In 1969 Sysco was a $116m regional wholesale food distributor operating in a highly fragmented segment. Today the company is the leader in its segment, with $20bn in revenues. Sysco has posted 18% annual average revenue growth in a segment with only 3% end-market growth by accomplishing all aspects of a three-pronged strategy—it gained high local market share, reduced operating expenses and increased gross margin.

The components of Sysco's strategy, when taken individually, certainly are not unique. However, their combination, and the company's tactics, have driven Sysco's remarkable success.

Sysco aggressively acquired companies to gain high local market share in attractive regions, built a huge private label business to increase gross margin and maintained P&L responsibility at the distribution centre level to secure tight inventory and operating expense management. The company posted a five-year total shareholder return of 252%, more than twice the S&P 500. Even more impressive, Sysco's shareholder return is almost 4.3 times greater than the food distribution segment average.

Ah, but here is the real question for investors: given the immense size and variety of the wholesale industry, how does one find a Sysco at the start of such a meteoric rise? And isn't the Internet gradually "disintermediating" wholesalers anyway?

Not quite. In fact, the Internet debacle confirmed that the industry is here to stay and that it is a critical part of the national supply chain. Wholesalers still account for $4trn in sales and one in every 20 US jobs. In addition, more than $10bn in wholesale distribution deals closed last year, ranging from acquisitions to minority equity stakes in sectors as diverse as restaurant supplies and speciality steel.

Unfortunately, most of these investments will generate poor returns. Over the past five years, only 20% of public wholesale distributors in a Bain survey sample managed to beat the S&P 500. As important, a majority actually destroyed shareholder value.

So should investors shun wholesale distribution altogether? That is a viable investment strategy. However, that strategy would be one that bypasses enormous possibilities for growth, given the 1000% difference in shareholder returns between the best and worst distribution performers in our sample. That tells us it is possible, albeit difficult, to find attractive investments.

We know that differences in strategy drive differential returns for competitors within a segment, be it steel, pharmaceuticals or otherwise. Therefore the nature of the segment itself must also play a key role in a company's success. If this is true, how can investors discern which segments are undervalued as opposed to fundamentally disadvantaged?

Contrary to popular belief, the answer lies not in differences in segment margin or cashflow, but rather in the shape of a segment's supply chain and the nature of the supplier/distributor relationship. So margins and cashflow don't directly explain a segment's shareholder returns? Not in wholesale distribution.

A new set of analytical tools was developed to determine which segments are fundamentally more attractive than others, and to predict investment returns across distribution segments. These metrics were designed specifically to provide investors with insight into which companies in which segments will realise the fastest income growth and highest shareholder returns (our measures of success).

The first metric requires some imagination. Picture three shapes: an hourglass, a slide (like one you rode down as a child) and an inverted funnel. These are the three typical shapes of a supply chain, or what we have termed "segment shapes".

Take the drug segment, which has an hourglass shape because markedly more drug suppliers exist than distributors. In contrast, the steel distribution segment is shaped like an inverted funnel, as few steel producers exist relative to steel distribution companies.

Finally, the grocery distribution segment is considered a slide, with a similar number of suppliers and distributors. How is this metric quantified? Simply by dividing the number of suppliers by the number of distributors, which provides a result we've termed a "leverage ratio". Of course, different leverage ratios coincide with different segment shapes. For example, a leverage ratio greater than 1.1x has an hourglass shape, while a leverage ratio below 0.9x has the shape of an inverted funnel; the slide is any leverage ratio in between these two. Surprisingly, this visual gauge is critical to understanding the relative attractiveness of distribution segments, but it is only half of the story.

The second and final metric quantifies the nature of the distributor/supplier relationship. Specifically, it is the percentage of shipments within a segment that are sold directly from the supplier to the end-customer, or those bypassing the distributor. Thinking about the drug distribution segment again, we estimate that only 43% of total segment shipments bypass drug distributors. In contrast, an estimated 80% of shipments in the office supplies segment go direct.

So the drug distribution sector is in great shape, and the new indicators show it. One of the sector's stars is Syncor. The $630m pharmaceutical distributor has managed to generate a 950% five-year return to shareholders, versus 275% for its industry segment. Quite simply, the growth is the result of a symbiotic relationship with DuPont, one of its key suppliers, for the distribution of its heart-imaging agent, Cardiolite.

To win preferential distribution rights to this exciting class of drugs—radio pharmaceuticals, or nuclear medicines—the company had to invest heavily in a sales and distribution infrastructure. Product exclusivity, along with strong drug demand and limited product substitutes, largely explains why Syncor has achieved a gross margin of nearly 36%, which is over five times as great as the segment average, despite being a fraction of the size of the largest drug distributors.

"We play a more important role than that of the traditional distributor," explains Robert Funari, Syncor's CEO. "Our local market presence and unique service model not only facilitate greater product demand, but they are critical when dealing with a product that is very expensive and can degrade in a matter of hours."

To meet this competitive condition, in most parts of the country Syncor can deliver a dosage-ready medicine kit directly to a patient's bedside in under an hour. Moreover, its dosage error rate is near zero, compared with as high as 5% in a hospital and 2% at retail.

While exclusive manufacturer-distributor arrangements of this sort are rare, Syncor did not come to occupy this position through luck. Rather, the company systematically built the capabilities that made it the distributor of choice for a manufacturer looking to quickly build a market for its radio pharmaceutical product. "Relative market share is the quintessential issue," says Syncor's CEO, "since all healthcare is about local market economics.

While not a perfect analogy, it's like being in the ice business in Texas in the summer. That's our business model. The closer you are to where the ice cube goes into the glass, the better." Today Syncor's pharmacies process 93% of all nuclear medicine procedures in the country.

So why are our metrics the critical measures, as opposed to the more traditional financial ratios? Unlike financial statement-based ratios, the leverage ratio measures the value a distributor plays as the aggregator of goods. Intuitively, this value should be high in an hourglass-shaped segment—customers can call one distributor instead of multiple suppliers for their product needs. Conversely, in a segment shaped like an inverted funnel, distributors not only are forced to compete against one another, but they also typically must undertake additional (capital intensive) investments to convince customers not to order direct.

Regarding the percentage of distributor-direct shipments, it is an excellent proxy for the degree of value added from the customer's perspective. Therefore an inverse correlation should exist between the percentage of distributor-direct shipments and the perceived value of the distribution function within that segment.

We looked across a diverse cross-section of distribution segments, including drug, steel, chemicals, food service, office supply and grocery. The metrics for each segment were used to derive a measure of predictability (regression) that was significant. In fact, segment shape and distributor-direct shipments explain almost all of the difference in segment shareholder returns (R2 = 95%) and income growth (R2 = 89%).

So what segments have historically made the best investments? Those with the highest leverage ratios and lowest percentage of distributor-direct shipments. As Syncor demonstrates, the drug segment is a near-perfect example of a distribution segment success story. With a leverage ratio of almost 24x, and a majority (57%) of shipments flowing through distributors, the average five-year shareholder return for a company in the segment was almost 350%.

The chemicals segment is at the opposite end of the spectrum, with a leverage ratio of only 0.1x and a vast minority of shipments moving through independent distributors. The average company in this segment generated a negative 44% return.

A critical application of this analysis lies in its ability to predict the success of "roll-ups". One resulting message is clear: do not attempt to roll up a segment with an inverted funnel shape. It is simply too hard to consolidate enough companies to bring the leverage ratio above 1x, particularly if suppliers are consolidating as well.

Essentially, most roll-ups to date have been too small to improve a segment's fundamentals. Metals USA highlights the importance of this point. Eight steel distribution companies merged in 1997 to form Metals USA, which immediately went public with the goal of consolidating steel distributors.

However, trying to roll up a segment with a leverage ratio of only 0.1x proved difficult. Today Metals USA has a negative 80% five-year shareholder return. In fairness, we are not saying that investors should avoid all segments with poor leverage ratios and distributor-direct metrics. But we are saying that successful investments in these segments are the exception and not the rule. Therefore finding targets with a unique and defensible strategy is of even greater importance in disadvantaged segments.

Gibraltar Steel provides proof. The steel distributor has been on a roll for the past five years. How does a small player in the steel segment differentiate itself? Quite simply, by investing in new processes and product lines that transform it from a lower-margin steel warehousing company to a higher-margin steel processing company.

At first glance, these tactics seem similar to many other steel companies that seek higher margins for their products. However, Gibraltar realised that its strategy would only be successful if it could perform value-added services at a lower cost than its suppliers or customers.

"We have found product differentiation to be critical, but we also had to perform these value-added types of services in a cost-effective manner," said Brian Lipke, Gibraltar's chairman and CEO. "While it is difficult to differentiate in the steel industry, doing so drives higher margins and also has a diversifying effect, since the finished product from one business segment becomes the raw material for another."

Case in point: over the past five years the company acquired downstream processors and heat-treaters of metal. The former enabled Gibraltar to deliver finished steel products directly to end-users in the construction market, while the latter was a complementary process that enabled its customers to reduce the time and expense associated with heat-treating metal, with no incremental inventory investment.

Impressive results followed. Products sold in finished form increased from 14% to 46% of revenues, which drove a 33% increase in gross margin. Today Gibraltar is one of only two steel distributors in our research study to have generated a positive five-year shareholder return. As important, Gibraltar is thinking about the final piece of the puzzle—high local relative market share—as it continues to make acquisitions.

We know that segments like drug and food service, with hourglass shapes, have rewarded investors. But how should returns be predicted? Again, by using the two key metrics to analyse a segment's migration path—its movement from, say, one segment shape to another.

For example, ten years ago the drug segment had a leverage ratio of 15x and only half of its total shipments moved through drug distributors (then giving a 150% shareholder return). Since this time, the drug segment return has increased with both metrics. Now, investors who understand the direction, and rate, of segment evolution can determine when the timing is best to undertake an investment in a segment.

Essentially, investors should seek segments that are evolving in shape from a slide to an hourglass. The grocery segment presents one such opportunity. Consolidation continues to increase the leverage ratio, but distributor-direct shipments are up slightly as the large retailers have moved distribution in-house over the past decade. The reverse of this trend (retailers shedding their distribution assets) would be an important signal that returns of companies in the segment are likely to increase.

In other words, investors ought to seek out companies with distribution plans like those of Sysco, and avoid those with plans like AmeriServe's. Here's what happened to AmeriServe. In early 1999 the company appeared to be the quintessential success story. In just three years, it had transformed itself from a $400m regional, chain restaurant food-service distributor to one of the nation's largest distributors, with sales of nearly $7.5bn. At the time, AmeriServe was serving more than 36,000 restaurant locations, and its client list included leading chains like Burger King, KFC, Taco Bell and Dairy Queen. Less than a year later, AmeriServe filed for bankruptcy and shocked the food-service distribution industry.

AmeriServe's major problem was believing, incorrectly, that national (not local) market share and scale economies were the keys to distribution success. Pursuing this strategy, AmeriServe made several large acquisitions to gain national presence, with a disastrous effect on profitability.

For example, as AmeriServe grew, its gross margin actually dropped by 10%. Unlike Sysco, which focused on smaller customers, AmeriServe focused on large restaurant chains, which narrowed its customer base to the industry's most competitive segment.

As a result, AmeriServe faced continual pricing pressure. To make matters worse, its operating expense (as a per cent of sales) increased because of its flawed distribution centre strategy. AmeriServe's customers required that it have a national distribution centre. As a result, the distributor built a national presence but did not focus on local market share. This left the company with an underutilised network of distribution centres.

Case in point: AmeriServe's average revenues per distribution centre totalled only $122m, versus $229m for Sysco. The operation of subscale distribution centres put it at a cost disadvantage, particularly relative to competitors with higher local market share. Why? Distribution scale economies exist at the distribution centre level, not at the national level.

In a last-ditch attempt to improve its cost position, AmeriServe reduced its number of distribution centres by more than 50%. However, as the company consolidated its network, many customers fell outside its radius of effective service. Soon after, AmeriServe faced widespread customer dissatisfaction and lost its largest account. The bankruptcy filing wasn't far off.

In every industry, benefits to scale exist (eg purchasing discounts). However, in the distribution industry, local market share is what drives shareholder returns. Growth is important, but not if an investment in a local geography fails to improve that local share. In the end, distribution is and will remain a fundamentally local business. Sysco has proved the point very well.

*Sam Rovit is a director in Bain & Company's Chicago office, and a cofounder of Bain's Worldwide Merger Integration practice. Ken Sweder is a consultant in the Chicago office.

Source: Submitted to the EIU ebusiness forum by Bain & Company.