DATA QUALITY – Before you start!
Data Quality is important as the; technical, operational, commercial, legal, effectively all teams within your business will testify. The advancement of data and it’s integration into decision-making processes as small as what soap to order in your bathrooms to the enormous, should we purchase/insure/charter these ships?
This article can be summarised as stating the obvious: If you want to make good data driven decisions, you need high quality data, but much as I’d drive for a concise overview, there are some important variables to consider when applying this to your business, and here I’ll dive into these.
Before you start…
Measuring data quality is, for most organisations, and enormous and expensive component within the procurement process, and can vary massively in complexity from dataset to dataset, and there are a few things that can be done before even beginning the process which can have an enormous impact:
1. Set clear objectives for what you want to achieve with the data, now and in the future
2. Explore the landscape of the data available
Setting clear objectives
The impacts of this process should define any form of assessment of quality down the line. It will determine what metrics you choose as well as the approach taken. Some examples of the kinds of questions that large acquirers of data will use to guide this are:
Is this a new, or replacement requirement?
This fundamentally alters the entire process dependent on the answer. If the data is a replacement the questions revolve around the cost (Opportunity, Technical and financial) of change, and ultimately will revolve around the question: Is this the right decision to make to change? (For example: Will replacing our vessel tracking platform with a dataset that’s cheaper but slightly worse in specific regions have an impact on our ability to serve our customers?
If the requirement is new, the questions tend to veer towards the art of the possible, and while the variables around cost remain the same, you’ve got the added risk of a hypothesis not being correct, so understanding if suppliers would support a pilot project become more important. (Will adding a comprehensive vessel database enable our sales people to have more specific commercial conversations with potential customers? – should we test this out with a subset of the fleet?)
Is this a short term/long term purchase?
The length of the proposed engagement will have a profound impact on the decision, in particular the support required by suppliers. An example could be a one-off report required for a particular customer/board engagement that supports your thoughts on your market position, vs a long-term engagement to provide data for your ERP. The supporting arm of the supplier, and the reliability of their delivery mechanisms and customer service responsiveness matter significantly less than they would in the second example.
What impact is this going to have on your overall business?
The risk of any purchase is naturally multiplied by how much of an effect it will have on your business. Implementing a dataset to enable the backbone of your global pricing model or delivering mission critical information to your customers will require more studious and careful decision making than purchasing a small dataset to provide more effective marketing in most cases.
Explore the landscape of the data available
The market for maritime data solutions has gone through a thorough reshape in the last few years.
The introduction of a raft of sanctions and compliance programs (https://home.treasury.gov/system/files/126/05142020_global_advisory_v1.pdf), significant new emission targets (https://ec.europa.eu/clima/eu-action/transport-emissions/reducing-emissions-shipping-sector_en)alongside the advent of new tech (AI, machine learning, increased processing power, cloud technologies). All of which has produced a proverbial mountain of suppliers eager to provide there services.
Against this backdrop it can be difficult to understand what is available, and what the market means for each of these suppliers.
Take for example, vessel tracking data for which there are several suppliers competing to provide you with their services. In this context, you naturally have more leverage due to the ability to shop around, but if you were to narrow this search further specifically to satellite AIS as a component within vessel tracking, your options decrease dramatically to two suppliers, and you naturally alter your expectations on all facets of the service, as you have a binary decision.
Summary
All these actions can be taken before even receiving a data sample and provide a framework for determining what value your organisation needs to see from the data. This sets you up to understand what areas there are for flex within the scope, and what are the red lines.
Answering the questions above should also give you an indication of the financial value of the data to your business, which will give you more confidence in negotiating both internally for budget, and externally with suppliers as the quotation process begins.
Comments