Client:
Porterdale Business Consultancy works with small and medium sized businesses on all aspects of their business finance. Through years of experience they are familiar with the challenges companies face when looking to improve their commercial decision making, profitability and cash flow.
One specific area that Porterdale recognise as being an issue is financial forecasting. Detailed forecasting helps explore many aspects of business operations and planning, and a forecast report is often essential to raise finance. Using their experience, Porterdale developed their own tools and now wanted to develop one of these to expand the services they offer. By enhancing an existing model, it could be securely distributed and also allow valuable time to be spent advising clients rather than on the model itself.
The challenge:
How could existing desktop tools for forecasting be secured and protected so that Porterdale could commercialise them? In addition, Porterdale needed to be able to refine and modify their offering without having to employ a software developer for the purpose.
Faced with these challenges, STEM and software experts at Sheffield Hallam University, were able to explore the best approach to creating commoditised software solutions that could be refined by non-experts and distributed securely to Porterdale clients.
What we did
Through the Sheffield Innovation Programme, experts in Computing reviewed Porterdale’s existing tools and helped re-engineer them, ironing out some inconsistencies, and developing prototype software to allow them to be sent to clients securely. The software plug-in was developed to manage and protect the Porterdale forecasting model. Finally, the infra-structure for licencing and distributing the tools was designed, this ensured that a model could not be shared further without Porterdale’s agreement.
The results
As a result of the support received through Sheffield Innovation Programme, Porterdale can now develop their own bespoke models and support packages, encrypt them and distribute them, with only paying customers being permitted to access them. This offers far stronger protection than normal password protection. They can continue to use tried and tested modelling technology, yet also distribute models securely. In addition, their clients benefit from being able to still work with familiar models and do not have to learn a new technology.
Porterdale are now in a position to productise and commercialise some of their tools. The next stage for them is to integrate their software with a licence server. They then can review their models and start promoting them to the diverse set of SMEs in need of forecasting expertise and more.
In addition, the forecasting model is more robust and reliable. Before the work the model consisted of 36,000 used cells (over a 900 x 40 region) with 9,269 individual formulae. After the engagement the same model was formed using only 1,170 individual formulae, and with a few corrections on the way.
There are number of ways this improvement to the model could be assessed, taking some common metrics:
- the size of formulae reduced slightly from 35.5 characters to 29.4
- the number of model inputs dropped dramatically by 97%
- the number of uses of conditionals also dropped by 93%
- the number of functions used dropped by slightly less 75%
Many of these improvements are due to using array formulae that define formulae conditions that apply over large regions of cells, and this perhaps explains the one measure that increased in the new model – the number of terms in formulae rose from 4.2 to 7.8.