Research Methodology

Section 1: An Integrated Framework for Market Channel Evaluation

This section establishes the methodological architecture for the study, "Impact of
Agri-Exhibitions and Demonstration Plots on Product Awareness, AND Retail Channel
Analysis." It outlines the philosophical underpinnings, the structural design of
the research, and the specific objectives and questions that guide the inquiry. The
framework is explicitly designed to generate robust, actionable insights for
Sangamitra Seeds Corporation by integrating quantitative and qualitative evidence.
1.1 Research Philosophy: A Pragmatist Stance
This study adopts a pragmatist research philosophy. Pragmatism is fundamentally
concerned with practical application and problem-solving, prioritizing the research
question over allegiance to a single methodological doctrine. This philosophy is
exceptionally well-suited for this project, as the ultimate goal is not merely to
describe the state of Sangamitra's marketing efforts but to generate evidence-
based, actionable recommendations for improvement. A pragmatist approach provides
the necessary flexibility to combine diverse data types and analytical techniques—
quantitative surveys, qualitative interviews, and secondary data analysis—in a
manner that most effectively addresses the central business challenges of farmer
conversion and retail channel effectiveness.
1.2 Research Design: Explanatory Sequential Mixed-Methods
To operationalize the pragmatist philosophy, this study employs an explanatory
sequential mixed-methods design. This design is structured in two distinct,
sequential phases, where the findings of the first phase inform the exploration in
the second. This structure is deliberately chosen to create a comprehensive
understanding of both the scale and the underlying causes of the phenomena being
investigated.
* Phase 1: Quantitative Investigation. The initial phase focuses on the collection
and analysis of quantitative data derived from farmer surveys and retail audits.
The primary goal of this phase is to answer "what" and "how much" questions. For
instance, it will quantify the percentage increase in product awareness following
an Agri-exhibition, measure the correlation between event attendance and purchase
intent, and calculate compliance scores for branding and promotional materials at
the retail level. This phase provides a broad, numerical overview of the
performance of Sangamitra's marketing initiatives.
* Phase 2: Qualitative Explanation. The second phase involves the collection and
analysis of qualitative data, including open-ended survey responses, in-depth
interviews with retailers, and documented field observations. This phase is
designed to explain the "why" behind the quantitative results identified in Phase
1. For example, if Phase 1 reveals a weak correlation between event attendance and
sales, Phase 2 will explore the reasons, such as the identified weakness in post-
event follow-ups. Similarly, if quantitative audits show low visibility scores,
qualitative interviews will uncover the root causes, such as inconsistent delivery
of branding materials or a lack of retailer training.
By sequencing the research in this manner, the qualitative data provides depth,
context, and explanatory power to the initial quantitative findings, leading to a
more nuanced and holistic understanding of the marketing ecosystem.
1.3 Research Objectives and Key Questions
The research is guided by three overarching objectives, each broken down into
specific, measurable research questions. This structure ensures that every data
collection and analysis activity is directly linked to the project's core aims.
Table 1: Research Design Matrix
The following matrix provides a comprehensive overview of the research plan,
mapping each objective to its corresponding research questions, data sources,
collection methods, and analytical techniques. This serves as the strategic
blueprint for the entire study.
| Research Objective | Key Research Questions | Data Source(s) | Data Collection
Method | Analytical Technique(s) |
|---|---|---|---|---|
| 1. To assess the impact of Agri-exhibitions and demonstration plots on farmer
awareness and purchase decisions. | 1.1 What is the baseline level of product
awareness among target farmers? <br> 1.2 How significantly do events increase
product awareness, recall, and purchase intent? <br> 1.3 What is the correlation
between event attendance and subsequent purchase decisions? | Farmers (N=30) who
attended Sangamitra's events; Sangamitra's historical event attendance and sales
data. | Structured Surveys (face-to-face/telephonic); Archival Data Review. |
Descriptive Statistics; Paired t-tests (pre/post awareness); Correlation Analysis;
Multiple Regression (Predictive Modeling). |
| 2. To analyze the effectiveness of the retail channel in supporting marketing
efforts. | 2.1 How visible are Sangamitra's products at the retail level (Share of
Shelf, branding)? <br> 2.2 Is promotional material consistently and correctly
deployed? <br> 2.3 What are the key knowledge gaps and support needs among
retailers? | Agri-retailers (N=15) in Telangana; Field Observations. | Retail
Audits using a structured checklist; Semi-structured interviews with retailers. |
Audit Scoring System; Descriptive Statistics (compliance rates); Thematic Analysis
of interview notes; SWOT Analysis. |
| 3. To provide actionable, data-driven recommendations for improving marketing
ROI. | 3.1 What are the primary weaknesses and bottlenecks in the current
marketing-to-sales funnel? <br> 3.2 What strategic interventions (digital,
training, etc.) are likely to have the greatest impact on farmer conversion and
retail performance? | All primary and secondary data; Competitor analysis reports;
Expert opinions. | Synthesis of all collected data. | Integrated SWOT Analysis;
Qualitative Scenario Analysis (projecting outcomes of proposed improvements). |
Section 2: The Multi-Source Data Collection Strategy
This section details the systematic processes for gathering empirical evidence. The
strategy is predicated on the principle of triangulation, leveraging multiple data
sources—primary field data, internal company records, and external market data—to
construct a robust and multi-faceted view of Sangamitra's marketing performance.
2.1 Primary Data Corpus: Field-Level Intelligence
Primary data forms the core of this investigation, providing direct insights from
key stakeholders: farmers and retailers.
2.1.1 Farmer-Centric Data: The Structured Survey
The collection of data from farmers is central to evaluating the impact of
promotional events.
* Sampling Strategy: A purposive sampling technique will be employed to select a
sample of 30 farmers (N=30). The primary inclusion criterion is documented
attendance at a Sangamitra Seeds Agri-exhibition or demonstration plot within the
last 6-12 months. This non-probability sampling method is appropriate as it ensures
all respondents have direct experience with the marketing activities under
investigation. To enhance the sample's diversity and the external validity of the
findings, efforts will be made to include farmers from different geographical
regions and crop segments within Telangana.
* Instrumentation: The primary data collection instrument is a structured
questionnaire, which has already been developed and deployed for the initial 15
farmers. The questionnaire is designed to capture data across several key domains:
* Demographics: Crop type, landholding size, and farming experience.
* Pre-Event Metrics: Baseline product awareness and perceptions, measured
through unaided and aided recall questions.
* Post-Event Metrics: Changes in product awareness, brand perception, and
information recall following event attendance.
* Purchase Influence: Factors influencing the final seed purchase decision.
* Post-Event Engagement: Based on the preliminary finding that post-event
follow-up is a significant weakness , a dedicated section will be included to probe
this area more deeply. Questions such as, "Did a representative from Sangamitra
Seeds contact you after the event you attended?" and "What type of information or
support, if any, would have been most helpful to receive after the demonstration?"
will be systematically asked to gather both quantitative (yes/no) and qualitative
data on this critical gap.
* Operationalization: To facilitate quantitative analysis, key constructs will be
operationalized using standardized scales. For example, 'Product Awareness' will be
measured on a 5-point Likert scale (from 1 = 'Not at all aware' to 5 = 'Extremely
aware'). 'Purchase Influence' will be assessed by asking farmers to rank the
importance of various factors (e.g., demonstration plot results, retailer advice,
neighbor's experience, price) in their decision-making process.
2.1.2 Retail Channel Data: The Audit and Interview Protocol
Understanding the retail environment is crucial for analyzing the final link in the
marketing-to-sales chain.
* Sampling Strategy: A hybrid sampling approach, combining convenience and
purposive methods, will be used to select a total of 15 Agri-retailers across
Telangana. The purposive aspect will ensure the sample includes retailers from
districts with varying levels of market maturity (e.g., high-growth, established,
and new-entry markets) to capture a comprehensive view of channel performance.
* Instrumentation: Data collection at the retail level will utilize a two-pronged
approach to capture both quantitative metrics and qualitative context:
* Retail Audit Checklist: A structured checklist is being used to conduct
audits, systematically capturing quantifiable data on key performance indicators.
This includes product visibility (e.g., shelf space allocated to Sangamitra
products measured in linear feet), branding consistency (e.g., presence and
condition of posters, banners), and availability of promotional support materials
(e.g., leaflets).
* Semi-Structured Interview Guide: To complement the audit, a semi-structured
interview guide will be used to engage retailers in a qualitative discussion. This
approach is specifically designed to navigate the documented challenge of
retailers' reluctance to share sensitive sales data. Instead of directly requesting
sales figures, the interview will focus on gathering perceptions and experiences.
To diagnose the root cause of inconsistent branding—a key preliminary finding —the
guide will include targeted questions like: "Can you describe the process for
receiving new promotional materials from Sangamitra?" and "Compared to other seed
companies, what support do you find most helpful for driving sales?"
2.2 Secondary Data: Contextual and Performance Benchmarks
Secondary data provides the historical and competitive context necessary to
properly interpret the primary findings.
2.2.1 Internal Historical Data (Sangamitra Seeds Corp.)
To establish performance benchmarks and analyze trends over time, a formal request
will be made to the company for access to archival data from the preceding two to
three fiscal years. This historical perspective is critical for evaluating the
efficiency of marketing expenditures. The data request will include:
* Territory-level sales data for products featured in promotional events.
* Marketing budgets specifically allocated to Agri-exhibitions and demonstration
plots.
* Historical attendance records for these events.
* Records of promotional material distribution to the retail channel.
Analyzing this data allows for a more sophisticated evaluation than a simple,
static snapshot. For instance, a basic correlation between marketing spend and
sales in a single year can be misleading, as it fails to account for lags, market
saturation, or execution quality. A more robust approach involves examining the
historical trend of marketing efficiency. By combining data on event costs,
attendance (leads generated), and subsequent territory sales, it becomes possible
to estimate metrics like Cost Per Lead (CPL) and marketing Return on Investment
(ROI) over time. Tracking the trajectory of this ROI provides a far richer
understanding of whether the effectiveness of these marketing tactics is improving,
declining, or stagnating.
2.2.2 External Market and Competitor Data (Literature Review)
The ongoing collection of secondary data on competitor strategies will be
formalized into a structured literature review. This review will serve as a
benchmarking exercise, comparing Sangamitra's practices against industry norms and
key competitors. Sources will include industry intelligence reports, competitor
websites and press releases, academic articles on agribusiness marketing, and
relevant news media. A specific focus of this review will be to benchmark the use
of digital engagement tools. The preliminary finding that competing brands
effectively use digital invites (WhatsApp/SMS) for demonstration plots highlights a
potential competitive gap. The review will systematically document these practices
to inform the final recommendations.
2.3 Macroeconomic and Environmental Data Integration
Farmer purchasing decisions are not made in a vacuum; they are profoundly
influenced by external factors beyond a company's control. To avoid misattributing
sales outcomes solely to marketing activities, this study will integrate key
macroeconomic and environmental variables into the analysis. The collection of this
data is essential for building a more accurate and context-aware model of market
performance. Data points will include:
* Meteorological Data: District-wise rainfall data for the relevant cropping
seasons, obtained from sources like the India Meteorological Department (IMD).
* Government Policy Data: Information on significant changes to agricultural
policies in Telangana, such as Minimum Support Prices (MSPs) or subsidy schemes.
* Agronomic Data: Reports of major pest or disease outbreaks in the study regions
that could impact seed choice and yield expectations.
The inclusion of these variables enables a more advanced level of analysis. The
conventional approach is to use such factors as control variables to isolate the
impact of marketing. However, a more insightful approach is to explore the
interaction effects between these external factors and Sangamitra's marketing
efforts. For example, a year with an excellent monsoon may not just independently
boost sales; it might also amplify the effectiveness of a demonstration plot by
making farmers more receptive to investing in high-quality seeds. The analytical
models will be designed to test for such interactions (e.g., by including a term
like (Event Attendance) x (Rainfall Deviation from Normal)). This allows for the
development of a dynamic marketing strategy, where Sangamitra can allocate
resources more effectively based on seasonal forecasts and prevailing market
conditions.
Section 3: A Triangulated Data Analysis and Interpretation Protocol
This section outlines the systematic, multi-stage process for transforming the
collected raw data into coherent findings, interpretations, and strategic
recommendations. The protocol is layered, beginning with qualitative data reduction
to uncover underlying themes, followed by statistical analysis to validate
patterns, and culminating in advanced modeling to provide strategic foresight.
3.1 Qualitative Data Reduction and Thematic Analysis
All non-numeric data, including open-ended survey responses, field notes from
observations, and transcripts from retailer interviews, will be analyzed using
thematic analysis. This method, as planned in the project's initial stages ,
provides a rigorous process for identifying, analyzing, and reporting patterns
(themes) within the data. The process will follow three distinct steps:
* Open Coding: The initial step involves a thorough review of the qualitative
data. As concepts and ideas emerge, they will be assigned a code (a short,
descriptive label). For example, a retailer's comment, "The banners they sent last
year are faded," would be coded as "outdated promotional material."
* Axial Coding: In this step, the initial codes are reviewed and grouped into
higher-level conceptual categories. For instance, codes like "outdated promotional
material," "no leaflets available," and "inconsistent branding" would be grouped
under the broader theme of "Inadequate Retail Merchandising Support."
* Selective Coding: The final step involves identifying the core themes that are
most salient and central to the research questions. These core themes, such as
"Post-Event Communication Gaps" or "Retailer Training Deficiencies," will form the
foundation of the qualitative findings.
The output of this process will be a rich, narrative summary of the key challenges
and opportunities within Sangamitra's marketing channels. These empirically
grounded themes will then be used to populate the SWOT (Strengths, Weaknesses,
Opportunities, Threats) analysis, ensuring that every point in the framework is
substantiated by direct evidence from the field, such as the finding that "60% of
retailers lack product training".
3.2 Quantitative Analysis: Statistical Validation of Market Performance
Quantitative data from the farmer surveys and retail audits will be analyzed using
statistical software (Excel and/or SPSS) to identify patterns and test
relationships.
* Descriptive Statistics: This will be the first step in the quantitative
analysis, used to summarize and describe the main features of the dataset. It will
involve calculating measures of central tendency (mean, median) and dispersion
(standard deviation) for key variables. For example, the mean pre-event and post-
event awareness scores will be calculated, along with the frequency distribution of
retailer audit scores. These summary statistics will be presented using tables, bar
graphs, and pie charts to provide a clear and concise overview of the findings.
* Inferential Statistics: To move beyond description and make inferences about the
relationships between variables, several statistical tests will be employed:
* Paired-Samples T-Test: This test will be used to compare the mean awareness
scores of the same group of farmers before and after attending a marketing event.
This allows for a statistically robust assessment of whether the observed increase
in awareness is significant or simply due to chance.
* Correlation Analysis: Pearson's correlation coefficient will be calculated to
measure the strength and direction of the linear relationship between key
variables. For instance, this will be used to test the hypothesis that there is a
positive correlation between the number of demonstration plots a farmer attends and
their subsequent purchase intent score, directly addressing a core project
objective.
3.3 Advanced Modeling for Strategic Foresight
To elevate the analysis from a descriptive and diagnostic level to a predictive and
strategic one, advanced modeling techniques will be utilized.
3.3.1 Predictive Modeling: Multiple Regression Analysis
A multiple regression model will be developed to identify and quantify the key
drivers of farmer purchase decisions. This statistical technique allows for the
examination of the relationship between a single dependent variable and multiple
independent variables.
* Objective: To build a predictive model that explains the variation in farmers'
purchase intent, thereby providing Sangamitra with a quantitative tool to guide
resource allocation.
* Model Specification: The model will be specified as follows:
* Dependent Variable (Y): Farmer Purchase Intent Score, a continuous variable
measured on a 1-10 scale, collected via the survey.
* Independent Variables (X_i):
* Event_Attendance: Number of Sangamitra events attended in the last year.
* Awareness_Score_Post: The farmer's product awareness score after attending
an event.
* Retailer_Visibility_Score: The composite score from the audit of the
farmer's primary retailer.
* Follow_Up_Received: A binary variable (1 = Yes, 0 = No) indicating if the
farmer received any post-event follow-up.
* Farmer_Landholding_Size: A demographic control variable.
The resulting equation will take the form:
\text{Purchase Intent} = \beta_0 + \beta_1(\text{Event\_Attendance}) + \beta_2(\
text{Awareness\_Score\_Post}) + \dots + \epsilon

The primary output of interest will be the standardized beta coefficients ({\
beta_i}). These coefficients represent the relative impact of each independent
variable on purchase intent, holding all other variables constant. For example, the
model might reveal that the coefficient for Follow_Up_Received is larger than the
coefficient for Event_Attendance. Such a finding would provide strong, data-driven
evidence that investing in a systematic post-event follow-up program could yield a
higher return on investment than simply increasing the number of events hosted.
3.3.2 Strategic Foresight: Qualitative Scenario Analysis
To translate the analytical findings into a strategic decision-making context, a
qualitative scenario analysis will be conducted. This structured foresight method
moves beyond a simple list of recommendations to explore plausible future
trajectories for Sangamitra's marketing strategy. Based on the integrated findings
from the entire study, three distinct and compelling scenarios will be constructed.
* Scenario A: "Business as Usual." This scenario will extrapolate current trends
and practices. It will likely project stagnating market share, continued leakage of
leads generated from events due to weak follow-up, and increasing vulnerability to
more digitally agile competitors who are already leveraging tools like WhatsApp for
farmer engagement.
* Scenario B: "Digital-First Farmer Engagement." This scenario will model the
potential impact of fully implementing the proposed digital integration strategies,
such as geo-tagged demo plot invitations and digitized real-time feedback forms.
The narrative would project improved event attendance rates, enhanced data quality
and velocity, and the development of a stronger direct-to-farmer relationship.
However, it might also highlight that challenges in the last-mile retail channel
could still cap the overall sales impact.
* Scenario C: "Channel Partnership Excellence." This scenario will explore the
consequences of a focused, strategic investment in strengthening the retail
channel. This includes the deployment of standardized branding kits and the
implementation of monthly retailer training programs on product benefits and sales
techniques. This path would likely lead to superior in-store execution, higher
product visibility, and stronger retailer advocacy, ultimately improving sales
conversion at the point of purchase.
Presenting these scenarios provides Sangamitra's leadership with a clear and
tangible framework for strategic discussion. It shifts the conversation from "What
should we do?" to "What future do we want to create?" and clarifies the trade-offs
and potential outcomes associated with different strategic priorities.
Section 4: Ensuring the Credibility and Integrity of Findings
This final section outlines the measures that will be taken to ensure the quality,
rigor, and ethical conduct of the research. Upholding these standards is paramount
for the final report to be considered credible and for its recommendations to be
trusted.
4.1 Validity and Reliability
* Validity: To ensure that the research accurately measures what it intends to
measure, several strategies will be employed.
* Triangulation: The core of the validity strategy rests on triangulation, which
involves cross-verifying findings from the study's multiple data sources. For
example, a qualitative theme from retailer interviews suggesting "inconsistent
promotional support" will be validated against the quantitative data from the
retail audit checklists measuring the presence of those materials.
* Instrument Pre-testing: Prior to full deployment, the survey questionnaire and
retail audit checklist will be piloted with a small, representative sample (e.g.,
two farmers and one retailer). Feedback from this pilot will be used to refine the
instruments, ensuring that questions are clear, unambiguous, and capture the
intended information.
* Reliability: To ensure that the data collection procedures are consistent and
repeatable, all processes will be standardized. Any individuals involved in data
collection will be trained on a single, uniform protocol for administering surveys
and conducting retail audits. This standardization minimizes procedural variance
and enhances the reliability of the data collected.
4.2 Addressing Research Limitations
Every research project has limitations, and transparently acknowledging them is a
hallmark of rigorous inquiry. The primary limitations of this study include:
* Sampling Method: The use of purposive and convenience sampling means that the
samples of farmers and retailers are not statistically representative of the entire
population in Telangana. Therefore, the findings cannot be generalized with
statistical confidence beyond the sample group, though they will be highly relevant
to Sangamitra's target audience.
 * Data Access: As noted in the preliminary phase, obtaining sensitive commercial
data, particularly precise sales figures from retailers, is a significant
challenge. The study mitigates this by using proxy variables (such as retailer
perceptions of sales velocity, shelf space allocation, and farmer inquiries) to
infer performance, but the absence of direct sales data is a limitation.
* Response Bias: There is a potential for social desirability bias in survey
responses, where farmers or retailers may provide answers they believe are
favorable to the researcher. The assurance of confidentiality is the primary
strategy to mitigate this bias.
4.3 Ethical Considerations
All research activities will be conducted in strict adherence to ethical principles
to protect the rights and welfare of all participants.
* Informed Consent: Before any data collection begins, each participant, whether a
farmer or a retailer, will be provided with a clear and concise explanation of the
study's purpose, the nature of their involvement, how the data will be used, and
their right to withdraw at any time. Verbal informed consent will be obtained and
documented before proceeding with any survey or interview.
* Confidentiality and Anonymity: The privacy of all participants will be strictly
protected. All personal identifiers will be removed from the data during the
analysis phase. In the final report, all data will be presented in an aggregated
and anonymized format. Names of individuals and specific retail locations will be
replaced with codes (e.g., Farmer 1, Retailer A) to prevent identification.
* Permission for Photography: In line with the project plan, photographs of retail
displays will only be taken after obtaining explicit permission from the retailer
or store manager.