How would the allocation look if we used revenues as the allocation base? Is this a good choice?

The situation

It was Thursday 2 January 2020 and DJ Ramanathan was worried. All his attempts to better understand the cost relations in his firm had been in vain all of his most current bids had bounced. His prospect clients claim that they get a better deal from DJ's competitors. DJ believes that one important reason to miss out is that they are insufficiently aware of the nature of their own costs. DJ is the commercial director of Brains Retail Intelligence and Consulting Ltd. (BRIC), a company specialized to (re-)develop retail networks. Their services include: strategy, marketing, organization, operations, technology, transformation, digital, advanced analytics, and sustainability. Over the years BRIC developed functional expertise and has found its way into how to best create conditions to optimize the sum of the parts of a retail network, not just the individual pieces. BRIC for instance was able to design spectacular cost savings programs and put into work systems to standardize customers experience in retail shops belonging to the same network.DJ is astonished by how little knowledge he and other decision makers have of BRIC's own cost structure while at the same time BRIC's consultants do successfully advise its clients on how to efficiently structure its operations and to develop costing systems that provide actionable insights into its clients costs structure. Since the consultants filed several complaints about the cost allocations imposed on their projects DJ had considered to ask the firm's own consultants to advise him on the firm's cost structure. However, instead he decided to first arrange a meeting with his financial controller Aishwarya Rai who said that she is able to shed light on the cost structure of BRIC. Aishwarya identifies the following service units: Finance (50,000/month); Performance improvement (80,000/month), Agile (10,000/month), Strategy (40,000/month) and Operations and ICT advice (90,000/month). BRIC produces five different service lines (S1, S2, S3, S4, S5). To perform their activities BRIC's consultants operate from within a service units. For each service line the firm creates a so-called engagement team who prepare and help implement a plan for the client firms. This engagement team is led by the engagement leader who prepares the bids based on the cost relations that are assumed to exist for each separate service line. Typically the client firms are retail chains that seek the advice of the BRIC consultants. All service units deliver their services primarily to each other and in conjunction they deliver the five basic service lines of BRIC. In the context of the firm each service unit is treated as a separate cost pool.

Current system

Aishwarya believes that a major issue resides in the complexity of the production structure the five product lines entail. It seems that the complexity cause the cost structure to become intractable. That is, the work performed in the 5 service units are interrelated to such an extent that it seems impossible to disentangle the different contributions to the five service lines. At the moment the firm has information on resource consumption for: Finance (50,000/month); Performance improvement (80,000/month), and Operations and ICT advice (90,000/month). Finance is assumed to deliver equal levels of activities to each of the 5 services, i.e. 20 percent to each). Performance improvement is assumed to deliver 50 percent of its activities to S1 and the other 50 percent to S3. Operations and ICT advice is assumed to deliver 60 percent to S1, 20 percent to S2, 5 percent to S3, 5 percent to S4 and 10 percent to S5. BRIC decided to pool Strategy (k40) in one department with finance (k50) and Agile (k10), and to keep Operations/ICT advice (k90), and Improvement (k80) as separate departments. The cost pool that combines three activities is assumed to follow the consumption pattern of finance.

The meeting

Aishwarya decides that the current allocation method is hard to defend and arranges a meeting with the best four cost consultants BRIC has. On 2 March 2020 Carine Lacor, Christine Shih, Isabelle Auger and Peter Cardinal join the "allocation meeting." They thank Aishwarya as they have been worried for years about the methods applied to allocate costs. Carine starts the discussion and argues that it would be too expensive to exactly determine the resource consumption patterns for all activities. However, according to the current method the five resources are almost completely randomly allocated to three cost pools. This method turns the likelihood of creating an accurate allocation basis to almost zero. "Bids we make based on this cost allocation principle begin to look a lot like a Russian Roulette played in reverse." Carine asserted. She therefore argues to first select the three biggest resources and then allocate the other resources to these biggest cost pools randomly. The firm can subsequently make a more accurate estimation of the consumption for these largest cost pools over the different services. Isabelle disagrees. In her opinion it is better to randomly create cost pools altogether. Peter argues that it is better to create just one cost pool and to determine an exact estimate of the resource consumption pattern of the biggest cost pool that is included in the system. Aishwarya disagrees with everyone. She advises that the current revenues are taken as a proxy for the demand for services and that the revenues are used as an allocation base. That is, all costs are collected in one big cost pool and the costs are subsequently allocated based on the revenue. The idea is that the activities that can bear the highest costs are allocated the highest cost level. Christine applauds the idea put forward by Aishwarya but she adds that the current revenue is biased as the revenue is determined by the current demand for the different services, and this demand is affected by the current allocation method. She, therefore, advises to take the potential revenue distribution that would come into being if every bid was won. She further advises to first subtract the variable costs of the three services and then allocate the costs accumulated in the cost pool.

Experiment

Aishwarya made also a bold proposal to her colleagues. This was based on an idea put forward by Christine. She had recently met a student named Ajanee Shed at the Amsterdam Business School. Ajanee had developed an algorithm that helped her ascertain the consumption patterns in activity pools. Ajanee offered on a no-cure-no-pay basis to develop an algorithm for BRIC. Among other data she collected data on how often the firm had meetings with clients on the services that the different departments delivered. More meetings would suggest more resource consumption. Her estimates resulted in the following resource consumption pattern estimates. From Performance improvement, 50% of the activities are used for S1 and 50% for S3. From Agile, 60% of the costs are used for S1, and the rest is equally distributed among 4 remaining services. From Strategy, 50% of the activities are used for S1, 10% for S2, 5% for S3, 30% for S4, and 5% for S5. From Operations and ICT advice, 60% of the costs are used for S1, 20% for S2, 5% for S3, 5% for S4, and 10% for S5. And for Finance, costs are equally distributed to all the services. Aishwarya and Christine were the only ones to believe that Ajanee could be right. The other colleagues expressed their skepticism as far as algorithms were concerned and insisted that all possible calculation were to be done in order for them to select the best method.