improve product or project performance.
There are two caveats here. Firstly, the failure in question is something we can afford, but do not aspire to create. We must be realistic and pragmatic. If we decide on a quick project with a limited scope and expect spectacular results, and then immediately migrate the entire client portfolio to the new system: failure is guaranteed from the outset; virtually irreversible, and no doubt with disastrous consequences for us. Secondly, failing fast is not always an option. If we must adapt a system to meet regulatory changes that come into force on a certain date, we don’t have the space to experiment and make mistakes.
If we are going to fail, we need to fail as early as possible to limit the cost and organisational impact. Going back to our example, if it turns out during the safety inspection that our car requires some repairs that we should have completed before our journey, and our budget does not take this into account, should we continue to waste time planning our trip and paying to book accommodation?
MVP
An MVP (Minimum Viable Product) is an early and maximally simple version of a new solution, sufficient to meet only the initial needs of the customer. The MVP allows you to assess the value of the product and set expectations for its further development. In other words, an MVP is the most basic version of a given solution that is useful to users.
The MVP approach aligns with the agile methodology, where customers and end users participate in product development and provide continuous feedback. MVP requires involvement at the project team level on both the IT and business sides. This ensures that product development runs smoothly, as the transfer of knowledge is continuous and the feedback is prompt.
The list of factors that determine the success of IT projects is very long and it is worth learning from the experiences of others and listening to the advice of people who have participated in similar projects; sharing knowledge is, after all, a good practice that should be encouraged. We will never come out badly if we do this while using common sense.
Krzysztof Ko?odziejczyk , Senior Managing Consultant
Technology in Insurance: Facts and comments?
Friederike Krieger writes about motor insurance in Germany in Versicherungsmonitor:
领英推荐
MSK anticipates a loss totalling EUR 1.5 billion for 2024.??
?
"Motor insurance is expected to post an underwriting loss for the second year in a row, and homeowners insurance could contribute to these challenges. Insurance companies will focus on increasing effectiveness and efficiency."
Marcin Mekler , Manager
?
Insurance Asia reports:
Synthetic data to power 40% of insurance AI by 2027
?
“The utilization of synthetic data is not simple. As of now, there are no engines to generate reliable synthetic insurance data to train machine learning algorithms. Synthetic data can carry inherent biases that may skew AI-driven decisions.”
Mateusz Stasiak, Data consultant
?
Figure of the Month?
?
*The amount of US-$ raised by the San Francisco based data annotation company ScaleAI in a Series F round from investors such as Amazon and Meta
?