feature article
Subscribe Now

The Changing Face of Distribution

From Box Shifters to Full Outsourcing

Once again, a news story has appeared in time to provide a topical twist to a story that was already underway. At the beginning of January, distribution giant Arrow Electronics announced that it had bought einfochips – a 1500-person “design and managed services” company. To quote the release, einfochips’ activities include, “developing custom hardware and software and new business models for the Internet of Things (IoT)”. And this announcement neatly encapsulates how distribution companies are changing.

Distribution has been a part of the electronics industry, one could argue, since before there was a recognisable electronics industry. For example, giant Avnet was founded in in 1921 by one Charles Avnet, who sold parts for people to build their own radios; Arrow Electronics came on the scene in 1935, and Electrocomponents was founded in 1937 as Radio Spares. But it was in the 1970s that they really began to play a role. Semiconductor manufacturers were geared to think in volume, and they preferred to deal with the handful of customers who wanted to buy in volume. At the same time, there were many customers who wanted only relatively small numbers of a particular device and didn’t want to negotiate deals with a number of different companies to get all the components that they needed for a product. Distributors filled that gap. Given the potential number of sales that they could generate, they were able to negotiate good deals with manufactures and hold stocks to meet immediate needs. (The volatile pricing, particularly of memory products, in the 80s and 90s was often a source of debate between manufacturers and distributors, and it still presents some challenges.)

At first, the distributors were content to accept contracts that banned their stocking of competing products, while manufacturers were free to appoint multiple distributors in a particular market. They continued to work pretty much as they had in their previous retail roles. They typically had limited marketing budgets and often spent much of it on catalogues. (Forests were felled to produce what became massive doorstops.) They typically had technical people as their sales force and gradually built up their technical support with application engineers.

These were pre-Internet days, so normal orders went by post, while urgent orders were sent by Fax or Telex (look them up on Google if you are lucky enough to be too young to know what these are/were), and fast turnaround was measured in days, not hours. By the 80s the large number of relatively small and relatively local distributors began to merge or be taken over, creating multi-national operations. The merging companies would often have competing products, and, at the same time, semiconductor companies began to both increase their product line-up and also to consolidate, so it became impractical to stock only non-competitive products

Alongside the new silicon components specialists, the older companies, while retaining retail outlets, began to build silicon portfolios. They were also moving into IT, providing both equipment and support services, which provided a healthy cash flow. Organic growth in silicon has continued, and so have the mergers.

Today, distribution is a complex field. Obviously, the Internet has replaced massive printed catalogues, and e-commerce is a key technology. Distributors boast about the breadth and depth of their stock. (Although there is disagreement about how to count the numbers. Mouser argues that some distributors’ count is based on their own part numbering, which assigns separate part numbers to the same product in tubes, on reels, in trays, or whatever. But the numbers are always a point of dispute.)

With components, life is relatively simple. Most sites allow comparison of different products and provide data sheets and app notes. But, a few years ago, they recognised that people want a component only because they are building a system, and so they began to offer design tools for PCB layout. Obviously, you would choose your components from that distributor’s offerings, and the design tools would have detailed knowledge of each component, not just its physical footprint and pin-outs, but, in the more advanced versions, they would also have an understanding of their behaviour, allowing modelling of the system, and letting the user play “what if?” games, changing devices to optimise performance or price. The distributor may also supply technical support, of varying levels. The end result would be a PCB design and a Bill of Materials (BoM). But again, what the designer really wants is a product, not a BoM or a Gerber file, so some distributors now offer a logistics and manufacturing service, using selected board manufacturers and board stuffing companies and taking responsibility for all the activities needed to build and deliver the completed product to a warehouse, or to multiple warehouses. In some cases, they may even add your new product to their stock.

The design service assumes certain skills on the part of the customer, which, particularly with the growth of the IoT and the explosion of “Things”, may not exist. So, alongside the design services has grown a mass of reference designs. These may come from device manufactures who create reference designs as an extension to application notes, from the distributors’ own development teams, or, in some cases, from users who contribute through user communities hosted on the distributor’s web site. There is an increasing number of development boards and kits for the purpose of supporting the reference designs. While these have always been a part of the device manufacturer’s portfolio, the rise of third party boards, such as Arduino, Beagle Bone, and, outstandingly, Raspberry Pi, has made a major transformation, with some of these boards, particularly Raspberry Pi, being used for even quite high-volume production products.

Following on from this, it is an obvious route for distributors to build their own implementations of their reference designs, leaving market differentiation to the users’ own software.

And then, as we started with, distributors are increasingly offering design services, whether by building their own teams, or, like Arrow, by acquiring an entire company. Couple this with the manufacturing and related logistic management and this can be an attractive outsourcing route. If you are carrying out your own manufacturing, then perhaps you would like a distributor to manage your stock room for you, automatically maintaining stocks of components, etc., at the levels you need for efficient running.

If you are overstocked or have surplus inventory, some distributors will take care of that for you, buying the surplus or arranging for it to be bought.

What else can distributors do for you? Well, for example, around a third of Arrow’s turnover last year was from enterprise computing services, so they can handle your IT requirements. And most of them have a great deal more than components and boards on their web sites. Power supplies, enclosures, and test and measurement instruments, for example. Hand tools (even soldering irons) are all there, and some, like Germany-based Conrad and RS in the UK, offer retail outlets with a great deal more.

Distributor websites are also increasingly carrying “content” – that is, technical material that goes well beyond the app notes that they have long hosted. Written by professional journalists, often through third party agencies, some have in-house teams – these are often interesting – but (as I am bound to say) don’t replace the objectivity of real journalism, such as what EEJournal provides.

Where are we going next? Last year, the semiconductor market grew significantly. Future Horizon’s Malcolm Penn believes that it passed the $400 billion mark, and he is bullish on 2018, arguing for 20% growth to reach $500 billion. This will be achieved partly though increased unit sales and partly through inflated average selling process, as fab capacity fails to keep pace with demand. Distributors will be cutting a good slice of that. They are also planning to get more out of the value-added services, such as design and support services. There is a feeling that consolidation, both of distribution (in 2016 Avnet swallowed Premier Farnell/element 14) and of semiconductor manufacturers (The EU has approved Qualcomm’s purchase of NXP, which only a year or so ago swallowed Freescale, and is now in the sights of Broadcom) will continue. Distribution will remain a highly competitive business. The IoT is seen as a significant area of growth with some thinking that the maker community is also a growth engine.

All in all, the changing face of distribution is set to continue to change.

Ed note: I have had input, directly and indirectly, from a number of people in preparing this article. Many thanks to them and their views

Leave a Reply

featured blogs
Dec 12, 2018
For years chip designers have dealt with ECO'€™s when their source code was written in RTL. But the move to high-level synthesis (HLS) means that their source code is now one step further removed from... [[ Click on the title to access the full blog on the Cadence Communit...
Dec 12, 2018
A joint demonstration between Samtec and eSilicon — an eSilicon 7 nm 56 Gbps DSP SerDes over a Samtec 5 meter ExaMAX® backplane cable assembly — caught a lot of attention at SC18. The demo showed a true long-reach capability with a high-performance, flexible, eas...
Dec 10, 2018
With Apple'€™s '€œWearable'€ category of sales setting a new record this September with growth over 50%, and FitBit seeing growth in both trackers......
Nov 14, 2018
  People of a certain age, who mindfully lived through the early microcomputer revolution during the first half of the 1970s, know about Bill Godbout. He was that guy who sent out crudely photocopied parts catalogs for all kinds of electronic components, sold from a Quon...