What’s holding your organisation back?
CEOs have it clear that businesses must become Analytics-driven. It is why they have been investing in resources and Artificial Intelligence (AI) and dedicating their own time in implementing Analytics programs or hiring skilled Data specialist. Many industry sectors are satisfied with the outcomes and the value they got form it.
In spite of this, many executives recognise that developing an Advanced-Analytics-driven company is frustrating; they do not see a way out, as only few Analytics initiatives are likely to involve effective scaling practices in the organisations. What’s not working? According to recent studies, there seems to be a failure pattern across industries of all sizes, proving that Advanced-Analytics attempts were not adapted well. Hereafter the causes of failure:
1. The team doesn’t have a clear vision of Advanced Analytics
Executive need to see-through Advanced Analytics as a powerful predictive and prescriptive tool. Studies revealed top managers fail to capitalise the central skills needed, neither do they work the spot-on troubles, and ignore the newest techniques. They ran lots of AI programs, but without scaling them.
The leader of Analytics initiatives in your company should set up a series of workshops intended for executives and all teams, to train them in the Advanced Analytics basic principles to clear misconceptions, something like an in-house academy or a continuing-educations program.
2. The value Advanced Analytics can bring is not enhanced.
Executives and employees alike must appreciate the value that Analytics initiatives can add to the entire organisation. It is necessary to identify and assess the viability of use cases through Analytics, to convince executives and all teams of the need for potential Analytics financings.
3. There’s no Analytics strategy, away from a few use cases
Don’t forget to identify potential use cases in which Analytics adds value. Building a digital platform comes next; connect with other industries that produce new products and services. Tackling Analytics in the wrong way is missing this significant opportunity, and in the long run, it will make more difficult to motivate your workforce.
4. Analytics roles are badly defined
Could you identify Analytics talent within your company, where it is located, how it is organised and whether they have the right skills? The proper way to approach it is to find Analytics talent with the specific skill sets. CEOs and HR teams must be in charge of these findings. Data Scientist hiring would help but, a skilled group, adequately ascribed in the right roles will do! You must carefully tailor job descriptions for all the Analytics roles needed, now and in the future. It will be useful to have an inventory of all talent working for you, to help to decide who meets those job qualifications, and to bring he/she aboard.
5. Organisations lack of Analytics translators
The person who best can start unlocking value is a TRANSLATOR with the right skills to interpret what Data Scientists are trying to say. Someone on the side of the business who can link all stakeholder. A Translator with a mix of business expertise, general technical fluency, project-management excellence, extensive company knowledge; and also, the training to understand mathematical models and to work with Data Scientists to bring out valuable insights. The shortage of this unique skilled individual, made many companies create their translator academies to train these internal candidates. Hire or trained translators right away!
6. Analytics know-hows are isolated from the business
Enterprises are driving Analytics capabilities into the centre of their business to create value. Over-centralisation creates conflict, leading to an absence of authority, as it causes data models disconnection. The C-suite must consider a hybrid organisational model, in which talented professionals, both from business and the Analytics side are blended. At some point, it would be advisable for the hybrid model to move aside to a more backing approach, granting above all the business’s autonomy.
7. Costly data-cleansing efforts
Business leaders think that all existing data should be clean before Analytics programs begin, but it could be expensive. The aim is to have one source of truth and a commonplace for data management. A Chief Data Officer’s (CDO) role is to coordinate data cleansing to drive the most valuable use cases while working to create a master-data model.
8. Analytics platforms are not “built to purpose"
Analytics platforms as Data Lakes, hold, process, and analyse structured and unstructured data. Studies say that more than half Data Lakes are “not fit to purpose.” These Data Lakes were built before considering the best ways to structure them, and designed it as a unit. It was a big mistake. By all means, it should be separated to tackle different types of use cases; CDO must ensure that data intake comes from multiple sources, that can be executed, and Analytics conducted on the platform, while Legacy IT system keeps working for the organisation transactional data need.
9. The mathematical impact provided by Analytics is not known
While enterprises are paying for Advanced Analytics, they cannot grant any impact from these initiatives. Both, the CEO and the Translator, must solve these issues, as it is their responsibility to spot specific use case delivering value. Refocusing their processes to automated systems for supervising use-case execution, embracing in progress model validation and advancements, could be the answer.
SUMMING UP: How to cope with the risks implied in algorithms?
It is crucial that businesses get a hold of a healthy Analytics transformation. Anticipate the Digital use of data, issues like the ethical, standard and regulatory requirement should be addressed. Close supervision to a risk-management program is the CDO responsibility. It is a significant challenge of collaborative work among CDO, CHRO and business-ethics, legal experts, plus the designated Translator, to set up checking services to be exposed straightaway and interpret the collateral effect of Analytics programs.
Dave Food
Subscribe to our emails & exclusive free content.
