HE-BCI major review user engagement activities

We held a webinar on Wednesday 1 May, the recording including a presentation and Q&A are now available

HE-BCI 1 May webinar - PowerPoint slides

Can I just check that for the new spin-out census and new Table 4a, we would only be required to list organisations that fit the spin-out definition only, and not other categories reported in Table 4 such as ‘Student Start-ups?’

This is correct. The new definition provides a baseline on what we think a spin-out company is.

The crucial thing is that it has come out of research from the institution. Have you considered that in small providers, the same person will be doing HE-BCI and the Student (23056) return?

The 23056 collection has to take priority over this activity. We also do not have this data to hand, so would require significant data collection activity at a time when we run on reduced academic staff. This is an important question. We have done an analysis of providers who have returned to Table 4 in the past 9 years. Through this analysis we’re anticipating that a small amount of providers will account for the majority of the data, whilst large amounts of providers will have little to no data, and smaller providers are more likely to be in this category. There is a middle group of around 70 providers who we estimate will return around 25 lines of data. We are working on a quick start guide to help providers to identify whether they are likely to have any of this data.

Can you provide some clarity on “other data sources,” which will be cross-referenced with the spin-out census? For data linking, will you be only relying on Companies House for data, or will you be using other sources too?

Companies House has some issues with incomplete data (like not requiring companies to report investment or employment figures) and accuracy (since it relies on companies to submit accurate information without verifying it). We are looking at ways that we can use the data that providers return to identify spin-outs in other third party data sources. For example, if a Companies House ID is provided, we are hoping to pick up a range of data from the Companies House database. There are also other commercial databases, and the possibility of using data held by the ONS to provide further information on employment, turnover and investment.

Does this data collection cover graduate startups or is there new guidance on these, or is the guidance the same? Do all of the slides around data collection on spin-outs apply to Graduate startups or is there any new guidance on these?

We are not collecting individual item line data on graduate start-ups. By improving the definition of a spin-out, this has implications for the definition of a graduate or student start-up. We have dealt with this in the guidance.

How will the data [for the Spin-out Census] actually be collected? Do we just provide a spreadsheet?

The format will influence who and how many people across the university need to be involved. Yes, a spreadsheet is how this data would be returned. There is a draft version of the template available for review as part of the open consultation.

What’s the difference between the HMRC vs Frascati definition of research and why do you use the HMRC definition?

The Frascati definition is: “Research and experimental development (R&D) comprise creative and systematic work undertaken in order to increase the stock of knowledge – including knowledge of humankind, culture and society – and to devise new applications of available knowledge”. In comparison the HMRC defines research as: “original investigation undertaken in order to gain knowledge and understanding”. We have used the HMRC definition given the financial nature of the information that is sought. HMRC define the rules around R&D tax credits, whereas Frascati offers a philosophical and principled position on what counts as research. The definitions are close in practice, but HMRC is the right one to choose in this context.

Table 4 - some of our spin-out are on stealth mode, how are you going to deal with confidentiality data?

We're considering this at the moment and hoping respondents will give us some background on prevalence of stealth-mode companies. We'd prefer to collect and protect this data rather than create an incentive for non-return of the data. Options for handling this are either 1) provider's raise this with us ad hoc during collection, or 2) we add a column to indicate that this company cannot be included in outputs, other than those for statutory users. Or 3) an alternative good idea proposed by a respondent.

So we wouldn't have to include any spin-outs formed before August 2012 which are no longer University spin-outs?

Yes, that is correct, proposing to just collect spin-outs during the reference period (01 Aug 2012 – 31 July 2023).

Table 4: Origins for benchmarking: Origins field: having free text describing the IP does not provide a good set of data for downstream analysis. Also, text analysis can lead to misinterpretation and incorrect conclusions. An alternative way would be the establishment of a robust classification system for economic sectors, such as MedTech, Healthcare, Engineering, Renewable and so on, which can provide actionable insights.

We agree - but no usable pre-existing typology has been identified. Free text will help us assess existing frameworks for suitability, or to create an appropriate new classification.

Is there any change to what you deem "active" spin-out?

No, we can’t see anything that we have proposed which changes what we deem as an active spin-out.

Will HESA be looking to reconcile the spin-out data vs past HE-BCI returns?

It is possible as part of the quality assurance process, but with the change in definition it’s not possible that these would completely match.

Table 4: Considering that the information regarding the spin-out census increases HE provider workload substantially, can the data be sourced by using the company number?

We are looking to use the identification data that providers return to help enrich the data we are able to pull in about spin-out companies. One of the key burden reduction areas we are looking at is can we produce estimates of turnover, employment and investment.

Why would reporting [the spin-out census] alongside normal HE-BCI timescales be more complex?

The population of interest for the spin-out census is historic data from 2012 – 2023. The population of interest for HE-BCI is only for the reporting year. The data fields between the spin-out census and HE-BCI are different to an extent and would create additional work in HE-BCI leading to burden. There is also a policy need for a national register, and having the data collection alongside HE-BCI would be later than a government would need the data for.

Will the guidance on staff and student start-ups be updated?

Arguably the wording around HE providers 'assisting' the formation of staff startups, or 'formal business/enterprise' support being provided to student start-ups is ambiguous and open to differing interpretations. We think that even if support has been provided, it won’t be a spin-out unless it meets the definition for a spin-out. We have not updated the definition substantially as we have heard from providers that they understand what is meant by formal business and enterprise support. It is not amenable to easy, simple sector-wide definitions because different approaches get taken in different institutions.

RE: Sub-head 4aii/4bii: Other spin-outs. "When a spin-out ceases to be a firm (whether independent or as a subsidiary of a group) or it ceases to be active, tracking should cease." Can we still include former spin-outs that are still recognisable in a larger corporate structure?

Depending on what is meant by recognisable, the idea is yes, you can if it is recognisable in the sense that it is still a company. For example, if it is within a group structure and is recognisable within that group structure. An edge case could be where the company is named as a department within a larger corporate organisation, but it does not have any separate or distinct identity. This would then fall out of tracking.

Can you tell us exactly who is going to see the 'disseminated data' with regard to spin-outs?

Our proposal is that we operate the standard onward supply categories that we have established with your provider. This gives you the opportunity to say that the statutory data that we've returned, we're happy to see that used onwardly. We give three options; just in research, in research and for sector benefit type activities, and for the widest range of uses that are legitimate for that data. It is your choice to say which one you would like to sign up to. Our proposal is that the data follows that logic, and we disseminate it according to that logic. We want to put some of the data including the identifiers including the Companies House number and equivalent, into a public data source so they can be used quite widely as open data. 

Will there be mandatory fields on the spin-out census spreadsheet? If we are unable to provide certain information will the submission be deemed incomplete?

The rules on mandatory fields are laid out in the guidance. All fields are required unless excluded by the rules spelled-out in the guidance. We will be measuring item nonresponse and raising queries where data is missing.

Can you send us an excel file template with an example for the spin-out census, to show the type of answers you are after?

This could form part of the some supporting information that we will look to publish to assist colleagues in reporting the data.

For Table 2 are we excluding overseas consultancy? If yes, why?

No, we are excluding consultancy for other HE providers (wherever they are based). There is no other restriction on international consultancy.

For Table 5 - Would you be able to indicate when a new set of data to collect would be finalised for implementation?

Not at this stage - we will indicate the direction of travel in the end-of-review report, this summer.

Please could you clarify what is the status with Table 5 – proposal to pause data collection?

Social and cultural interactions are an important part of knowledge exchange activity, so this is only a pause whilst we conduct further work into the area to improve metrics and definitions. If providers agree with this proposal, we are asking which year providers would want the pause to come into effect for. We have presented an alternative option if providers do not agree with this pause in the consultation.

Will all these changes (subject to consultation responses) be implemented for the 23/24 return? Considering we are nearly at the end of the year and may not have been capturing data throughout the year in line with changes, this could be difficult.

Yes, we are intending to implement this for the 23/24 return. The majority of changes have been tightening guidance. Where we’re collecting additional data on spin-outs we have been engaging with providers to see what data is captured. We are basing our proposal on what providers have told us they hold.