Leadership review
Yeshi Qelzani
Chief Human Resources Officer · India
Highlights
34 people colleagues
Four functional heads
₹5.4 crore FY 2026–27 expense; separate ₹52 lakh capital provision
30 years
Employee services to CHRO
Five continuous appointments from July 1996
42 service points
Lender operating context
780 company employees; ₹1,460 crore outstanding loans, not personal authority
A letter from Yeshi Qelzani
I ask who has to find the money, absorb the delay or carry the responsibility when a people policy looks inexpensive. The answer is sometimes a candidate, a field colleague or a specialist pushed into a management job they do not want. My work is to make that transfer visible and decide whether the company should fund a different arrangement. I have also had to correct policies I approved before asking that question carefully enough.
I grew up in Tawang in a Buddhist family and studied sociology at Vezmuthi Social Studies College, completing my degree in June 1996. I speak Monpa, English and Hindi, with working Kannada developed after moving to Mangaluru. The geographical moves are part of my story, but I would not turn them into a claim to understand every colleague's circumstances. What shaped my professional approach was the practical experience of seeing a small administrative delay become a substantial problem for someone whose household had little room to absorb it.
Yeshi Qelzani
Chief Human Resources Officer
Strategic priorities
01
02
03
04
Performance
Priority outcome 01
In 2022, eleven people from a group of 42 recent recruits supplied evidence that they had paid recruitment charges connected with their applications. The documented amounts totalled ₹2.2 lakh. Our arrangement with the recruitment supplier did not authorise charging candidates. The discovery therefore created two questions: what the company should do for the people who had paid, and how it should investigate the supplier relationship. They were connected, but they did not have to move at the same speed. Procurement preferred to establish recovery from the supplier before approving reimbursement. That would avoid paying twice if responsibility remained disputed, and I understood the concern about unsupported claims. I proposed checking the eleven records and reimbursing the evidenced amounts from company funds while the commercial review continued. The recruits had already carried the cost of an arrangement the employer had selected. Making them wait for a dispute they could not control would preserve our cash position by extending their burden. The CEO approved ₹2.2 lakh and the payments were made. Legal and procurement retained the supplier review, including questions about further evidence and the future contract. We paused new referrals through that supplier during the review. HR contacted the other 31 recruits through a confidential process that did not require a public allegation or assume that everyone had retained the same documents. We did not label those cases clear merely because they were not part of the initial evidenced group. The immediate outcome was eleven completed reimbursements and a continuing investigation, not a declaration that every concern had been resolved. No recovery from the supplier is counted in this case. The company accepted the cash cost and a slower recruitment channel while it examined the facts. I learned to keep remedy, evidence gathering and contractual responsibility visible as separate decisions. That made it possible to act on what we knew without claiming knowledge about the cases still open.
Priority outcome 02
By 2023, 112 field colleagues were routinely advancing travel money and reclaiming it through the monthly expense cycle. I had approved that cycle during my HR director years because it reduced processing interruptions. A review of 96 settled claims found a median 24 calendar days between submission and reimbursement. The number did not show every employee's cash position, but conversations with colleagues made clear that the company was relying on personal funds to carry ordinary operating work. Finance wanted to improve claim completeness before introducing advances. Incomplete claims did create reconciliation work, and an advance would not remove that problem. COO Lalita Zevnathi also worried that a new process might leave managers chasing unsettled balances. I proposed an ₹18 lakh revolving float with advances tied to planned field activity, defined reconciliation and a named operating owner. It would fund work before travel while retaining scrutiny of the actual expense. The CEO approved the float. We explained that it was recoverable working cash circulating through the process, not an additional ₹18 lakh annual travel cost or an employee allowance to keep. Managers reviewed unsettled items, and finance retained control of expense recognition and payment. HR helped colleagues understand the arrangement and recorded problems without treating every delay as evidence of misconduct. I acknowledged that my earlier approval had considered processing convenience more carefully than the person financing the journey. A later sample of 104 settled claims showed a median seven-day reimbursement period. The populations and operating conditions were not identical, so we reported the observation without attributing all improvement to the advance arrangement. Some claims still required clarification. The decision's immediate value was that approved travel no longer depended on the same expectation of personal funding. It also required working cash, supervision and continued review. I would not describe those obligations as eliminated merely because the headline delay became shorter.
Priority outcome 03
In 2024, a role review examined 78 relationship officers whose job descriptions combined customer development, application preparation and a limited delegated credit-decision responsibility. The credit leader wanted a clearer separation as the business and product mix changed. Some regional managers preferred the combined role because customers received a faster answer from someone who knew their business. Several employees had also joined with an expectation of broad responsibility and saw the proposed change as a reduction in professional standing. I worked with the credit leader and COO to distinguish the operating decision from the employment response. Qualified credit and compliance colleagues determined the appropriate authority model. HR could not decide that model by rewriting a title. We could, however, describe the new work honestly, assess the staffing requirement and make the transition intelligible to the people affected. I opposed presenting the change as a simple clarification that would cost nothing and leave every career expectation untouched. The company approved eight additional credit-analyst positions at an annualised employee cost of ₹92 lakh, outside the people-function expense budget. Relationship officers retained customer development and application responsibilities; the revised delegated decisions sat with the designated credit roles. We explained the revised remits individually and made relevant specialist opportunities available through the normal role process. No employee was described as unsuitable merely because the organisation had changed the work it wanted the position to perform. The new structure went into operation after the credit and operating owners completed their required preparations. It added cost and another handover rather than promising faster decisions in every case. This account does not claim a measured fall in credit losses or certify regulatory compliance from the staffing change. My contribution was to ensure that the authority decision had a credible employment design, including a funded team and an honest explanation to colleagues. A changed diagram would not have supplied either on its own.
Priority outcome 04
Our 2025 employee-listening exercise received 534 responses from 780 invited employees. The initial reporting proposal offered detailed breakdowns for each service point and role group. In some combinations fewer than five people had responded. Removing names did not remove the possibility that a manager could infer who had raised a concern from the role, location and wording together. The listening provider described the report as confidential, but its proposed detail still required examination. Regional managers wanted the smallest available breakdown so they could act locally. That was a reasonable operating preference, especially where an enterprise average might conceal a specific issue. I asked what decision each additional level would support and whether the same action could follow from a broader cluster or a facilitated discussion. We chose to suppress the small-group breakdowns, edit identifying details from shared comments and limit access to the underlying records. A ₹9 lakh external-analysis budget funded the revised work. The COO supported the approach but required a useful follow-up, not a general report that nobody could act on. We therefore retained enterprise and sufficiently broad cluster themes, assigned management responses and explained to employees how their feedback would be used. The 246 nonrespondents remained outside the respondent account. We did not describe the results as the opinion of all 780 people or treat participation as proof of confidence in the employer. The revised report gave managers less granular material than they had requested. That was an intentional limit, not a missing analytical feature to be restored informally. The case does not establish perfect anonymity or a subsequent increase in trust. It shows a decision to reduce unnecessary exposure while preserving a route to action. I wanted the company to judge the usefulness of listening by the changes it could responsibly consider, rather than by how closely a manager could locate the person behind a comment.
Priority outcome 05
In 2025, five experienced credit professionals were eligible for progression, but the available next grade was tied to supervising staff. The credit leader needed their judgement on difficult cases and their support for less experienced colleagues. Only some wanted the employment responsibilities of a line manager. I had previously accepted the grade structure as a convenient way to compare roles. The discussion showed that it could make people choose between advancement and the kind of contribution they were best prepared to make. Finance questioned whether a specialist route would become a pay increase without a demonstrable change in responsibility. I agreed that a new title alone would not justify it. Working with the credit leader, we defined two specialist bands around the complexity of assigned work, professional review and mentoring expectations. Delegated lending authority still followed the credit policy and individual appointment; a higher career band did not automatically confer a larger approval limit. The CEO approved an annualised ₹28 lakh pay increase for the three colleagues who chose the new route after assessment. Two preferred to remain on their existing path, with management opportunities considered separately. The cost belonged to enterprise employee expense, not the CHRO's function budget. We explained the responsibilities and the review process before asking anyone to choose, and did not describe the specialist option as the superior route for every experienced employee. The immediate result was three appointments and an additional career choice. We did not count all five as retained because of the programme or claim a reduction in credit losses. The company accepted a continuing pay cost and a more demanding explanation of work value than a simple count of direct reports. My lesson was that a structure can be administratively clear while narrowing the careers the business actually needs. The alternative had to be specific enough to evaluate, rather than a flattering label that postponed the same argument.
Stakeholder voices
Yeshi separated the evidence needed to reimburse recruits from the unresolved supplier dispute. She asked the company to fund an immediate remedy while keeping the wider questions open, rather than treating the first payment as the end of the investigation.
She acknowledged her part in the old travel cycle and helped design a funded alternative with clear reconciliation. The conversation became more useful when we could discuss both the employee burden and the operating work that still had to be done.
Outlook
A reimbursement process may look inexpensive because the person doing the work supplies the working cash. I had to reconsider a cycle I had approved.
Read the essayMore granular feedback is not always more useful. A manager needs enough evidence to act, not every clue that might lead back to a respondent.
Read the essayCareer progression can recognise deeper professional contribution without making supervision compulsory. The alternative needs a real remit, a price and an honest choice.
Read the essayAn invitation
I am preparing for one first outside board contribution on people systems and organisational responsibility. I hold no statutory seat, and my fictional databank registration does not mean the proficiency assessment is complete. Twenty-four days a year is a provisional allowance to test with my employer and a prospective chair.
office@yeshiqelzani.example