Advocacy organizations and nutrition policy in Nigeria: identifying metrics for enhanced efficacy

Abstract Advocacy organizations have played a significant role in the field of nutrition in recent years. However, why are some advocates viewed as more effective than others? This paper derives metrics for assessing advocacy efficacy by first drawing on key insights from the nutrition and public policy scholarship. A set of metrics is proposed to capture the constitutive elements of three concepts that often emerge as critical from that literature: organizational capacity, strong networks and external outreach. Based on a survey of 66 nutrition stakeholders in Nigeria, including at the federal level and within the states of Kaduna and Kano, the metrics are then applied to a set of advocacy organizations within the country. We show that the metrics can provide insights into why some advocacy organizations are perceived as more effective than others by policymakers. Specifically, we find that geographical reach, the share of budget allocated to advocacy, action plans with clear objectives, large networks that include government and non-governmental policy champions, multiple media and dissemination outputs and numerous training events collectively increase nutrition advocates’ visibility to, and influence on, policymakers. Although the metrics are subject to further testing in other country settings and need to be interpreted based on a country’s underlying policy system, they offer a useful starting point for more systematic, comparative advocacy analysis and learning within the nutrition field and beyond.


Introduction
Over the last decade, nutrition has received elevated attention in international and national policy arenas. For instance, nutrition's importance is affirmed by inclusion in the Sustainable Development Goals and elevated by financial commitments from global leaders through initiatives such as Nutrition for Growth. Policy advocacy has played a significant role in this shift by raising awareness about underlying causes of malnutrition, emphasizing associated economic and social costs and creating targets by which governments should be held accountable (Pelletier et al., 2013;Mejía Acosta and Haddad, 2014;te Lintelo et al., 2016). However, in the crowded arena of nutrition advocacy, why are some advocates viewed as more effective by policymakers than others?
To address this question, this paper reviews existing literature on the components of effective nutrition advocacy and proposes a way of operationalizing these components. Subsequently, it discusses how the proposed metrics fared in the case of Nigeria. After describing how data was collected from 66 key informants, including advocacy organizations, government officials, donors and academics, we show how policymakers perceive the efficacy of different advocacy organizations. Variation in perceived effectiveness of different organizations is then assessed vis-à-vis the proposed metrics to identify which are most robust and should be considered by those interested in leveraging advocacy in the nutrition arena.

Theoretical context
There are numerous definitions of advocacy (see te Lintelo et al., 2016). For instance, Gen and Wright (2013, p. 165) note that advocacy consists of 'intentional activities initiated by the public to affect the policymaking process. ' Pelletier et al. (2013, p. 86) define advocacy as 'a continuous and adaptive process of gathering, organizing, and formulating information into argument, to be communicated to decisionmakers through various interpersonal and media channels, with a view to influencing their decision… ' Cullerton et al. (2018, p. 83) noted that advocacy is 'The process of undertaking active interventions with the explicit goal of influencing government policy.' At their core, these and other definitions all see advocacy being an ongoing and interactive process, targeted primarily at policymakers.
Many different stakeholders can be advocates for particular positions, including multilateral and bilateral donors and the business community. However, we define advocacy organizations as non-profit agencies whose core mandate involves promoting particular causes, ideas and norms (Keck and Sikkink, 1998) or who participate in such promotional activities in addition to providing direct services, including technical training, community education and program implementation (see Kimberlin, 2010). Advocates may have achievements in multiple ways, such as by building trust within local communities or fostering dialogue on polarizing issues. However, following Raynor et al. (2009), this paper defines advocacy efficacy as attaining some type of policy outcome, such as through identifying a policy problem, shifting the policy agenda, changing policy design, or facilitating policy implementation. Such an explicit focus on policy outcomes corresponds with much of the empirical scholarship on nutrition advocacy discussed below.
While there has been an expansion of advocacy evaluation methods over the last decade (e.g. Gen and Wright, 2013), these approaches aim to guide advocacy organizations to selfreflect on their goals, benchmarks and strategies. However, there is less attention to how advocacy organizations can be assessed in relation to one another. Therefore, we elaborate on three interrelated components that extant literature collectively and consistently suggests are necessary-although by no means sufficient depending on the enabling environment-for effective advocacy: organizational capacity, strong networks and external outreach.

Organizational capacity
Advocates in any field require a minimum set of human and fiscal resources to build momentum and scale-up activities (McCarthy and Zald, 2002). Comparative case studies of nutrition policy and broader advocacy show that organizations' achievements rely on, inter-alia, staff with relevant skills and knowledge about the specific policy domain who can determine which research and information is credible (Elbers and Kamstra, 2020;Pelletier et al., 2013). If the aim is to influence policy at the subnational level as well, the organization requires the presence of trained staff outside the capital city (Harris et al., 2016). Others find that advocates are likely to be more successful if they have strategic capacity, including a clear, tactical vision, demonstrated by time-bound objectives that are supported by action plans or an organizational strategy (Reisman et al., 2007).
As Prakash and Gugerty (2010) observe, advocacy organizations are motivated not only by normative concerns, but also by instrumental ones related to their organizational survival and growth. Therefore, financial resources are another key sub-element of organizational capacity, which need to be available for the duration of the advocacy effort, whether a one-time campaign or a more long-term engagement (McCarthy and Zald, 2002;Shiffman, 2016).

Strong networks
Individual organizations can enhance their resilience and adaptability by establishing strong coalitions or networks, which have proved essential for advancing policy change in spheres as diverse as nutrition, education and the environment (e.g. Di Gregorio et al., 2019;Harris, 2019;Sabatier and Weible, 2007). One of the major public policy perspectives is the Advocacy Coalition Framework (ACF) advanced by Sabatier and Jenkins-Smith (1993). In this view, such coalitions are united by a set of policy beliefs within their particular policy subsystem, such as nutrition policy and a geographical focus, such as a state or country. The ACF argues that coalitions are founded on a hierarchy of beliefs. Specifically, deep core beliefs are overarching normative assumptions about how governments should act. Policy core beliefs refer to the expected role of actors in addressing the deep core beliefs while secondary beliefs are narrow and focused on policy mechanisms (Sabatier and Weible, 2007). For the nutrition community, there is unity in deep core beliefs that tackling malnutrition is a fundamental responsibility of governments and increasingly of the private sector (Development Initiatives, 2020). There is also shared policy core beliefs that malnutrition needs to be tackled through multi-sectoral interventions. However, disagreements occur at the secondary belief level regarding the policy instruments, such as behavior change campaigns, ready-to-use therapeutic food (RUTF), or fortification strategies, that should be prioritized.
A critical consideration is the composition of the network. A dense network ensures the incorporation of diverse skills and policy contacts, potentially leading to more creative policy solutions and a more holistic understanding of the enabling environment. However, the trade-off is that more diversity can also contribute to conflicting perspectives among network members (Shiffman, 2016;te Lintelo et al., 2016). Disputes can result in mixed messaging to decisionmakers, leading to either policy inertia or contradictory policy interventions. A formal governance structure can assist with inter-organizational cooperation by mitigating potential conflicts among network members and enhancing collective action among members (Shiffman, 2016).
Such networks often require building alliances with a range of policymakers to better tailor policy options to the decisionmaking environment (Cullerton et al., 2016). Often, building linkages with a powerful policy champion has helped advocates establish local legitimacy and build momentum (Balarajan and Reich, 2016;Resnick et al., 2018). Policy champions often are 'insiders' in the reigning power structures and derive their influence from their status and power. Such champions should not only be limited to elected officials but also include senior bureaucrats from relevant ministries (Pelletier et al., 2013;te Lintelo et al., 2016)

External outreach
To both build networks and exert policy influence, advocates need to pursue a variety of external outreach activities that communicate their positions to targeted stakeholders. Such communications must be derived from evidence that is viewed as credible by the local communities in which they are operating (Harris et al., 2017). Moreover, the way in which advocates frame their positions and their policy issue is essential. Frames ideally reorient thinking about an issue by identifying a problem and a solution as well as suggesting who is affected and the consequences of inaction (Chong and Druckman, 2007).
The selected frame needs to not only resonate with external actors, including political elites (Gillespie et al., 2013;Shiffman et al., 2016), but also reflect a consensus within the broader advocacy network about the policy beliefs discussed above. Depending on the issue, such framing strategies can be difficult for nutrition communities to agree on, as shown by Pelletier et al. (2012) in Guatemala and Harris (2019) in Zambia. Shelley (2012) argues that framing over obesity has led to a 'policy cacophony,' creating confusion among decisionmakers about which interventions should be prioritized. Frames that work best appear to convey simple messages digestible to both the public and politicians and build on values respected in the target society (Freudenberg et al., 2009;Cullerton et al., 2016).
Advocates also need to disseminate their work and positions in multiple ways to engage in frame 'amplification' (Cullerton et al., 2018). This requires both inside and outside tactics (Dellmuth and Tallberg, 2017); the former involves direct interaction with decision makers through phone calls, email exchanges and closed-door meetings while the latter focuses on mobilizing public opinion through awareness campaigns, op-eds and other modalities to indirectly pressure decisionmakers. For instance, the Advocacy Working Group in Uganda pursued its efforts around stunting policy through policy briefs, newspaper articles, a documentary on local television stations and radio call-in programs (Pelletier et al., 2013). Active forms of policy engagement, such as high-level events, parliamentary briefings or journalist training workshops, can help to reinforce a particular frame and communicate evidence about complex nutrition issues (Harris et al., 2017). Figure 1 summarizes how these factors contribute to policy outcomes and interact with each other: organizational capacity allows for scaling and sustained momentum, strong networks provide resilience and adaptation to new policy developments and outreach ensures visibility and memorability in an increasingly crowded marketplace of ideas and priorities. Each of these factors is embedded in and influenced by the enabling environment, inclusive of the external institutional, political and legal settings in which advocates operate (e.g. Harris et al., 2017;Resnick et al., 2021). Yet, given that many dimensions of the enabling environment are constant in our case study due to the use of a single country case study, the current paper only focuses on metrics for advocacy that are more internal to organizations and over which they can exert agency to directly change. Similarly, while there are other potential factors that shape advocacy efficacy, such as seizing windows of policy opportunity and honing skills as  (Raynor et al., 2009;Cullerton et al., 2018), such dynamics can be idiosyncratic and difficult to measure at the organizational level.

Materials and methods
The preceding section shows there has been significant learning across diverse contexts about what makes advocacy effective in general and particularly with respect to nutrition policy. However, there is less effort to operationalize these lessons, which prevents advocates and their supporters from determining where they should invest resources to improve their efficacy. Moreover, there is often a lack of information about whether some of these dimensions of advocacy are more salient and consequential than others.
To address this gap, we analyzed the nutrition advocacy landscape in Nigeria where 37% of children under five are stunted (National Population Commission NPC and ICF, 2019), exceeding the African average of 30% (Development Initiatives, 2020). In addition, 68% of Nigerian children between 6 and 59 months are anemic and only 29% of infants under 6 months of age are exclusively breastfed (National Population Commission (NPC) and ICF, 2019). The importance of nutrition is widely acknowledged by different government ministries, and there are at least 19 nutritionrelevant national policies within the country (Vanderkooy et al., 2019). The advocacy community is also extremely rich; one study found that there were more than 200 international and domestic non-profit and non-governmental organizations working in the nutrition domain (CS-SUNN and UNICEF, 2017).
Key informant interviews (KIIs) were conducted with 66 stakeholders to identify which advocacy organizations were most effective at influencing policymakers and, in turn, whether those organizations had specific characteristics that correspond to the three sets of factors reviewed above, i.e. organizational capacity, strong networks and external outreach. Since Nigeria is a federal country, many policy interventions are formulated at the national level but implemented at the state and local government area (LGA) levels. Thus, explicit attention was given to both levels, with a focus on the states of Kaduna and Kano. These two states in Nigeria's northwest region are where many advocacy organizations are concentrated because they fare much worse on malnutrition indicators than states in the south of the country. For instance, child stunting prevalence ranges from 46-56% in Kaduna state and as high as 57-66% in Kano. Comparative ranges in states such as Delta, Lagos or Cross River are 14-24% (National Population Commission (NPC) and ICF, 2019). The KIIs were conducted according to a detailed and consistent interview guide that is publicly available upon request. The identified stakeholders were selected from a list provided by a longstanding West Africa nutrition initiative led by one of the author's institutes. We then shortlisted those that were still operational and located in Kaduna, Kano and/or the Federal Capital Territory of Abuja and focused on those with a focus on either infant and young child feeding (IYCF) or food fortification advocacy. These two areas are not only key priorities in Nigeria's National Policy on Food and Nutrition but also involve appealing to different groups of policymakers. As such, advocacy efficacy can be assessed vis-à-vis stakeholders within the broad nutrition community as well as by those who work in one of the two specific nutrition domains. This purposive sampling ensured that our conclusions did not just reflect the views of one sub-community within the nutrition advocacy landscape.
In addition to conducting interviews in the capital of each state, health and agricultural departments in two LGAs each (Giwa and Kachia in Kaduna, Bichi and Wudil in Kano) were selected in consultation with each state's nutrition officer and to reflect geographical variation within each state (i.e. an LGA from the North and South of each state).
LGA departments are frontline nutrition service providers and those selected were already implementing Community Infant & Young Child Feeding and Community Management of Acute Malnutrition interventions, ensuring that LGA stakeholders would be knowledgeable enough to speak about health advocacy and government support for IYCF and food fortification. This process was important for ensuring that our findings did not simply reflect the perspectives of a narrow group of actors, i.e. national policymakers in the capital city, but rather accounted for the opinions of a broader range of actors in the policy process.
The interviews occurred from October to December 2019, and as shown in Table 1, spanned five main categories. While other studies have looked at stakeholder advocacy in Nigeria (e.g. Allcock and Barker, 2012), this study differed by approaching respondents bilaterally rather than in a workshop setting. By conducting face-to-face interviews, respondents could be more forthcoming about which advocacy organizations they felt were more effective. Appendix Table A1 provides a full list of stakeholder organizations that were interviewed, and Appendix Table A2 offers a full elaboration of organizational acronyms.

Perceived efficacy of advocacy organizations
Our main outcome variable is perceived effectiveness of advocacy organizations by other, non-advocacy organizations, including donors, government decision makers and implementers and private sector actors. This was assessed in two ways. First, these non-advocacy respondents were requested to name the top three most effective nutrition advocacy organizations in either Nigeria or their state. Figure 2 illustrates the range of organizations that were identified. Those organizations in boldface were those that were also included in the survey as respondents while those in non-boldface were not. The figure highlights that organizations such as UNICEF, Save the Children and FHI 360 were perceived as the most effective, in addition to the Civil Society Scaling-Up Nutrition in Nigeria (CS-SUNN) non-profit, which is a hybrid organization that includes international and domestic nutrition advocacy organizations.
Second, the same set of respondents were asked whether they could attribute a policy accomplishment to the advocacy organizations that they viewed as most effective. Figure 3 illustrates responses per organization based on a three-point coding of the interviewees' qualitative answers. For instance, if a respondent could not attribute any accomplishment to an organization they designated as effective, that was coded as 'don't know.' Answers that were relatively generic, such as 'Domestication of food and nutrition policy in the state' or 'Generate discussion on policy review,' were coded as providing 'moderate specificity.' Detailed responses, including 'Part of the team that ensured six months of maternity leave' or 'brought forth the issue of hidden hunger which is now a serious policy objective of the Federal Ministry of Health,' are coded as 'high specificity.' Generally, the organizations viewed as more effective in Figure 2 also have a larger share of respondents who can identify accomplishments with high specificity in Figure 3. In other words, respondents did not simply gravitate towards certain organizations because of name recognition, but they could associate specific actions with those organizations. Although we did not carry out detailed case studies to confirm each organization's objective achievements, we believe that this approach captures how well advocates are making their positions and interventions known to a diverse range of policy actors.
Combined, Figures 2 and 3 illustrate that while some organizations are very visible and perceived as highly effective, others lie at the other end of the spectrum. To understand why, we follow best practice for comparative case study research (George and Bennett, 2005;Seawright and Gerring, 2008) and focus on six organizations that fall at differing levels of perceived effectiveness. Specifically, we are following the 'diverse case' study method for hypothesis testing, which Seawright and Gerring (2008) define as selecting a minimum of two cases that represent extreme values on the dependent variable as well as additional cases that represent the mean or median. Because this method encompasses the full range of variation, it is likely to enhance the representativeness of the sample cases to others in the broader population of cases.
At the one extreme, UNICEF and CS-SUNN were identified as among the most effective organizations in the country on nutrition. For instance, one observer noted that 'The Health Policy and Planning, 2022, Vol. 37, No. 8  totality of the nutrition policies that we have had in Nigeria was led by UNICEF. The first policy, the second and the third was led by UNICEF, including the establishment of the coordination structures' 1 while CS-SUNN. 'Played a very critical role in the development of the nutrition policy and the multisectoral food and nutrition plan of action…They created an opportunity for us to know the funding gaps at the state level.' 2 At the other extreme, no one identified the KAF Care Foundation or the Partnership for the Promotion of Maternal and Child Health (PPMCH), even by stakeholders in Kano state where those organizations are based. FHI360 and Nutrition International (NI) provide interim cases between these two extremes.
To understand the variation across these six cases, the next section applies the concepts reviewed earlier-organizational capacity, strong networks and external outreach-and shows which metrics could be used to reflect those concepts in a meaningful way. Table 2 above summarizes the metrics in general terms while Tables 3 through 5 provide details for each of the six organizations as well as the interview questions used to elicit responses. To enhance transparency and avoid assumptions of an omitted variable bias, we provide the range of data collected but boldface those metrics that demonstrate the greatest salience in explaining variance in perceived efficacy.

Organizational capacity
Organizational capacity can be disaggregated into three subcomponents: human resource capacity, financial capacity and strategic capacity. These issues were observed as essential for advocacy in Nigeria: 'The question is do organizations have the time and budget to advocate in a way that will establish the minimum number of contacts points needed? This requires funding and human resources.' 3 For human resource capacity, having a sufficient ratio of staff with experience in nutrition compared to the number of employees within the organization feasibly allows for greater outreach, as does the number of offices and programs For UNICEF, CS-SUNN and FHI 360, having offices in both the capital and in multiple states contributed to their higher recognition by both federal government stakeholders and their counterparts in Kaduna and Kano states (see Table 3). Not surprisingly, more offices are correlated with more activities at the LGA level as well.
The two financing metrics need to be examined in tandem; ideally, advocacy efforts could be more impactful if a higher share of resources can be allocated to this goal and if organizations already possess those resources rather than rely on anticipated appropriations. While NI claimed they had 85% of their needed resources in an average year, only 20% is intended for advocacy. By contrast, UNICEF targets threequarters of their resources for advocacy, even if they report having less than half of needed funding already in place within an average year.
As seen in Table 3, the length of an organization's existence does not play an obvious role in explaining its perceived efficacy, and all six organizations note that they have an organizational strategy. The main difference is the specificity of their goals. KAF only claims that it has a program on IYCF and PPMCH notes it supports maternal health but neither organization provides details on the way in which it is promoted. FHI 360, NI and UNICEF all have very specific outcomes they plan to promote in certain constituencies (e.g. women, children), and FHI 360 and UNICEF have targeted geographical areas of focus (e.g. certain states, vulnerable areas).

Strong networks
To identify the composition of advocacy networks, all 23 advocacy organizations in the survey were asked to identify the names of up to three organizations with which they partner most frequently to advance their objectives (see Table 4). Imposing a ceiling reduces the likelihood of respondents listing large numbers of minor partners and concentrates attention to the most important. Figure 4 maps the responses listed in Table 4 to provide a visual interpretation of the nutrition advocacy network. The size of the circles approximates the 'degree centrality,' which counts the number of links by each organization in a network. The shapes of the advocacy organizations indicate whether they are international (circle), domestic (diamond), or hybrid (triangle) entities. UNICEF is clearly the partner that most organizations mentioned as a partner, followed by CS-SUNN and FHI360. By contrast, PPMCH and KAF could only identify two advocacy organization partners, and KAF was not identified as a partner by any other organizations in the sample. Despite not having a high level of recognition by policy stakeholders, NI has a large number of advocacy partners, and spans the divide between the IYCF and food fortification communities. This reflects the organization's transition in mandate in 2017: ' [We] used to focus only on micronutrients. With the change of name to Nutrition International, the organization recently expanded to broader nutrition issues such as adolescent nutrition and institutional coordination.' 4 However, the intensity of the network activities is not equivalent. For instance, both UNICEF and CS-SUNN work with their partners to engage in both inside tactics directly with the government and outside tactics with communities on specific issues of breastfeeding, budget allocations, resource mobilization and micronutrient supplementation.
As noted earlier, policy champions are central to networks' legitimacy and visibility. We assessed policy champions from two perspectives. The first focused on institutional policy champions, particularly government ministries. To improve perceived effectiveness, more policy champions are better than fewer, and creating linkages at multiple levels in Nigeria theoretically increases the likelihood of influencing both policy formulation and implementation. CS-SUNN, FHI 360 and UNICEF have found ministerial partners at these dual levels. Yet, instead of noting a public health or nutrition champion, CS-SUNN and UNICEF view bureaucratic actors within budget and planning as major supporters for their activities. This is significant because when all 66 survey respondents were asked which government institution is most important for nutrition policy in Nigeria, 45% (30 respondents) identified either the Federal Ministry of Finance, Budget and National Planning (FMFBNP) or the State Planning and Budget Commission. A key reason for this is because the National (state) Committee on Food and Nutrition is housed within the Federal (state) budget ministries, which convenes the committees' meetings, and has been critical for pushing for nutrition interventions to be costed within the federal and state budgets. Moreover, the federal MFBNP leads multi-sectoral coordination around the National Policy on Food and Nutrition in Nigeria and reports directly to the Vice-President on nutrition activities (World Bank, 2018). In other words, CS-SUNN and UNICEF partnered with a government actor with substantial power in the decision-making space in Nigeria, reflecting an understanding of the underlying policymaking system within the country and states. By contrast, KAF lacked any government champions in its network while PPMCH pointed to working with the National Orientation Agency (NOA) whose mandate encompasses issues beyond nutrition, including electoral violence and the environment.
Advocates also were asked about their linkages with specific individuals, including politicians, celebrities and other prominent public figures, who support their activities. Notably, KAF did not identify any such champions of this type. However, the other organizations work with the wives of state governors, particularly Kaduna state governor's wife, Hajia Ummi El-Ruffai, as well as the Nigerian first lady (Aisha Buhari) and the former Emir of Kano and former Central Bank Governor (Sanusi Lamido Sanusi) (see Figure 4). These are three of the most active, well-known public figures in the nutrition sphere in Nigeria: El-Ruffai established in 2017 the Kaduna State Emergency Nutrition Action Plan (KADENAP) to fast track the work of ministries that deal with nutrition, women and children's health, Sanusi frequently speaks about the scourge of malnutrition on the country's development, and Buhari established the Future Assured Foundation, which focuses on child well-being.
As noted earlier, while density can be favorable for a network, the downside is that too many participants with different perspectives may affect cohesion. For KAF and PPMCH, no key areas of disagreement within their networks could be identified, likely reflecting the narrowness of their networks. PPMCH notes that its network agrees on the need to improve nutrition in Kano but provides no specificity for how members believe this should be achieved. This can be problematic for identifying areas of commonality that can be used as a springboard for action. By contrast, UNICEF's network's focus on the first 1000 days is specific enough to focus on the urgency of early childhood interventions, thereby serving as a unifying, policy core belief. This can help overcome divisions within its network over secondary beliefs, including whether to prioritize treatment (e.g. ready-to-use therapeutic foods) or prevention (e.g. IYCF) of malnutrition in certain circumstances. Both FHI360 and NI have narrower axes of agreement focused on policy levers they see as most effective to promote nutrition, including IYCF and vitamin A supplementation, respectively. The potential trade-off is that they can offer focused targets for government action at the expense of engaging cooperatively with a broader range of advocates.

External outreach
As elaborated earlier, external outreach is most effective when it involves communicating messages based on credible evidence that are framed appropriately and delivered through multiple written, audiovisual media and in-person activities. The credibility of evidence can depend on whether there is a process for ensuring its rigor and transparency, including a process of peer review by experts or availability for public scrutiny. As shown in Table 5, FHI360, NI, CS-SUNN and UNICEF rely heavily on national survey data, such as publicly accessible Demographic Health Surveys and the National Nutrition and Health Survey. Three of these also conduct their own in-house data collection where needed. KAF conveyed a limited source of data outlets that inform its advocacy work.

None
There is a divide on whether the focus should be on treatment versus prevention. Also, there is a debate on fortification of certain food vehicles.

Figure 4. Network of Advocacy Organizations
When developing a frame for mobilization, actionable outcomes with clear processes are more useful because they suggest a clear policy position that decision makers can digest. While all six organizations provided a frame, the positions of KAF and PPMCH are quite vague and offer no actions. NI provides an aspirational frame but does not demonstrate how to reach the goal for which it strives. On the other hand, UNICEF identifies actions-domestic resource mobilization and policy/strategy development-but did not articulate the ultimate objective for such actions. Of the six organizations, FHI360 offers a clear objective and processes with measurable targets.
The metrics on dissemination and policy engagement activities have a high correlation with the patterns of efficacy observed earlier. CS-SUNN, UNICEF and FHI360 conduct a wide variety of media activities and produce multiple written and visual outputs while also holding regular outreach activities with decision makers and civil society. CS-SUNN in particular targets particular events, such as World Food Day and World Breastfeeding Week, holds 'Nutrition Tuesdays' on Twitter, and capacity training with media and legislators. As the CS-SUNN informant noted, 'Meetings with the legislators had an impact because of their oversight func-tions…Meeting with the people who also do the budget had an impact because they may otherwise remove the budget [for nutrition] if they are not aware.' 5 By contrast, KAF and PPMCH lack operational websites, and the former is limited to newspaper op-eds for much of its outreach, which can undermine engagement in low-literacy communities. These patterns are further reinforced in Table 6, which shows that non-advocacy respondents recall attending many more events, engaging in more informal interactions and receiving regular written updates by CS-SUNN and UNICEF, followed by FHI 360.

Discussion
Collectively, the metrics presented in Tables 2 through 5 reinforce that multiple measures are needed to assess when and why some advocacy organizations are perceived as more efficacious than others. No one indicator alone will provide the full story of why some advocates are more visible and impactful. Therefore, a comprehensive and systematic stocktaking of  an advocacy organization's internal capacities, networks, and outreach activities is needed. At the same time, some metrics were more informative than others to understand variation in efficacy across the six focus organizations, and these were indicated in boldface within the corresponding tables. Specifically, many staff with relevant nutrition skills, a broad geographical reach, sufficient financing devoted to advocacy from extant resources, and an action plan with clear objectives differentiated CS-SUNN and UNICEF and to some extent, FHI360, from the organizational capacity of KAF, NI and PPMCH. Network strength is conveyed by the degree of centrality an organization plays in the nutrition landscape and its integration of powerful government and non-governmental policy champions. For external outreach, using multiple forms of media, ensuring an operational website, producing diverse outputs and holding multiple training events a year increases visibility. When this particular sub-set of metrics are examined together, the reasons why these organizations fall along a spectrum of perceived efficacy by other in-country stakeholders-with CS-SUNN and UNICEF at one end and KAF and PPMCH at the other-becomes more apparent.
Notably, KAF and PPMCH are domestic advocacy organizations but were not identified as top advocacy organizations even by LGA and state-level actors within Kano, where they are based. The application of the same metrics to other domestic organizations, including FOMWAN, TDI and Wazobia, reveals that they fare worse than international ones and are less recognized by other stakeholders. This finding may reflect our narrow definition of advocacy efficacy, which focuses on nutrition policy outcomes and therefore does not capture broader societal achievements; indeed, Kurfu (2018) shows that FOMWAN has been instrumental in enhancing Muslim women's empowerment in Nigeria. Nevertheless, our analysis vis-à-vis nutrition policy suggests worrisome cyclical effects whereby domestic organizations are overlooked by donors for financial support on nutrition advocacy that potentially would allow them to improve their metrics, therefore causing them to continue to be viewed as less effective. This raises questions about the local ownership of nutrition policy and the balance of power between international and domestic organizations that is increasingly recognized as problematic within public health (Shiffman, 2014;Harris, 2019;Storeng et al., 2019). Simultaneously, CS-SUNN's ability to achieve a high level of perceived efficacy among policymakers highlights the advantages of its hybrid modality.
Additional validation of the metrics is required in other policy settings where the underlying enabling environment may further streamline the requisites for effective advocacy. In settings with more centralized or more authoritarian decisionmaking structures, fewer policy champions may be needed or external outreach through multiple media outlets might not be feasible. In other words, as highlighted in Figure 1, the enabling environment will structure how the metrics should be interpreted. By way of example, the relevance of the FMBNP as one of CS-SUNN's and UNICEF's perceived policy champions is not obvious without knowing the significance of that ministry in Nigeria's nutrition policy landscape.

Conclusion
Neither the importance of advocacy for nutrition policy reforms nor the discussion of some of the advocacy metrics reviewed here, may be surprising. Yet, in practice, there are few attempts to operationalize many of the best principles for nutrition advocacy. Establishing metrics that can be refined, tested over time and applied in different settings can both assist with comparative research on advocacy efforts in a more a systematic way, as well as contribute to organizational learning about gaps that need to be addressed to exert more influence in both the nutrition field and other policy spheres.

Supplementary data
Supplementary data are available at Health Policy and Planning online.

Data availability
The data underlying this article cannot be shared publicly out of respect for the privacy of the individuals who participated in the study and their roles within the advocacy organizations that they represent. The data will be shared on reasonable request to the corresponding author.