ANALYTICS CAN FACE LOW ACCEPTANCE WHEN THE DEVELOPED SOLUTION IS NOT ALIGNED WITH CUSTOMER EXPECTATIONS :
Originally Posted: Analyticsindiamag
Today, analytics remains limited to the top level of hierarchy in most enterprises. Adoption is typically better at service organisations such as Banking & Financial Services, IT enabled Services organisations, compared to traditional industries. In conventional industries, analytics culture is still MIS-driven. Here, business users are not encouraged to do data discovery by themselves.
There are a lot of positives ahead. Organisations have started to understand the importance of democratised analytics. The reduced total cost of ownership of analytics platforms today is helping the trend. Modern analytical platforms come with flexible and affordable usage-based pricing.
Data democracy will require meticulous change management program. It is essential so that users are enabled and motivated to use analytics.
Adoption will accelerate when analytics is embedded in web & mobile applications used by individual departments, bringing analytics closer to the individual points of action.
I am optimistic about analytics in the near future where we shall see organisations, across all spectrums, incorporating data as their way of life.
AIM: Understanding from a client’s standpoint, what are the biggest challenges with respect to adoption? What are the steps to address these?
CA: A major reason why analytics gets low user acceptance is when the developed solution is not aligned with user expectations. This may happen when all the stakeholders were not involved in early requirement gathering stages.
Analytics is a continuous journey. It will require a well-defined mechanism to capture feedback from users. Companies will require a program management team which will prioritise and incorporate the feedback in a structured manner. Without this, the requests can get lost in flood of demands. Without the user feedbacks getting incorporated, it will lead to the reduced motivation of the user with analytics.
A big impediment is the lack of user training. Without training, the user will not be comfortable with the platform & technology and analytics will be seen as an uphill task which increases workload with little productivity benefit. The role of training is to enable the user on the platform so that they are able to reduce their workload and devote more time to do strategic analysis of their data.
Using the right analytics platform as well as choosing the right partner is extremely important. Organizations must ensure that program output is aligned with corporate goals and the solution is scalable as well as easy to use.
AIM: Is user training important even with the modern analytics platforms which promise easy self-service capabilities for end users?
CA: Yes, even with self-service analytics platforms today, users need to be enabled to use the platform. Modern data architecture gives users the ability to play with their data for deriving analysis. But for governance & security, the user should not be allowed to do data massaging & pull elements at the data warehouse and database level.
With this apparent lack of control in modern analytics platforms, the user may resort to excel for data analysis. To avoid this, users must be trained on using the platform to maximum benefit, within a defined organisational governance framework.
If the user is asking for some specific analysis, which will cause changes to the underlying data, then the governance committee should ensure that this is not done in an ad-hoc manner, taking in random user requests.
AIM: When analytics is delivered at enterprises, do businesses have an understanding of what they actually need?
CA: When starting, organisations need a blueprint for the overall analytics journey. But the journey cannot be cast in stone. The organisation needs a framework to incorporate the suggestions. It will help tie the changes to organisational goals and corporate vision. The management must ensure that the analytics solution should not start replicating individual way of work. It must not become just a personal productivity tool. The organisation needs to do a thorough introspection about why things are done in a certain manner and whether things can be done in a better manner?
AIM: What factors differentiate a company’s effectiveness on analytics?
CA: There are some identifiable factors which set organisations apart in their success with analytics:
1. Analytics gets successful in organisations where senior management is closely involved. Here, the management gives clear directions to the detractors that the initiative will not be rolled back
2. Sunset of the existing MIS process is very important. Executives must roll-out a cut off period after which decisions and reviews will only be supported by the analytics platform
3. Successful organisations create a culture of analytics within the organisation where there is a reward & recognition mechanism for adopters of the technology.
4. Analytics fails at organisations where analytics is seen as a one-time exercise. It is a continuously evolving process.
AIM: Are data scientists and analysts detached from real business challenges and domain knowledge? How can companies overcome this?
CA: There is a skill gap globally on analytics. Analytics requires strong functional understanding as well as technical expertise. This can be solved with the right training. The organisation should go for an analytics partner who will partner during the entire journey from the stage of solution conceptualisation till the stage of adoption. I also recommend that sooner or later they should have an internal team in place for support & maintenance. This will ensure that the response time is quick.
How can organisations set-up an internal team? They should align an inside team to work with the analytics partner for 3-6 months. This should not be a part-time responsibility. In case the partner is not available for this, the organisation should conduct a series of training. The training program must be tied to measurable goals. Take a survey after and capture the feedback from participants about the program. The level of learning should be tied to individual rewards.
The company should assess whether they have been able to convert 80% of people after training into analytics promoters. Organisations should go for a training partner that ensures strong process convergence with training.
AIM: Are data scientists and analysts detached from real business challenges and domain knowledge? How can companies overcome this?
CA: With analytics, organisations need to keep business users closely involved with the program. Organisations need people who come from a business background and who have a fair knack of data. These people can then be trained to do a certain degree of data massaging, data manipulation and data analysis. For hardcore data scientists, they should have training on the business aspects. This will make sure that they become aware of the entire business process, pain areas and factors.
With such strong functional grounding, they will be able to do a meaningful analysis for the business. Data scientists need a varied set of skills today, from understanding their data, the tool and technologies so they can create analysis on the data and finally link it to the business outcome.
AIM: Does the pace of analytics adoption differ across industries?
CA: It differs from company to company and from industry to industry. There are organisations where people at every level are very comfortable with technology. In such organisations, there are usually integrated platforms and applications. Pushing analytics here is a lot easier because such organisations have a culture of being technology adopters on applications where it can be embedded.
With embedded analytics, it serves them analytics on their most used applications. This makes adoption & usage much easier and better. On the other hand, there are also industries where technology has not made significant inroads. In such organisations, the management is MIS -driven by MIS. In such organisations, the pace of analytics adoption is less as compared to other industries.