Futrli unveils Predict hybrid forecasting appby
Following an epic development effort, Futrli has released Futrli Predict, a sat nav-like hybrid forecasting app that blends business advice into financial forecasting software.
Futrli Predict has finally landed in the forecasting software market on a mission to redefine the traditional user experience and remove the time spent preparing forecasts.
Futrli founder and CEO Hannah Dawson likened Futrli Predict to a forecasting “sat nav”. Using custom algorithms, the app combines from data outstanding invoices and bills with historical profit and loss and balance sheet trends to create a holistic picture of short, medium and long term performance.
“It’s been a long time in the making – too long,” Hannah Dawson told AccountingWEB. “Predict is really quite special. Well-known tools like Facebook Profit use machine learning algorithms that require lots of data, much of it irregular, to spit out accurate predictions.”
During the R&D phase, Futrli focused on the complexity and diversity of small businesses. “They often have very irregular data, which means that we needed to create our own algorithms to negate this. The really nerdy piece of the puzzle is the hybrid forecasting,” said Dawson.
“Most forecasting software looks at the cashflow side of things, but you can’t run a business on gross cashflow. You need to know you know your actual sales trends and expense trends, what you’re actually spending on VAT and so on.”
Hybrid forecasting combines both forecasting approaches to create an early warning system that removes complexity from the forecasting experience. “Hybrid forecasting helps guide businesses on their day-to-day decisions and doesn’t require any extra work on the accountant’s or client’s part,” explained Dawson.
Predict can test forecasting decisions within the system and users can customise forecasts and predictions to increase the level of accuracy.
“In today’s Covid-ravaged word, this is completely essential with all the loan scheme payments, knowing when you have to pay lenders back or knowing what will happen if we have a second wave,” she added.
So how does it work?
The hybrid app looks at each individual transaction in cloud-based ledgers and categorises them into different types. The app categorises the records by type of account, such as rent or wage, and applies a set of custom algorithms for each account. Then it splits each account’s data into invoices and bills, spending, receipts and other transactions like journal entries.
Customer and supplier payments are also aggregated “to make the payments side of things live and give a complete view of performance and cashflow,” added Dawson.
For each dataset, Predict looks for trends such as seasonality and Covid impact, which are factored into the forecast. If the software detects any pandemic-related effects within any data, it will alert users.
“The hybrid approach means we’ve combined everything forecasting software does and everything an advisor does into one product. If the prediction is off, you can toggle off and create your own which makes customisation potential vast.”
The VAT calculation module is extremely accurate, Dawson emphasised, and includes relevant options like VAT deferment and the ability to override and customise the data.
“Predict is the only forecasting app that actually calculates VAT accurately from actual and projected data,” she claimed.
Transparency and AI
For Dawson, data transparency was the last piece of the hybrid forecasting puzzle and market awareness is the only impediment holding back adoption of AI-based tools like Futrli Predict.
“I genuinely don’t think there is any barrier to adoption. Forecasting has lots of negative connotations because previously it was time-consuming or funding event-driven. We’re not using AI. We’ve created our own algorithms because AI tools in the market can’t explain where the numbers have come from, which is really, really important.
“To establish trust, you can click on every single prediction and every single cell in Predict and it will tell you where the data has come from.”