Archive for August, 2007
According to a report in Computing.
How is this relevant to designers and implementors of e-sourcing software?
In contrast to a primarily internal-focussed system like accounts payable processing, successful e-sourcing is all about putting buyers in touch with new suppliers. Suppliers represent the whole spectrum of technical savvy. If 25% of the population has trouble identifying phishing emails then it follows that a significant proportion of the potential supply base, globally, will find e-sourcing systems too complex and unintuitive.
The moral is clear: make your software simpler and clearer. Always keep making your software simpler and clearer.
It is very easy for software designers (who live and breathe software) to massively misunderstand the comparative lack of technical savvy of the target audience for their software. And also equally easy for buyers of software systems (who are themselves usually pretty tech-friendly) to massively misunderstand the appetite within their organisation for learning a new IT system.
As a slight tangent: This is one reason why I believe that anthropology has a lot to offer software development: after all, anthropology studies how people really behave in real life. [Disclosure: My degree is in anthropology rather than any IT-related (or purchasing-related, or business-related) subject]
Two contrasting views:
- Been there, done that. The vast majority of companies have now done aucions and taken all the margin out. The age of the reverse auction as a revolutionary procurement tool is over. Time now for buyers to focus on other things raher than continuing to beat suppliers on price.
- My god, we’ve barely scratched the surface. The vast majority of companies have either done no auctions, or just one auction and really haven’t exploited the benefits of auctions to lower price and increase quality.
As you might think, my views tend towards the latter
Aberdeen reckons the average auction saving is around the 12% mark. TradingPartners’ average auction savings are nearer the 20% mark. I’m not sure how meaningful one rolled-up average number is, but suffice to say that this number covers categories that have been auctioned twice, three times or more as well as rising markets, falling markets etc. Reasons as to why this should be the case are for another blog but, even taking Aberdeen’s number there is clearly still some margin left in 2007 for buyers to take out.
Admittedly if I didn’t believe this then I would not be working for an auction provider – so feel free to provide me some evidence to the contrary if you think I’m wrong.
I’m currently reading Jared Diamond’s Guns, Germs and Steel, a book that sets out to explain “why history unfolded differently on different continents” – about as “big picture” as you can get. One of his chapters is specifically about the rise of new technologies and is useful background reading for anyone involved in e-sourcing software (or enterprise software in general).
The common view [of new technologies is] expressed in the saying “Necessity is the mother of invention”. That is, inventions supposedly arise when a society has an unfulfilled need …. In fact many or most inventions were developed by people driven by curiosity or by a love of tinkering, in the absence of any initial demand for the product they had in mind. Once a device had been invented, the inventor then had to find an application for it. Only after it had been in use a considerable time did consumers feel that they “needed” it. Still other devices, invented to serve one purpose, eventually found most of their use for other, unanticipated purposes. It may come as a surprise to learn that many of these inventions in search of a use include most of the technological breakthroughs of modern times, ranging from the airplane and automobile, through the internal combustion engine and electric light bulb, to the phonograph and transistor. Thus, invention is often the mother of necessity, raher than vice versa.
A couple of examples:
- Thomas Edison pushed for his phonograph to have serious uses such as recording books for blind people to hear. He gave up on it having a commercial application. But others then started using his phonograph to play music, and this became the main use of the phonograph - a use to which Edison apparently objected for a good 20 years or so.
- Nikolaus Otto built his first gas engine in 1866 at a time when there were more than enough horses and railroads to transport all the goods that needed transporting. It took until 1885 before Gottfried Daimler put one in a bicycle and, voila, a motorbike.
In the e-sourcing or e-procurement space substitute ”inventor” with “Ariba” and “invention” with “e-sourcing system” and you have a picture in which: The e-sourcing software out there today is either massively ahead of its time (like Otto’s gas engine) or may not even be used as intended (like Edison’s phonograph). Either way, the one thing you can say with any confidence is that the real importance of new technology often is not what the inventor thinks it is. For those of you implementing e-sourcing software this means: spend more time talking to your target audience and less time talking to your vendor/implementation consultant.
This happened to me yesterday. I was watcing BBC News 24 over lunch and I saw Ali Percy, apparently a random member of the public, giving her views on proposed new waste policies. I knew her when we were at university together in the mid 90s.First reaction was “Wow, I know her”.
Second was that “And she looks exactly the same now as she did a decade and a half ago”.
Is it something about getting older yourself that you can’t recognise aging in people that you have known?
I recall the implementation of a purchasing module of an ERP system from a past life. The intention had been for individual requisitioners to enter their requisitions directly into the ERP system. But the software was too cumbersome for end-user requisitioners, so they continued filling out their requisitions on paper and the organisation then assigned “super-requisitioners” to enter those requisitions onto the system.
Measuring such a project against “number of users” would show a failure. But change the metric to “number of reqs online” and you might have a success on your hands. Albeit at the cost of the super-requisitioners to do the typing for you. And hope no one asks about why you are using a super-expensive ERP system to store your requisitions when you could have got your super-requisitioners to type them into Excel. Still, you have a good metric of project success that you can put on your resume which is the most important thing, right?
This is an eProcurement example, I know.
eSourcing is very similar. Often companies start off with a plan to roll out extensive adoption of new RFX functionality. The roll-out then stalls for all the usual reasons. Changing the success metrics is usually an easy way of pulling the wool over the business’s eyes and might confuse them enough not to challenge you about the value of enterprise sourcing software if the bulk of the process continues to take place offline.
OK, you’ve invested in a license for a new eSourcing system. You love it, your team loves it and the CPO has given it her blessing. Even the CEO turned out at the launch and came to the pilot reverse auction you did and was very excited by the results.
But 6 months later and you’re struggling to keep momentum going. You’ve only managed one more reverse auction, only a handful of staff are actually using the new system with most still happily using Excel for everything. Where to go from here?
In the simplest of terms, people take to new technology (applies to pretty much everything) either when:
- It is fun and/or makes their lives easier (Nintendo is an example of the first, washing machines are an example of the second).
- It is mandated from above and their job/career/compensation depends upon it (Enterprise Software in general usually falls into this area).
Most eSourcing systems are not really easier for the individual buyer who is perfectly happy with Excel and Email. The people who benefit most from eSourcing systems tend to be managers who achieve better visibility and control.
Therefore the top-down approach is the one generally taken. The trouble with the top-down approach is that it requires dedicated attention from the top to continue ramming the new way of doing things down people’s throats until they give up resisting. Or Change Management as it’s sometimes called. Most organisations drastically underestimate the need for this.
What can you do about this?
- Review your implementation plan. Your implementation plan needs to go well beyond the initial customisation, training and pilot stage, through to how it will be rolled out across the organisation. Ensure functional heads are bought in as early as possible, understand the benefits of the system, are bought into the plan and consider it their system, not just yours. This point is often mentioned, but rarely acted upon because it is hard to do. I know it’s hard and time-consuming, but if you don’t do it you will struggle to reach the adoption levels you want.
- Ensure that you genuinely understand the issues that real staff members are facing, as opposed to what their managers claim are the issues. If you can demonstrate your system addressing these specific issues, or even make the software fun to use, then you will need less top-down attention.
- Don’t lose track of your objectives. Too many software implementations forget the original objective behind the system and get bogged down in indefinite debates with functional heads about requirements whose cost, in a sober analysis, would outweigh their benefits.
- Be prepared to fail. If those resisters are resisting your software then maybe they have a point. They are probably smart guys, after all. Sometimes, taking their objections seriously and adjusting your approach to adoption and roll-out (whilst continually ensuring you have agreement on the objectives in 3) can help in the long run.
Some more cynical comments on the subject tomorrow …
So claims a report by Open Europe just out and available here.
It’s a pretty hefty report – but here’s a quote for those with an interest in auctions:
[T]he ETS in phase one was not a real market – instead of auctioning off permits to pollute, member states allocated them free of charge to companies based on how many the government believed they needed. This created severe distortions. Large companies which lobbied for more permits than they needed were able to sell them on at a profit. Other institutions – particularly smaller institutions like hospital trusts – proved less effective at lobbying. They got too few permits and therefore had to pay into the system.
In other words, you would have got a much better result from the scheme if it had harnessed the power of a (well-designed) auction to create the market.
Whether buying or selling, (well-designed) auctions are the best way to allocate goods and services.
Just been reading another article in the current issue of The Economist about Procter & Gamble’s approach to product development, entitled “Will she, won’t she?”.
[M]any people at P&G do not use the word “consumer”. Nor might they ask if a “customer” or “shopper” would buy a putative product. They are more like to ask: “Would ‘she’ buy it?” … Women have long accounted for four fifths of P&G’s customers.
Staff from [P&G's] Consumer and Market Knowledge division tour the world and spend entire days with women to observe how they shop, clean, eat, apply their make-up or put nappies on their babies. They try to understand how a woman reacts in the first three to seven seconds after she sees an item in a shop (the “First Moment of Truth”, in P&G speak) and when she tries it at home (the “Second Moment of Truth”).
Some insights for those building enterprise software and esourcing software systems:
1. Be clear on who you are targeting. Certainly avoid generic terms like ”user”. A particular challenge for enterprise software developers is that the person buying the system (could well be a Purchasing Director or CFO, for example) is often far removed from the people who actually get saddled with the software. Which one is the real target? Probably both. Most enterprise systems I’ve seen are great for one or the other, but struggle at being good for both.
2. Once you’ve worked out who you are targeting get to know how they really operate during The First and Second Moments of Truth, i.e. When they see the demo (they first moment of truth) and When they pilot it for the first time (the second moment of truth). You won’t achieve this by reading analyst reports and case studies. You’ll only achieve this by being on the ground with people as they work with the software.
In a time of sliding stock prices, VMWare’s stellar IPO stands out:
for the story and
for some more details
Good news for technology companies with viable business models.
The Economist on 11 August has a feature on the uptake of roundabouts in the USA.
Apparently roundabouts are safer (less misjudged left-hand turns, or right-hand turns in the UK), cheaper (no traffic lights), more fuel efficient (cars only need to stop if there is traffic coming onto the roundabout).
What has this to do with a blog on e-auctions and enterprise software? Anyone who has attempted to implemented ERP software or eSourcing software will recognise the issues knows that getting users to adopt new technology can be a very long, hard, drawn-out process. It’s heartening to remember that exactly the same change management issues can occur in all kinds of areas – not just software. According to The Economist:
In Deepest Washington state, trouper Dusty Pierpont stands in front of a roundabout trying to persuade motorists to like them. Washington started building roundabouts in 1997. By 2001 there were 17; now there are over 100, according to Brian Walsh of the state’s transport department. But trooper Pierpont is still needed to soothe those first-time (or even tenth-time) nerves.