You picked a company-based HR project — attrition, employee engagement, training effectiveness — and the HR manager who agreed to host you three months ago has gone quiet on the actual data. Appraisal scores, exit-interview data, attrition numbers by department: all “confidential,” all off the table two weeks before your submission deadline. This is one of the most common mid-project crises in an Indian MBA HR specialisation, and it has a specific, workable fix that does not require restarting your project from zero.
The problem: real HR data is more sensitive than most students expect
An MBA proposal that says “I will analyse the company’s attrition and appraisal data” sounds reasonable in September and turns out to be legally and practically difficult for the company to grant in November. HR records touch individual employee identity, appraisal ratings and sometimes salary — data an HR department is genuinely restricted from handing to a student researcher outside a formal, often lengthy, data-sharing agreement, regardless of how cooperative your original contact was. This is not usually about the company distrusting you; it is about real internal policy that a mid-level HR manager cannot override on their own authority.
Fix it now: four ways to complete a company-based HR project without raw HR records
- Switch from company records to primary survey data. Instead of analysing the company’s own attrition figures, design and administer your own structured questionnaire to a sample of employees (with the company’s permission to distribute it) on engagement, job satisfaction or perceived reasons for turnover. This is primary data you collected yourself, not secondary company records, and most Indian B-school HR project guidelines accept this as a legitimate, sometimes stronger, methodological choice
- Ask for aggregated, anonymised figures instead of raw records. A company that cannot share individual appraisal scores can often share an aggregated departmental attrition rate, an average engagement-survey score, or a training-completion percentage — numbers with no individual identifiable data attached. Ask specifically for the aggregate rather than assuming the whole request was refused
- Use the company’s own published or internal-communication materials as secondary sources. Annual reports, CSR reports, internal newsletters and public statements often contain HR-adjacent figures (headcount, diversity ratios, training investment) that are already cleared for external use and can support a project’s context or discussion chapter even where operational HR data is unavailable
- Narrow your scope to what the company can actually grant. A project originally framed as “attrition analysis at Company X” can often be reframed as “employee perception of retention factors at Company X” — a scope shift that keeps the company and the site but replaces a data source you cannot get with one you can, and this reframing is usually acceptable to a guide with two or three weeks’ notice, far less acceptable with two days’
Why this happens so often in Indian MBA HR projects
Most company-based HR project agreements are made at the proposal stage with whichever contact was easiest to reach — often a junior HR executive, a friend’s parent, or an alumnus willing to help — someone who can genuinely offer access to the building and to informal conversations, but who frequently lacks the authority to approve release of records that touch individual employee data. The gap surfaces later specifically because the initial “yes” was real and well-intentioned, but was never actually a data-sharing approval from whoever legally controls that decision. Knowing this pattern in advance is useful for your next project even if it does not undo the current one: at the proposal stage, ask directly whether your contact has authority to approve data sharing, or whether that approval sits with someone else you should also loop in early.
What to say to the company when you ask again
A second ask framed narrowly and formally succeeds far more often than the first informal one. A workable approach: put the request in writing (email, not a hallway conversation), address it to whoever actually has signing authority over data-sharing rather than only your original contact, specify exactly what you need (a named metric, an aggregation level, a time period) rather than a broad request for “HR data,” and offer a signed confidentiality undertaking or non-disclosure agreement from your institution if one exists. Many Indian B-schools have a standard student NDA template for exactly this situation — check with your placement or academic office before assuming none exists.
What does a defensible methodology look like once you pivot?
Whichever fix you land on, an examiner reads a pivoted methodology the same way they read any other: does the method match the objective, and is the limitation of the substitute data source acknowledged rather than hidden?
- If you moved to primary survey data, state your sample size, sampling method (a company employee list is usually your sampling frame) and instrument clearly, and cite whether it is a validated engagement/satisfaction scale or one you built for this study
- If you moved to aggregated figures, state explicitly that individual-level data was unavailable and aggregated figures were used instead, and note in your limitations section what this means for the granularity of your analysis — you cannot run individual-level statistical tests on data you only received as a department-level average
- If you narrowed your scope, restate your objectives to match the narrower question precisely, so there is no gap between what your introduction promises and what your methodology can actually deliver
An examiner who sees the pivot acknowledged honestly in your methodology and limitations sections reads it as competent project management under real-world constraints — exactly the skill an MBA is meant to demonstrate — rather than as a shortcoming to be hidden.
The cost of waiting to fix this
Every week you wait on an unresolved data-access problem is a week your literature review, methodology chapter and front matter could have been progressing around whatever data source you eventually land on. The scholars who recover fastest from a mid-project data refusal are the ones who pivot their methodology within days, not the ones who keep re-asking the same contact for the same refused data while the calendar runs out. If your project is genuinely stuck, tell your guide this week, not in the final fortnight — a guide who knows early has far more room to approve a scope change than one who hears about it three days before submission.
How Tesify helps once your data plan is settled
Once you have decided which of the four fixes above fits your situation, Tesify’s AI thesis assistant can help you rewrite your methodology chapter to match the new data source — turning a primary-survey pivot into a properly structured Chapter 3 with a defensible sample, instrument and analysis plan, rather than a patched-together explanation of what changed and why. It can also help you draft the written data-access request itself, in the specific, formally scoped format that gets a second ask answered rather than ignored. Start rebuilding your methodology chapter now — see how quickly a scope change turns into a clean, defensible Chapter 3 on Tesify’s free tier.
If your revised project now leans on a primary questionnaire, how to build a conceptual framework for your MBA dissertation covers turning your revised objectives into a testable framework, and how to structure an MBA project report covers the chapter-by-chapter format your revised project still needs to follow. If your revised methodology needs a sample-size justification for your new survey, how to calculate sample size for a thesis covers the same G*Power and Krejcie-Morgan methods that apply here, and thesis vs dissertation vs project report in India settles the vocabulary if your revised scope also changes how your project should be labelled. Ready to fix your Chapter 3 tonight? Try Tesify’s AI thesis assistant on your own draft — no card required for the free tier.
Frequently asked questions
Can I still call my project “company-based” if I only used a survey I designed myself?
Yes — a primary survey administered to a company’s employees, with the company’s permission, still constitutes company-based, primary research; what changed is the data source, not whether the project is grounded in a real organisation.
Will my guide accept a scope change this late in the semester?
Most guides prefer an honest, timely pivot to a stalled project that misses its deadline — raise it as soon as you know the original data is unavailable, with a specific alternative already proposed, rather than waiting until you have exhausted every attempt to get the original data.
What if the company refuses to let me survey employees too?
Ask whether a shorter, fully anonymous survey distributed digitally (rather than in person, which can feel more intrusive to a company) changes their answer, and have a backup company or a published-secondary-data pivot ready if the answer is still no — do not let a single company’s refusal be the reason your entire project stalls.
Is it acceptable to use industry-wide HR statistics instead of company-specific data?
Yes, for context and comparison, but check your department’s expectations — most company-based project formats still expect some primary or company-specific element (even a small survey or a set of interviews) rather than a purely industry-level secondary-data analysis with no direct company engagement.
How much does Tesify cost, and is it worth it for a rewrite this late in my project?
Tesify has a free tier so you can try restructuring a chapter before deciding whether a paid plan is worth it for the rest of your project — check the current pricing page for exact rupee figures, since plans can change.
Is it academic misconduct to change my methodology after my proposal was already approved?
No, provided you inform your guide and get the change approved through whatever process your department uses for methodology amendments — a documented, guide-approved pivot is standard practice when field conditions change, not a violation of academic integrity.
Should I approach a different department in the same company instead of HR?
Sometimes — if HR specifically cannot share operational data, a related function (learning and development, or an internal communications team running an engagement initiative) may hold adjacent, less restricted data that still supports an HR-specialisation project, though check with your guide whether this still fits your specialisation’s expected scope.
What if I cannot get any company to agree to even a survey?
A published-secondary-data project — built entirely from industry reports, published HR statistics and academic literature rather than any single company’s primary data — is a legitimate fallback in most Indian MBA HR programmes, though check your specific department’s requirements, since some formats mandate a company-engagement element that a purely secondary study would not satisfy.