Indrani Poddar

The Net Promoter Score is an index ranging from -100 to 100. It reflects the tendency or likelihood of customers to recommend an organization’s products or services to others. It is used to measure the customer's overall satisfaction of the product or service and brand loyalty. ‘Word of mouth’ being an integral part for any buyer community NPS is a critical scoring that every organization has embraced to make their brand larger and memorable for customers. All of this is calculated with ordinal data as the scoring is between 1-10 however this data is later modified to three discrete categories (categorized as promoter, neutral or detractor). NPS survey score of 9 or 10 are categorized as ‘Promoters’ which means they are very satisfied with the product or service and are very likely to recommend the product or services to other organisations. NPS of 7 or 8 are categorized as ‘ Neutral’ or ‘Passives’. They are not too satisfied with the product or services and will not be sharing any negative opinion though but are less likely to recommend to other organisations. NPS score less than 7 are termed as ‘Detractors’ which means these customers neither will repurchase or avail any further services of the company and could most likely negatively impact the reputation of the company. This is because categorization of the type of customers gives us a qualitative view of the customer experience and what kind of measures one must take to ensure maximum protection to the company’s reputation, higher growth and market share and better customer experience. Hence the judgement to promote a brand or not cannot just depend on the ordinal data which are independent variables. There has to be an accumulation of the customer experiences which comes through the categorical data only.

The whole purpose of a service or manufacturing of a product starts and ends with customer. Hence customer is the integral part of every product or service. Organisations have become far more flexible in today's time to meet customer goals and objectives. Driving value beyond expectation is something that the leaders in the market are thriving for to maximize growth and hence their top line and market share. Very rarely do we see a service provider or a manufacturer reject a customer. However citing key examples and criteria which is actually leading to a customer rejection in the value chain and offerings. 1) Credibility of the customer in the market, their credit rating and financial position determines whether one would deal with or reject the customer 2) Every organisation has a vision and a mission. If there is alignment to the vision with that of the customer a partnership can start and sustain. However any de-alignment of the same which leads to deviation of the vision and values of the organisation can lead to rejection of the customer 3) Confidentiality is very key between a provider and the customer. If there is any deviation to the confidentiality terms and conditions would definitely end up with customer rejection and legal suits 4) Customers supporting directly/ indirectly illegal organisations or customer themselves are involved in illegal activities or transactions, unethical activities or behaviour is another critical factor for customer rejection 5)Delivering value or service to a customer if the profitability falls below the company's minimum threshold can lead to customer rejection 6) Specification of a product for a manufacturing company undergoing too many changes w.r.t LSL and MSL very frequently could lead to customer rejection

Pareto analysis is a statistical technique used in decision making mostly where there is competition amongst the root causes, contributors to the cause and we need to identify the number of such causes or contributors, which has a significant overall impact. ‘‘In the late 1940s Romanian-born American engineer and management consultant, Joseph M. Juran suggested the principle and named it after Italian economist Vilfredo Pareto, who observed that 80% of income in Italy went to 20% of the population. Pareto later carried out surveys in some other countries and found to his surprise that a similar distribution applied.’’ – as mentioned in the website By Duncan Haughey Hence, we use the Pareto Principle (also known as the 80/20 rule) which means focusing on 20% of the work/cause/contributor; one can impact 80% of the overall job/problem/benefit. Hence Pareto Diagram is one of the essential tools used in total quality control and Six Sigma. While performing Pareto Analysis there could be some common misuses as follows – a) All problems/causes/ factors may not be included before doing the analysis. For example; if we are doing a Pareto of delay in booking of goods receipt. If I consider only few plants for the analysis the Pareto Diagram will only generate basis the data hence decision making would be inaccurate in case some of the major plants contributing to this problem has been left out of the analysis. b) Inaccurate, insufficient, or inconsistent data collection. For example, if the causal factors are not measured accurately and consistently for the same period and the 80/20 principle will also give inaccurate results. c) Only those probable root causes are included in the analysis that we assume might be the contributors and not the actuals. This means if we have not identified all root causes, which is actually affecting the problem it could lead to misleading output hence the problem would not get resolved or would be resolved for a temporary period. d) If the data points have interdependence, the outcome of the Pareto Analysis will not help. For example both data points must be addressed if they are interlinked even if one of them do not fall under the 80% contributing bucket. e) Some of the lower contributing factors could be quick wins however they get ignored again because these do not fall under the 80% contributing bucket. f) If weightage is added to the attributes considered for Pareto Analysis the outcome of such would be biased and misleading.

Severity is nothing but the impact or seriousness of a failure. When it comes to PFMEA, Severity assessment or ranking is of high importance as it is one of the factors affecting RPN score. Some of the common challenges we face while doing Severity assessment could be - a) Severity ranking is provided basis immediate or short-term view without considering long-term view. For instance, there are balances being accumulated month on month in suspense accounts. If one provides a low Severity rating keeping in mind a short-term perspective as there is no impact on overall financial reporting. However, as a long term impact such balances retained would be aged. In an Insurance company if the suspense account is for client funds, retaining aged balances will have critical implication due to Escheatment laws and other regulatory requirements hence severity rating should be high. Thereby, long-term impact should definitely be considered. b) Severity ranking is often determined from the immediate perspective instead of end user/customer perspective. For example for invoice processing the processing team is only interested in meeting SLA target hence in such cases invoice processing or first touch would have a higher severity rating however invoices pending in exception queue would have a lower severity rating. However, in reality, high exception invoices leads to lower paid on time, hence vendor dissatisfaction and escalation and often a threat on production. Hence considering end user/ customer view for severity assessment is critical. c) Severity rating is determined based on a limited group of individuals’ perspective and whims of the participants. If participants creating the FMEA are experienced, knowledgeable and determines the severity rating basis the organization’s nature of business, objectives, priorities then such FMEAs would have a more accurate severity assessment. d) Upstream activities has severe impact on downstream activities. Hence, severity assessment would grossly go wrong if processes are considered in silos and not with an end-to-end view. For example; the accounts payable team has to close certain aged Open POs within a certain time frame as per the policy of the company. Only considering this activity in silo might determine a low severity rating. However when seen from an end-to-end perspective all open POs determine the accrual to be booked by the Reconciliation team. If the accrual entry is incorrectly done then the financial reporting is incorrect which has high implications hence both upstream and downstream processes are to be considered. e) Severity ranking cannot be established if the failure effect is not well defined. Often, all scenarios of effect of failure mode is not captured leading to one severity ranking for the failure mode. Below is an example, which shows how severity ranking can change basis effect of failure for the same process step and failure mode. Process Step Failure Mode Failure Effect Severity Ranking Duplicate account is created for same supplier More than one account is created for same supplier when the team receives creation request from different requestors 1. Duplicate account leads to duplicate payment. 8 2. Decreases auto match process since invoice might get processed in X account and GR might get booked in Y account 4 Complexities and disagreements arise when the severity ranking is based on an improperly defined effect or inadequate severity scale. Hence, failure effect must be structured, explained and defined well which determines a more accurate severity scale. Severity rating must be determined basis effect of failure mode and not the failure mode by itself. The above examples clearly explains why we end up with inaccurate severity rankings and how one can overcome the same while doing a process FMEA.

An algorithm is a process or set of rules to be followed in calculations or step by step problem-solving methodology followed. In our organisation we have built algorithms using Random Forest model on R, wherein the quality check process has significantly improved efficiency and accuracy of the process. Typically in a quality check process either samples are randomly picked without any logic or rules being or rules defined are typically; all employees with less than 6 months experience 100% audit to be done, critical process 100% audit, 100% audit of all transactions above a certain threshold value, 50% audit between a given value and another so on and so forth. This conventional audit process uses maximum effort and cannot ensure high coverage of errors. What we have done is basis past trend of errors for 1.5 years, identified critical variables and combination of errors and developed an algorithm that gives the list of transactions for the audit team to audit which has highest potential of error occurrence. The model developed provides the audit team the transactions to be audited from the total number of transactions processed. Thereby optimizes effort spent and has lead accuracy rate closer to 100% using this predictive analytics. This algorithm also has self learning capability with any new type of error occurrence. Another example where we have effectively used algorithm as predictive analytics again used Random Forest Model was to predict the number of emails that can be responded in first attempt within a shorter turn around time. In a query ticketing process we get queries from customers in a ticketing tool inquiring about invoice payment status, payment not received, deduction amount, disputed payments etc. The inflow of such emails is about 250 per day with an SLA of 3 days to be responded or resolved. The current process was facing extreme challenges with backlog building up on the queries to be handled leading to miss of SLA and high vendor and customer dissatisfaction. Also simpler queries were pending for a longer turn around time as the team was handling the queries one by one basis ageing. We developed an algorithm using past history of emails / queries closed to predict which queries can be resolved in first attempt and which queries would need more research and hence passed on to the resolution team. This has helped improve current TAT of emails being resolved, higher First Pass Yield, reduction of backlog of emails without adding additional resource and team is having more qualitative time for research of complex queries or doing effective follow up with dependencies on customer organisation.