Relational strategies for working with parliaments in Africa

Unlike ministries or research institutions, Parliaments are political institutions first, whose priorities shift with electoral cycles. Their top leadership turns over frequently, and the structure is hierarchical and charged. Demand for evidence is episodic and member-driven. This doesn’t mean evidence doesn’t belong in Parliament. 

Legislative and oversight roles shape public outcomes at scale, and citizens’ expectations of accountable governance rest, in part, on decisions being grounded in credible information. Evidence is pivotal to Parliamentary processes, but to be useful, it needs to be demand-driven, objective, timely, and patiently embedded in the rhythms of political life.

We worked with the Parliaments of Kenya and Zimbabwe for two years from November 2023 to October 2025, and the National Assembly of Zambia (NAZ) from November 2025 to date, with support from the Hewlett Foundation. As is characteristic of our Learning Partnerships (LPs), we were responding to our clients’ needs to establish an evidence-driven decision-making culture.

• In Kenya and Zimbabwe, we trained 112 Parliamentary staff, secured formal partnership agreements, and presented IDinsight’s work on AI in legislative processes to 15 African Parliaments at the 2025 Association of Parliamentary Libraries of Eastern and Southern Africa (APLESA) conference.
• In Zambia, we launched the five-year MoU with a Monitoring, Evaluation and Learning (MEL) advisory workstream, deliberately structured around co-creation to embed capacity building at the heart of the engagement.

We discovered that the true measure of success in improved evidence use lay in the relational strategies employed and the ability to navigate the political landscape inherent to Parliamentary institutions. In this blog, we share five key lessons that highlight the principles necessary for genuine partnership and sustained uptake of evidence.

1. Neutrality is a competitive advantage

When working in a politically oriented environment such as a Parliament, it is essential to remain neutral and avoid the politics of your own client ecosystem. Everyone who walks into a Parliament is read for their political motive. A development partner, Civil Society Organization (CSO), or government entity often has an agenda. What Parliaments rarely encounter, and genuinely value, is a partner who shows up without one.

In Kenya, the Parliamentary Research Service asked IDinsight for robust evidence to inform the drafting of the 2022 Startup Bill and evaluation of the National Policy on Women’s Economic Empowerment (NPWEE). We supported them in laying out the evidence on multiple sides by sharing IDinsight’s process evaluation of the Women Enterprise Fund (WEF). This provided insights into the barriers women micro-entrepreneurs face in accessing finance and into the effectiveness of delivery mechanisms to inform deliberations on the Startup Bill. We also conducted a rapid comparative analysis of women’s economic empowerment policies in similar countries to enable the Social Policy Committee to deliberate on NPWEE with relevant evidence.

In Zimbabwe, partnerships with Parliament are built carefully and over time. Our Five Year Rolling  MoU was developed over months of working together, showing that we were there to serve their evidence needs and strategic priorities, not our own.

In Kenya and Zimbabwe, our work spanned two distinct institutions, each with multiple departments and reporting structures. If something came up about a stakeholder, we listened without judgment. We offered no opinion, even when a joke landed at the other party’s expense. Instead, we adapted our working style to ensure each party felt like a genuine partner. The moment you signal, even lightly, that you have taken a side, your neutrality is compromised. And in government partnerships, once that doubt exists, it’s difficult to undo.

2. Demand pull beats supply push almost every time

Externally defined priorities rarely map onto institutional reality. A brief that answers a question no one is wrestling with just sits on a shelf. In Kenya and Zimbabwe, the Parliamentary Research Services approached us with specific capacity requests, so we tailored our scope of work to meet those needs. When clerks saw what their research officers actually needed and asked us to adjust the training focus, we did so rather than holding rigidly to the initial scope. When AI became a live interest among Members of Parliament (MPs) and Parliamentary researchers, we also leaned in with a presentation to 15 Parliaments at APLESA, and sessions with Kenyan clerks on bringing technology into their work. In Zambia, NAZ requested that we broaden our capacity-building support to include strategic planning and asked us to align the development of the Theory of Change (ToC) and Monitoring, Evaluation and Learning (MEL) Framework with their 2027–2031 strategic plan to ensure relevance.

Overall, all Parliaments asked us to extend the training to staff groups that were not in the original scope. It was clear to us that they were pulling the work forward instead of us pushing it.  Getting there requires proximity, responsiveness, and a willingness to start with whatever the institution actually needs right now. As the scope changes, we can still bring our own expertise to bear on where it should go next.

3. Being locally grounded matters

Parliaments are not only evaluating our technical credentials. They are also often quietly assessing whether we understand the local context. For instance, when a Kenyan researcher describes the pressures of working within a particular governance structure, a team member who has lived in similar environments does not just follow the conversation; they understand what is not being said. The client can also sense when the person across the table has a personal stake in the outcome, not just professional interest. 

Our team was deliberately locally grounded, with staff from Kenya, Malawi, and Zambia who brought a genuine allegiance to the well-being of the countries they serve. Contextual familiarity and personal stake in a region’s future are markers of expertise in the Parliament’s eye. Being locally grounded helps in two ways: understanding the leadership and level of influence, and showing respect the way the culture expects:

• Knowing who decides and who drives, then keeping both in the loop: Parliamentary institutions have two types of actors. Drivers carry the day-to-day momentum. Deciders determine whether the engagement is institutionally sanctioned and ensure that resources are allocated. The risk is building a productive relationship with drivers and assuming progress will cascade upward on its own. A senior official (decider) who learns about new partnership plans from a colleague (driver) may feel sidelined and can quietly withdraw the institutional cover that makes everything else possible, without ever saying so explicitly. 

  For instance, in Kenya, our first formal meeting was with the Director General of Parliament.  Only after he was satisfied did he direct his team to begin the paperwork. Keep deciders informed before activities, seek their endorsement, and maintain both relationships simultaneously, not sequentially.

• Reading the room on titles: How you address people carries weight, and this is deeply cultural. In the Parliamentary contexts we worked in, addressing senior officials by title and ensuring equivalent seniority on our side during meetings were both markers of respect that built institutional credibility over time. Where hierarchy is pronounced, getting it right consistently matters. Some officials will, over time, invite you to use their first name. Wait for that cue; do not assume it. Importantly, your team needs to reasonably align with the client’s norms. A colleague you would address by first name internally may warrant their full title in a client environment. The goal is to ensure people feel the respect they are due, on their own terms.

4. Political turbulence can be disruptive; design for it accordingly.

In Kenya, protests against Parliament disrupted our training schedule. In Zimbabwe, a government-wide due diligence process stalled our MoU for months. In Zambia, the demanding work schedule for staff towards the dissolution of Parliament, and the need for selected staff  to facilitate the establishment of new constituencies delayed our MEL workshop. In all countries, Parliamentary sittings, budget debates, and shifting priorities reshuffled our timelines more than once. None of this was unusual – Parliamentary calendars can be volatile by nature. Rather than bracing for disruption, we recommend designing the work so disruption doesn’t derail it. 

First, build flexible timelines and budget buffers from the start. Treat contingency windows as a structural assumption, not a fallback. If your project design assumes a reasonably stable environment, you will spend much of the engagement trying to catch up. 

Second, when funding allows, choose an engagement model that absorbs change rather than resisting it. Our LP model fits this well: it’s iterative and embedded rather than tied to fixed, front-loaded deliverables, and it flexes with shifts in scope and timeline without straining the underlying relationship. When the institution’s priorities move, the model moves with them. 

Practical operational and legal considerations may include: using modular training formats so that sessions can be rescheduled without losing coherence; leveraging interim letters of collaboration to maintain legitimacy while formal instruments are processed through government chains; and building relational trust during stable periods so that the institution is willing to re-engage when things settle down.

5. Staff transitions carry more relationship risk than most people plan for

Staff changes happen internally and externally. When a key Parliamentary counterpart moves on, the working relationship does not automatically transfer. A new Director arrives with their own priorities and their own need to understand why the partnership is worth continuing. If the work has not been grounded in institutional systems, a single departure can set things back considerably. To mitigate this, establish formal agreements, documented frameworks, and institutionalize the tools. To further safeguard the partnership’s momentum, cultivate multiple champions at various levels within the client team, so that knowledge and support are institutionalized and not dependent on a single individual’s tenure. 

In Zimbabwe, leadership transitions meant we had to reintroduce ourselves at the senior level before we could move forward. The same discipline applies on our side: When an IDinsight team member leaves the engagement, we communicate early, introduce successors in person where possible, and ensure the incoming person is adequately onboarded on all aspects, including the client relationship: A well-handled transition signals that the institution is the partner, not a single individual. 

A note for partners interested in working with Parliaments

The gap between the evidence available and the evidence used in legislative processes can be closed if partners earn the right to be heard by being demand-driven rather than agenda-led, staying neutral under pressure, reading what goes unsaid, and designing for turbulence rather than assuming stability. What Kenya, Zambia, and Zimbabwe showed us is that evidence gets used when the institution trusts the messenger, recognises the task as its own, and has the internal capacity to act on what it receives. For organisations willing to work on those terms, patiently, contextually, Parliaments are among the most consequential drivers of evidence-driven policy making.