Push or Pull? : The ongoing debate
Every sales manager in a consumer goods company faces a moment of truth in the last week of every month – on one hand he has to meet his sales targets but on the other hand the stocks available with him is not exactly what the market wants (either in terms of quantity or range). Faced with this conflict, he uses his “influence” with the distributor to sell volumes more than the immediate requirement. In return, the distributor gets some additional discount/scheme or extra credit period as a sweetener to “deal” with problem of excess stock. This way of managing sales is typically termed as “push sales”. The push sales syndrome is best seen in a skewed sales pattern across the month – the last week is usually more than 50% of the month’s sales. The week beginning next month, the sales dip again as the dealer does not have free cash to buy more. The skew is also reinforced by monthly batches of production, which leads to arrival of range of SKUs towards the second half of the month. The phenomenon of skewed sales is acute and wide spread in consumer goods industry. (The entire logistics industry supplying trucks to consumer goods industry face an overload of requirements in last week of every month.) Many of them tried the trick of converting the month end measure into weekly measures to break the skew. However the “monthly” production planning method comes in the way of breaking the skew as the SKU set arrival gets skewed. But with dealer off take also being skewed, the push pattern is a self-reinforcing cycle. The sales managers know the problems of push based supply chain. They also intuitively understand that if they stop the push strategy and supply only as per pure consumption (or only immediate requirement) they will have a much consistent sales pattern across time periods. Most of them know the advantages of smoothening the sales on receivables and capacity planning.
The know-how of how to implement pull is readily available and is also advocated by many consultants. However most players in consumer goods industry are still operating on the “push mode”. One of prime reasons why most companies do not switch to some form of pull distribution is the fear of sale loss. If they implement a pull system and supply what is immediately required, the fear is that the released working capital will be used by the distributor/retailer to buy more of competition products and hence the sales of company implementing the pull system will be jeopardized. This fear of losing immediate sales is holding back organizations to switch from “push” to “pull” mode of sales.
The Fear: Unfounded?
The Theory of Constraints approach towards implementation of pull distribution combines pull supply chain implementation with a win-win market offer, which has made the implementation of pull systems risk free in the transition phase.
If on one hand the supplier is helping distributors with frequent replenishments only as per consumption (pure pull), the distributor uses the released capital to stock more variety. In many environments the distributors only deal with small sub-set of SKUs (out of the total company portfolio). They restrict the range to reduce risks to capital stuck in non-moving SKUs.
When the supplier company supplies small quantities of new range while releasing capital from the excessively stocked items, a win-win partnership is created and pull system becomes a reality. This has to be coupled with two other enabling supply chain paradigms:
- A production system has to move away from producing large batches as per monthly forecast to production based on consumption from central warehouse.
- The central warehouse should move away from being a flow-through warehouse (or a trans-shipment point) to being an aggregator of inventory. This implies that the inventory at central warehouse is higher than that with the distributors.
The implementation of pull system provides staggering results both to the supplier and the distributor. TOC pull systems have been successfully implemented in India in many companies in varying environments (auto spare parts, garments, steel, fashion shoes, FMCG). In each case the results were staggering (around 30 to 40% jump in sales in period of implementation) particularly when the implementation was done along with a Win-Win offer with distributors. The ability to use the released capital to extend the range provided the practical approach for companies to implement replenishment with the distributors. The fear of capital flight to competition is prevented while at the same time; the ROI of distributors also improves due to frequent rotation of inventory.
In some environments the risks of the distributors in new range is further alleviated by an option of “return back” if the inventory of new range does not move beyond a specific time. Such a radical approach helps distributors in trying out new range in the market thus increasing the chances of new range hitting the shelve space. This sounds almost like a silver bullet.
The Silver Bullet: Not so Silver?
In first stage of most implementations, the sales growth is beyond the expectations of management. So the first phase of implementing pull with the distributors gets stabilized sooner than later. However the real power of this pull solution lies at the retail point primarily because at the retail point the sales fluctuations (and forecast accuracy) is worse as compared to that at a comparatively aggregated levels like distributor or a further aggregated level like the central warehouse of the company. At the same time, the direct reach of most companies in India at a retail level is very limited. When the distributors struggle with problems of working capital, they tend to limit the exposure to retail outlets. As a result the “real reach” (where retailers are serviced on a regular basis) is very limited. This also indirectly shows that the real power of a pull system lies in retail – sales potential is many times higher than what is experienced at the distributor level.
However the implementation of any pull system requires an overhead of frequent communication of requirement. The essence of pull system is aligning distribution to daily sales. Hence getting the daily sales data is key towards transition. To pull a system, In case of distributors, it is easier to get this information as most of them being large entities, have computer systems and the daily business transactions are usually with an IT system. With minimal IT intervention, one can get the IT systems of distributor to transfer daily data to supplier company’s IT system. This is crucial as we are moving to a system of daily communication of requirements as opposed to once or twice a month ordering by distributor (frequency at an SKU level).
However at the retail level, particularly in India, the level of use of IT is extremely low. Most of the retail outlets are mom and pop stores (The modern retail is relatively a small portion of the total retail population). The retail market is spread out and number of the outlets run into few millions. When retail outlets are spread out, wholesalers also dominate the channel instead of distributors. (As compared to the distributors, the wholesalers do not usually have a territory allegiance and sell high volumes to highest bidder). The “open market” conditions at the retail level coupled with difficulty in setting up the “daily-sales-data-collection” infrastructure has dissuaded many companies from trying pull based implementation at the retail level.
When companies stop at implementing pull systems only between them and the distributors, the pressure of increasing sales targets hits them hard. The first phase of implementation of pull systems with distributor does give a significant sales jump and takes away the pressure of increasing targets. The “honeymoon” period of the first phase continues till all large distributors are on-board on the replenishment program. With more sales coming in, the targets also grow. The problem starts when one wants the next jump – the next jump has to come from implementation in retail but if it is not done and targets are high, invariably companies slowly start creeping back to the push mode. You would surely know some companies who have claimed to implement pull systems, for the first 3 weeks it is a pure pull and in the last week, the situation deteriorates to push mode (as they fulfill their targets!). Availability at central warehouse also deteriorates in the first week and soon distributors start having higher inventory per SKU and the ROI drops to old levels. They go back to the mode of restricting range and also restricting reach.
Companies starting on the journey of pull systems actually do not have an option to stop the implementation at the distributor level only. They have to go to the next level of opportunity and enjoy the higher level of sales. They have to get away from the “pressure of meeting monthly targets” and hence the need to push. The biggest compromise to any operational discipline comes from pressure of meeting targets. The only way one can avoid the pressure is to ensure there is no scarcity of sales. So, companies starting on the journey of a pull system have to find a way to implement the pull systems in the vast environment of mom and pop retail outlets else it is better to not even start on the path of the replenishment program.
The Good Enough Approach
IT systems can help us get daily data but having IT systems for a large population of mom and pop stores is practically infeasible. So daily sales data (daily ordering for an SKU) is infeasible but the current environment is far from the best approach of daily ordering at a SKU level. Even when retailers are actively buying from wholesalers, the maximum frequency is once a month per SKU (exceptions are fast perishable items like bread or milk). In many cases the buying frequency (at SKU level) is much more than a month. Getting daily data is difficult but there is huge gap between current practice (monthly) to desired level (daily). We can make a huge improvement by adopting good enough instead of the best. A weekly frequency of ordering is a huge improvement. The weekly frequency or once in a fortnight can be implemented with help of sales man visiting retail outlets to collect order. This weekly sales man can be on the pay-roll of the distributor as he has earned enough from the replenishment program to take the next jump.
The word “sales man” here is a misnomer. The person is expected to visit a retail outlet at a pre-determined frequency and define requirement based on the difference between stock levels at retailer and the norm. While one may argue that the job can be left to the retailer. But in most mom and pop stores, the inventory management and ordering is usually very ad-hoc. So left to the retailer, he might tend to order more quantities per SKU than required. The job of the “sales man” also called the DSR (Dealer sales representative) is to ensure that he does not collect orders more than the norm. In a way he is not a typical sales man who should be happy getting more quantities.
The focus on the DSR is two fold:
- Ensure ordering does not go beyond norm and
- Visit frequency is adhered to without any compromise (so that the retailer experiences the frequent service to change his behavior).
Allowing the distributor/wholesaler sales man, an outsider, to check stocks and create orders requires a very high level of buy-in. The retailer has obvious advantages of increased ROI for the category of business. But since each category of products in his portfolio occupies a small percentage of his overall business, he has a poor intuition of ROI at category level. So in this case, each retailer has to be explained how his buying decisions impact his ROI at a category level. Such buy-in sessions have to be customized at a retailer level with data of the specific retailer. This involves a huge effort and the people trying to convince the retailers have to be competent in explaining the concept to the retailer. The efforts are huge but if well assisted by templates and a training program, we can convert an art to a process, which can be rolled out to the masses.
The above approach also has another effect, when we get a wholesaler to service a retail outlet with a predefined frequency; we have to limit the geography of retailers for the wholesaler. This means that the wholesaler gets indirectly converted to a distributor. This goes a long way in avoiding territory conflicts and price wars across territory.
A case of extension of pull solution unto the retail point.
Fleetguard started the journey of implementing pull replenishment right from suppliers to production shop unto the distributors. Within one year, it completed the first phase of implementation and got significant benefits. Then it decided to get into the most difficult phase of implementing at the retail points – mom and pop stores spread across the country.
Conceptually the sales managers understood the potential gain. They fanned out across the length and breadth of the country communicating to more and more retailers. As a first step, Fleetguard’s sales managers just convinced retailers to order frequently with the conviction that the dealer sales man will come at a defined frequency. In the first phase, FG did not get into the process of checking stocks and comparing with norms to generate the retail orders.
The retailers liked what they heard and more and more retailers agreed to work with the Fleetguard distributors. The company’s reach, as measured by the retail outlets that stock goods bought from the distributors increased from about 10000 outlets to over 27000 outlets over a period of three years. Though there was corresponding sales increase resulting from an increase in the number of outlets, it was clear that the essence of replenishment was not in place. Retailers were still ordering based on their judgment and tended to carry large stocks of a few products thus blocking their capital from more SKUs. The moot question is why does the retailer not see the obvious benefits that the company is trying so hard to give them?
The answer probably is a combination of two thoughts in the mind of the retailer:
One that any company accounts for at the most 5% of the retailer’s business, so any improvement there is akin to be noise in the mind of the retailer. Two, that perception of unavailability in the mind of the retailer was not as severe as in the mind of the company. They tend to compensate the lack of availability of an SKU from one company with the availability of a similar SKU from another company thus believing that at the retail outlet level, the range and availability is not so bad as it is made it to be.
The above two facts coupled together led the retailer to believe that his judgment based ordering was the best way of doing business and any improvements would be minuscule with the efforts of monitoring stocks and sales to order as per consumption being too much to justify.
The Fleetguard team decided to take this issue head on and disprove the hypothesis of each retailer based on actual data of each retailer. Having access to over one year’s weekly purchase data of each retailer, the Fleetguard team constructed a customized analysis for each retailer and demonstrated that there were some products where the retailer was carrying more than four to six weeks of stocks of some products (for some as high as 16 weeks) when the supply frequency was only one week!
In addition based on a one time stock take, they also demonstrated to the retailer that there were some loss. And then they nailed the issue by demonstrating that retailer was not out of stock on any product that he was buying from Fleetguard owing to the excellent availability and weekly visit of the DSR. So if the retailer spread his capital better by reducing stocks of existing SKUs and using the money to buy more products from the Fleetguard portfolio, he stands to make more money.
The approach called for new challenges in sales processes – since FG is dealing with large mass of retailers. It had to have processes, trained manpower to convince retailer with the new way of working where orders are generated frequently based on stock and limited to a norm and capital redeployed to a bigger range. Fleetguard has one of fastest growing reach in the auto industry. At the same time it is able to convince retailers with the new way of working. Almost always, the retailer concludes with one thought: if this is the situation for the products of Fleetguard – a company that is working so well, what would be the situation of other companies which are still operating in the old paradigm?
As Fleetguard gets more retailers on board and gets their buy-in on new way of working, they should reach a stage where a mass of retailers will try to change the way they operate and force dealers, across different brands of auto parts to supply as per consumption and who in turn should force the auto spare parts supplier to move to the pull system. Maybe one day, the actions of Fleetguard will change the way entire auto spare parts industry operates.
The efforts of Fleetguard managers are unique in the entire industry. They are perhaps the only company in the world where the sales men do not have sales targets – neither primary nor secondary. They have targets to ensure the enabling actions to implement pull in the vast market of India is done at the fastest possible time. These efforts (over the last three years) have resulted in their ability to reach over 35000 retailers every week. These retailers are now increasing the range of the company’s products to nearly double of what they held earlier. The combined effect of increasing reach and range is that their sales growth is far outstripping the rate at which the market is growing. However, they cannot rest on their laurels as the numbers of auto retailers in the country is more than 1,00,000 and the range that the company has is much larger than an average retailer’s holding.
They have started a journey on a path where no one has gone before. Their approach is best verbalized by following lines of the famous poem by Robert Frost – the road not taken.
“I shall be telling this with a sigh Somewhere ages and ages hence: Two roads diverged in a wood, and I took the one less traveled by, and that has made all the difference.”
Build to buffer
One challenge in the fashion industry is that forecasts often are wrong, especially at the individual item and retailer levels. Additionally, only about 15 percent of new designs turn out to be winners, selling out in about six to eight weeks.Read more