When working with data, people think that visuals like charts and graphs speak for themselves. Well, they don’t. People still have to communicate what they mean and represent. And those people have priorities, blind spots, and power dynamics to navigate to explain the data to their audiences. That’s where stakeholder analysis comes in.
As I work through weekly data projects using public datasets from sources like KNBS and government portals, I’ve realized something: A solid insight is useless if it doesn’t get to the right person at the right time — in a way they understand and apply.
That’s why I’m now baking the stakeholder
analysis process into every
brief I build.
This article is both a guide for
myself and a documentation of how I’m applying stakeholder thinking to make
public data relevant, usable, and acted upon — especially in the Kenyan
context.
Why Stakeholder Analysis Matters in Data Projects
If you build data products
(dashboards, briefs, analyses) without knowing who they’re meant to help,
you’ll likely end up with something technically polished but practically useless.
Stakeholder analysis gives you a way
to:
- Identify who’s affected by or influential in the issue your data project touches or addresses.
- Understand their interests, expectations, and fears in their work.
- Map out their power and influence.
- Strategize how to engage or involve them.
This matters because every data
project is, at its core, a change
initiative. Even if subtle, it’s asking someone to understand something
differently, act differently, or allocate differently. That can’t happen
without having stakeholder clarity.
The Goal: From Raw Data to Relevant Briefs
My goal isn’t to just “analyze
unemployment trends” or “visualize housing uptake.”
It’s to produce data briefs
that:
- Are aligned with someone’s real-world decisions
- Speak the language of their roles within an organization
- Account for who might resist or support the
insight
This is where stakeholder analysis
shifts from abstract theory to practical application in your data work.
What Is Stakeholder Analysis (in Data Terms)?
Stakeholder analysis is the process of identifying, understanding, and engaging the people or groups affected by your data project — turning it from a shiny project into something genuinely useful in decision making or understanding concepts about any topic. Here’s how I now understand it,
adapted to the data projects I run weekly:
🔹 Step 1: Identify and Categorize Stakeholders
This involves:
- Brainstorming all individuals, groups, or orgs affected
by or influencing the topic in the data.
- Looking at government structures — ministries, county
departments, SACCOs, public agencies.
- Considering external actors — NGOs, funders, advocacy
groups, even citizens or media players.
- Categorizing them based on their role, interest, or
influence.
📌 Example: If I’m analyzing youth unemployment data from
KNBS, key stakeholders might include Ministry of Labour officers, county youth
directors, FKE (Federation of Kenya Employers), vocational training centers,
and private recruiters.
🔹 Step 2: Collect Info on Each Stakeholder
You need to know:
- What do they care about (interests)?
- How engaged are they now, and how engaged should they
be?
- What do they already know or believe about the issue?
- How might the data project impact or threaten them?
Methods to achieve this are: using AI
to brainstorm the stakeholders, do LinkedIn research, look into old project
reports, research news archives, and ideally — have real conversations or email
exchanges with relevant stakeholders.
🔹 Step 3: Map Influence and Impact
This helps me know who to
prioritize when creating or sharing a data brief. To be honest here, I’m
still learning to figure this method out. Some mapping tools I use:
- Salience Model:
Based on Power, Legitimacy, Urgency
- Influence/Impact Matrix: Who can influence vs. who is affected
- Power/Interest Grid:
Who cares, and who has sway?
Each tool will show you how to allocate
energy, frame insights, and decide whose input matters at
what level of the data project.
How I Apply stakeholder analysis in My Weekly Data Briefs
Here’s what it looks like in action:
🎯 Example 1: Unemployment Data
- Source: KNBS Quarterly Labour Report
- Signal/indicator : Youth unemployment in urban counties
remains high
- Stakeholders: Ministry of Labour (policy), County Youth
Offices (programs), recruiters (jobs), NGOs (training support)
- Result: I draft a 1-page data brief tailored to county-level
employment officers, showing how this insight can guide program or
budget adjustments.
🏠 Example 2: SACCO Housing Uptake
- Source: KNBS Real Estate Report
- Signal/indicator : Drop in uptake of SACCO housing
loans in 2023
- Stakeholders: SACCO board members, Ministry of Housing,
financial inclusion actors (FSD Kenya)
- Result: I shape the brief to raise a simple question: What
factors are driving low uptake — and what interventions could change the
trend?
Here the point is: each project
isn’t just about data analysis. It’s about alignment of the data project
with what stakeholders want. Stakeholder analysis therefore helps me position
the insight where it can actually be heard and applied in decision-making.
Why This Especially Matters in Kenya & Africa
We have data, reports but what we
often lack is translation of these data into the decision space. Stakeholder
analysis is my way of:
- Getting public data out of the ivory towers and vaults to
where decisions are made.
- Giving it a pulse in boardrooms, ministries, SACCO
AGMs, or NGO planning meetings
- Creating traction where it counts when these insights
reach the relevant stakeholders.
In our context, influence isn’t
always written down as part of job descriptions and power is social.
Stakeholder mapping therefore lets one account for that reality.
Conclusion: This Is How I’ll Build Going Forward
I don’t want to build “impressive”
data projects.
I want to build useful ones —
the kind that shape conversations, trigger programs, or help someone in a room
make a clearer decision.
That’s why stakeholder analysis
isn’t just a management buzzword to me. It’s now a built-in part of my weekly
process. And as I refine this method, I’ll keep sharing what works — and what
doesn’t.