Data, Insights, Knowledge

“We need a lot more research about what's happening on the ground, not necessarily just funding research about how much money every company is raising, but really understanding how a particular industry operates in a particular market. How does regulation affect this particular market? And what does that say for investors and entrepreneurs? Or even just data on the numbers of this kind of asset, or the number of this kind of business in this region.” - interviewee

“There's no library where you can go and find good solutions to scale. Information is spread out. And it's often very hard to identify a solution, especially when there's not enough aggregated data to look at and compare projects in say Congo and South Africa. We don't have data sets to compare like for like. This makes it harder to get to scale.” - interviewee

Ad hoc and on-the-job learning are vital for sustained professional growth, all the more so in scale-ups, when learning trajectories are steep and rapid. Feedback from entrepreneurs to whom we spoke suggests that many founders do not know where to find practical, high quality information, data and knowledge at low or no cost. Moreover, given their time-starved founder roles, they want that information in bite-sized chunks to give them the quickest return on their time. They also want quick results. As mentioned elsewhere, there is often a reluctance to share information, stemming - sometimes incorrectly - from a protectionist view of their intellectual property. These factors lead to information asymmetry, which affects the equity of negotiations if one party has much more or higher quality knowledge than the other.

Collaboration is not normalised. Institutional efforts are uncoordinated, or not focused on commercial scaling. Few mechanics lend themselves to new data collaborations. Academic institutions potentially perceive competition, rather than collaboration, as their default position. Siloed behaviours will surely subsist, connections will not be discovered, and valuable opportunities will continue to be missed if there is a reliance on the old, rather than forging a new, different direction ahead. 

Collaboration would allow stakeholders to access extremely valuable benchmarked data about how they should be running and measuring their businesses. Scaling programmes would also be evaluated and assessed better for the benefits of investors, donors and corporate sponsors. Excellent expertise and models for designing data collaboration are evident and working. What is missing is a combination of strong leadership, alongside appropriate funding, which could help take the ecosystem forward strategically. 

The phenomenon of commercial scaling is new across academic, government, and research institutes. There are no approaches or consistent frameworks for capturing industry data; no dedicated databases exist. All the ecosystem focus has been on startups, not scale-ups. 

“There are islands - archipelagos of knowledge - but some good bridge building can allow a knowledge exchange of insights, ideas, data, and people coming together around certain themes.” - interviewee

Without a dedicated effort it will remain very hard to bring together the most useful data for understanding problems and how to solve them. Often data remains in private companies, and the current platforms and tools are proprietary. Furthermore academic researchers don't have the capacity or resources to do continent-wide research, meta-analysis, or cross-country comparisons. Whilst this situation persists there will remain huge gaps in evaluating how scaling ventures grow (and continue to do so), with no tracking or trend analysis. This leaves policy interventions consistently blunt, unverified, and likely ineffectual.

There have been recent calls to establish a public database as the absence of official databases in Africa adversely affects the regularity and rhythm of investment on the continent, increases the transactions and due diligence costs, and reduces investor confidence. Leaving such responsibilities to national government would be folly. 

Instead it is entirely feasible to establish an African scaling index database by bringing together a ‘task and finish’ group featuring expert researchers, such as Briter Bridges, Disrupt Africa, Asoko Insight et al. Championing financial support could be provided by the African Union, United Nations Economic Commission for Africa (UNECA), the African Development Bank (AfDB), and/or the African Private Equity and Venture Capital Association (AVCA). Data should be available so that public organisations and private actors can identify, target, evaluate and benchmark their support to scaling ventures. Joint efforts by multiple stakeholders such as governments, universities, research institutes, investors - in unison - would be highly beneficial. 

This would inform a greater understanding of scaling venture needs and how private and public sector engagement, resources and investment can help propel business growth. It would ensure scale-up ventures are properly placed on the economic map and provide suitable benchmarking each year to assess what has happened, and what more should be done. It would also act as an international barometer and assessor.

It’s no-one’s job to orchestrate and curate knowledge, evidence and data. There is no dedicated research institute nor body which provides strategic planning advisory support to the entrepreneurial ecosystem at large. The result is that activities remain largely reactive rather than strategic. Scaling businesses can't draw on examples - to understand and from which to learn - because they often have to orchestrate knowledge and data from scratch. This is incredibly inefficient and wasteful. 

The ecosystem could work to build a repository of data, knowledge and insights to ensure continuous (and rigorous) learning whilst also helping to assess interventions and programmes, based on verifiable evidence and measurable impact. 

“Communities of practice are highly beneficial for the cross-pollination of ideas.” - interviewee

Collective intelligence assemblies are a means of collecting and diffusing/ distributing data and knowledge. Much of the recent literature points to the need of making better use of public data sources, and to improve the governance of data, as well as better governance with data. Common themes include the need to establish data commons and using data exchange markets to open up the vast amounts of data that are available in the private sector and the public sector. Significant thinking has been developed which considers the governance with data, including augmented collective intelligence (ACI) frameworks, which build on state-of-the-art research and applications in AI, machine learning, internet of things (IoT), smart objects and smart contracts, as indicated by Figure 27. 

Specific examples of how collective intelligence approaches work in practice are usefully detailed by UNDP’s Accelerator Labs. A forward thinking entrepreneurial ecosystem could consider how to advance both human and machine intelligence to best serve its interests but this will require significant coordination, coherence, and collaboration commitments.

Presently, institutional deficiencies prevent sufficient knowledge sharing. It has been suggested that how we organise has always been interwoven with how we perceive ourselves. Our prevalent models of organisation are heavily designed with a bias to self-reinforcing ideas of humans as separate and self-interested individuals and our systems designed with an emphasis to stratify and control. But there are alternative ways to govern and organise that go beyond the rules that currently inhibit us. Future efforts need to be anchored in networks and movements, shared agency, trust and continuous learning.

Some simple first steps should be sharing basic intelligence for collective interests. There are a range of approaches to deploy knowledge to scaling ventures. Properly assessing demand factors - such as relevance, utility, potential effectiveness, technology principles, amongst other considerations - would be a helpful start.  Figure 28. points to a spectrum of how knowledge can be deployed and diffused to suit the scaling entrepreneurial context and its underlying principles. 

There are very limited case studies and stories of scaling success and failure, leading to one interviewee telling us that “there were no role models, essentially - nobody knew how to”. As more scale-ups emerge in Africa, this issue will start to be addressed. As will tackling the failure narrative.

In the US, prior experience of failure is usually seen as an asset, and this is backed up by studies. Banks and venture capitalists are more willing to invest in such entrepreneurs. Past entrepreneurial failures come with key learnings which play a vital role in adapting to dynamic environmental changes. Research indicates motivated and determined entrepreneurs bounce back from errors, failures, and setbacks.  

In most African cultures, on the other hand, failure is largely stigmatised. An expert on entrepreneurial failure told us, “There's very strong negativity around failure. So when people fail, often they don't share the underlying story. They don't share their experiences.” This is understandable, given how much is usually riding on the success or failure of a startup in the African context. Entrepreneurs’ parents don’t have a garage or basement from which they can launch their venture. There is no safety net, and no Plan B. The human (personal) cost of failure can typically be far higher.

The manner in which failure is attributed in Africa is another important factor. Strong mentorship and leadership development will assist in better internal attribution. 

“Their source of attribution of failure is interesting as it will often be linked to some outside conditions. Some people, especially those that are religious or have some superstitious beliefs, suggest this is the reason. There are some people who blame the economy. There is a wide range of attributions - from inflation to government subsidies, or rivals set up next to them, as sources of attribution. Where you have individuals who have an internal attribution, such as recognising that maybe they didn't manage things properly or lack specific expertise or qualifications, they are more likely to take up some kind of training or a government course initiative before they establish a new venture. The source of attribution is the most important differentiator.” - interviewee

Limited research resources are allocated to the development of helpful scaling “insight lessons”, in digestible, easy-to-access formats. More generally, ecosystem actors and investors have taken a competitive, rather than cooperative, approach to shared learning. Unfortunately, it’s still very hard to access reliable evidence about what works, where, why and how. This results in an over reliance on anecdotal evidence rather than core data and insights. There is no collective understanding, and few  frameworks on which to base performance. The lessons learnt are held by a few, which means invariably other businesses in the future will make costly mistakes; correspondingly, investors retain more risk.

“We need a lot more - and a variety - of research done about the ecosystem. But core research is always relevant. And understanding the past and how things have been done in the past, and what things are happening right now, actually gives us a great visibility on how things can go in the future. It also teaches us how to learn from our mistakes.” - interviewee

“Talk to the entrepreneurs who have scaled and try to trace back the steps of what they encountered, plus the type of support they received, which was essential for their growth.” - interviewee

“Fuckup Nights” - a global movement and event series was created to share stories of professional failure. Several chapters exist across African cities. We point to the Failure Hub too, which was designed by a young Argentine student (note that no African examples exist on the site). A full repository of scaling related case studies would be useful if properly signposted and contextualised. The result is success and failure are understood better, with helpful lessons learned and 'black box' knowledge diffused widely.  

New experimental approaches to public dialogue are needed. A central feature needs to be active deliberation, which allows time for participants to become informed, reflect on their own and others’ views, uncover issues in depth with other people and come to a viewpoint. The role of innovative techniques to improve public dialogue is an emergent field. 

New models are needed to trial constructive conversation, connection and cooperation on research and innovation related to scaling topics. Cutting edge digital and creative techniques can extend reach and facilitate ‘bottom-up’ and ‘informal’ participatory engagement. Opportunities are needed to reflect openly on the evidence on what is working (or not), what changes are needed, and how to shape and share practical responsibilities. UNDP and AFriLabs have started an Africa Innovation Policy Dialogues series, which is a helpful start, even if it appears a ‘broadcast’, rather than interactive, design.

Dialogue should not be episodic and ephemeral. Rather, pathways that cultivate and sustain functional working relationships between those who build solutions (and the associated emerging technologies), alongside the people who regulate them, is a necessity. Some excellent models, such as Open Loop, have been established; but few are yet active in Africa.  

We imagine a series of interactive online convenings to assess the latest scaling evidence. The Government Outcomes Lab -  a research and policy centre based in the Blavatnik School of Government, University of Oxford, is one model which can be explored. A dedicated series of virtual events would offer an open platform for policymakers, practitioners and researchers to engage with key scaling findings from the field. We also point to the Innovation Growth Lab, which has compiled a series of short and actionable “Evidence Bites” for policymakers and practitioners in the entrepreneurship and business support space. Developing a series of relevant insights from experimental research, actions which are contextualised for African ecosystem actors is an entirely achievable goal. It would be beneficial for the AU Policy and Regulation Initiative for Digital Africa (PRIDA) to actively absorb new participants, as well as open and participatory approaches, to their work.