Skip to content

Blog

Seeds and saplings

Preliminary findings from an evaluation of business support programmes

16 July 2026

Maria Brackin, James Phipps

Experimental research

Share this page

Several months ago, we began an analysis of the long-term impacts of business support programmes on participating companies. We evaluated: 

The premise of these programmes was to attract businesses with the potential for growth or innovation, and to contribute to their success.

Our analysis was motivated by 2 overarching questions:

  1. Did the programmes spot high-potential businesses?
  2. Did the programmes make winners by supporting these businesses?

Disclaimer: This work was undertaken in the Office for National Statistics’ (ONS’s) Secure Research Service using data from ONS and other owners and does not imply the endorsement of the ONS or other data owners.

Let all flowers bloom


Before we present the results, take a moment to imagine a participating company. What type of business is it? How large is it? Where is it located? 

Having trouble? Indeed there is no one answer to these questions, and therefore no “typical” business that accessed support. Not every business was eligible: companies in GrowthAccelerator (and the Growth Impact Pilot) had to meet turnover and employment criteria, and applicants to the Innovation Vouchers Programme needed an innovative project with a knowledge provider. But apart from these basic requirements, any business was able to participate.

And so they did. The SMEs within the support programmes were hugely diverse:

In line with Nesta’s influential research, this huge diversity of businesses was an intentional feature of the programmes, which aimed to support all sorts of high-potential endeavours. However, this diversity had a profound effect on our analysis, and our findings must be understood in this context.

The garden of forking paths

From huge diversity of participants, to a huge range of outcomes: after their involvement in the support programmes, businesses went on to widely diverging fates, from shutting down entirely to becoming superstar high-growth companies. This variation was much higher than expected from the interim data used for the original power calculations, highlighting how outcomes can diverge strongly over longer timescales (our primary outcomes were measured over the 4-5 years after programme participation).

Why does this matter? First, it highlights the importance of measuring outcomes over longer timescales to capture the true potential, and pitfalls, of interventions. Second, there are powerful methodological implications. The more variable and unpredictable are outcomes, the lower the statistical power and higher the risk that we cannot identify a true difference between participants and comparison SMEs, or between experimental conditions. And when considering diversity itself – for instance, when comparing outcomes in different regions – this problem is even more pronounced. 

Another challenge comes from the datasets we used. Our analysis used the ONS Longitudinal Business Database (LBD), which is a valuable resource containing turnover and employment data for millions of businesses, but it does not capture other, more nuanced information. Businesses vary in ways that we cannot observe in this data – like looking at only the stem of a plant visible above the ground, unable to see the roots.

Answering key questions

Overall, we found strong evidence that the support programmes spotted high-potential businesses, but no evidence that the programmes made them into winners.

Did the programmes spot high-potential businesses?

Yes! Our results are consistent with the idea that the business support programmes attracted and identified high-potential companies.

In both the Growth Impact Pilot and the Innovation Vouchers Programme, a propensity score weighting analysis showed that participating companies generally outperformed other SMEs:

In GrowthAccelerator, a comparison between participants and a matched comparison group (identified via propensity score weighting) produced even stronger results: 

These results strongly suggest that the programmes attracted businesses that were high-quality and more likely to succeed than other SMEs.

We also found a strong relationship between the expert assessor scores and business potential:

This suggests that GrowthAccelerator not only attracted, but correctly identified those businesses with the most potential. 

Did the programmes make winners by supporting these businesses?

We don’t know for sure. A lack of statistical power and concerns about matching quality make drawing clear conclusions difficult. We found:

On the surface, the findings we presented above suggest huge impacts for the GrowthAccelerator programme. But as discussed, the challenge of finding comparison twins undermines the credibility of these headline figures. Additional difference-in-difference analyses comparing sub-groups of participants also present a more mixed picture:

Overall, we cannot confidently say that participating companies in GrowthAccelerator experienced better performance by taking part in the programme  nor can we say that they did not.. 

Taken together, the null results from the RCTs and the mixed results from GrowthAccelerator should be interpreted as no evidence of impact on our primary outcomes of interest, rather than evidence of no impact.  

However, for IVP our research collaborators have already shown how vouchers had beneficial impacts on earlier outcomes (innovation collaborations, product and service development, internal processes, and intellectual property protection) and we find new evidence of additional innovation funding.

Given the statistical challenges but also the rapid policy cycles (GrowthAccelerator was closed before any of these findings became available), we believe that future experiments should focus on testing early impacts, critical for success, even when planned as full impact evaluations.

Reaping and sowing

Our experience shows that evaluation, particularly over long timescales, is challenging. But there is much we can and will do in the coming weeks, months and years:

What happens next?

Although we are analysing old programmes, the questions we have asked are as timely and relevant as ever. Across the UK government, there is a renewed interest in how best to target, and support, high-potential businesses. Our analysis suggests that business support programmes are able to find and attract high-potential businesses, but it is much less clear whether they are able to accelerate growth. 

Our work is ongoing: