Across industries, overstock is a big issue; in the UK alone it amounts to £5bn and often when stock is manufactured and unsold, it either sits in warehouses for years – at a cost to the manufacturer – or at worst is destroyed. Buyfair is a start-up which has identified a solution to this issue and a way to close this gap in the market; allowing manufacturers to auction their overstock in a way which is fair to both themselves and to buyers. It gives a range of buyer types access to opportunities they may not have otherwise had (for example, smaller retailers who wouldn’t be able to buy stock in bulk) and reduces the amount of wasted stock.
We were briefed with creating a straightforward solution to the issue of supplier overstock by building a platform which would allow sellers and buyers to manage the process of auctioning overstock in a fair and entirely self-service way. The platform needed to create a network effect, to attract buyers to the platform and allow for auctions to be extended to a wider network in a way in which allowed for total control by the sellers over which buyers can see their auctions.
The overarching challenge was creating a completely automated system which was easy to use while facilitating what is a complex transactional workflow.
Within this, the first challenge was in condensing the process of uploading an entire inventory in one go to allow sellers to easily set up auctions themselves. As sellers will often have a range of products to submit to auction – with a number of different details per item - the process of setting up a lot could become unwieldy and time-consuming.
The second challenge was managing the payment flow in a fair way for both parties; this meant committing buyers to paying for lots they have won and allowing for a secure transfer of funds to sellers on delivery of the goods.
The third, and biggest challenge, was creating a methodology to manage a one-to-many auction environment. To maximise the efficiency of the online auction system, all the stock in an auction would be allocated to buyers in such a way as would maximise the value of the entire auction for the seller and ideally sell the entire inventory being auctioned. The challenge is allowing for this to happen in a real time auction environment, accommodating fluctuations in bids and quantity requests and ensuring that the right amount of stock is allocated to each buyer to sell the entire amount of stock being auctioned. Allowing for the maximisation of the value of an auction based on real time bidding turned out to not be a simple algorithm.
The platform addresses the issue of setting up auctions quickly by allowing sellers to add their inventory in bulk using a simple automated spreadsheet upload. A range of items, with varying details and multiple product images can be uploaded in one go and can then be attributed to multiple auctions for each type of product.
The payment flow was designed with security in mind, both to keep the funds themselves secure and to ensure commitment to payment on behalf of a buyer whose bid has been accepted. Buyers are required to pre-authorise a card, which is then used to commit funds once a bid is made. The funds are then held in an escrow account until the buyer receives the goods, protecting both parties. All invoicing is automatically created and distributed, with the commission for the sale being automatically transferred to Buyfair’s account.
As part of the UX / UI process, we designed a graphical interface that informs the buyer of current, real time stock allocation at the time when they submit their bid and employs gamification of the bidding process. We developed a unique, Patent Pending algorithm to calculate buyer allocations, always maximising the value of the auction to the seller and updating buyers in real time of the success of their bids, making it clear when they have bid too low to be allocated any stock:
To attract buyers, we created a network which allows buyers and sellers to connect, view each other’s profiles and access associated auctions. Algorithmic principles are used to suggest connections and auctions which buyers may be interested in and allows each seller to invite their current list of buyers onto the platform, giving them exclusive access to private auctions which are not available to all buyers but also enabling them to see stock from other sellers.
‘AVAMAE was asked to provide a software solution to the problem of allocating partial bids for surplus stock in a manner that gave the maximum financial return to the seller. Their solution was inspired involving the use of a hitherto little known area of mathematics (mixed integer linear mathematics). The problem was solved and it led to an international patent application. We were very pleased by AVAMAE’s refreshing and innovative approach.’ - Guy Tritton, IP Barrister, Buyfair
‘Taking an idea from conception through to delivery is something that I have done a few times before across a number of businesses. In order for this to be done successfully, it is crucial to work with the right partners and team. From the start AVAMAE proved that they grasped what was a needed from a technological point of view in a way that left their competitors way behind. It was for this reason that we moved forward with them.
AVAMAE have since delivered an exceptional product and a service to match. We are grateful to them in providing a springboard to our future success.’ – Dominic Speelman, CEO, Buyfair