In the first generation of outsourcing contracts this term was a bit of an oxymoron. If it was mentioned at all in was purely as a side note in the contract. Neither party knew how to, nor planned on delivering this aim, but the lack of achieving it is still mentioned as a contributing factor in the dissatisfaction of many executives within the client. Things are getting slightly better in some of the more forward thinking relationships but there is much still to do before both parties can truly deliver on this aim.

Some of the problems are in the pricing structures in any contract. Innovation costs money it is often expected that the vendor will foot the bill. What is even worse is that there are clauses that state that any savings in innovation will be shared between the client and the vendor. So, not only will the vendor pay for it, but they will share any rewards. Innovation in inherently risky so if the vendor has to fund it and there is a chance it won't work and their margins will be further reduced for no compensation. As you can see this is not attractive for the vendor and so most didn't actually put any effort into innovation. If any happened it was more as a consequence of changes in the operating norm during the lifetime of the contract. These could be technology or infrastructure improvements brought about by a requirement to upgrade to the latest version of a particular piece of software.

To actually make innovation work, there are a few options. Assign a certain proportion of the cost of the project to innovation. This innovation budget will then be spent in agreement with both parties towards the benefit of the project as a whole. The vendor will actually have money to work with and the client will have a sense of ownership of any work which is undertaken. It could be the vendor's responsibility to identify and propose areas of innovation investment as they are closer to the day to day running of the project. An alternate benefit is to have the vendor pay for it, but they see all the rewards. In other words, any cost savings or efficiencies as a consequence of the innovation will not reduce the costs to the client.

There is an added benefit to innovation which is often overlooked and that is staff retention in the vendor. Attrition in vendors is often a massive issue, especially in tier 1 markets like India. If staff have an opportunity through innovation to learn new skills or participate in innovative but possibly risky projects then their interest level will be retained and there is more chance of them staying for longer. This is of course a benefit to both the client in that the people delivering the project are retained and the level of service maintained and a benefit to the vendor as they have less cost in continuous training for new members of staff.

Both of these options listed above have to actually make it into the contract. As long as the contracts only mention innovation without actually explaining how it is to be delivered on executives will continue to be disappointed with the results.

Steven Arbogast has written a seminal article about Process Towers and how they can spoil the Process Improvement you can effect in any part of an end-to-end business process.

Steven uses an excellent example of a waiter taking your order at a restaurant with many variations, and special requests you may ask for, after looking at a menu. They usually have very little time to take down all these variations on a little slip of paper that they pass on to the kitchen. Between your order and the eating, many things can go wrong in the handoffs between the waiter, the kitchen, and the one or more chefs that may prepare what you ask for!

End-to-end processes may have many process towers in between, that can make a process go horribly wrong. In the end, it does not matter to the end customer, who among the process towers got things wrong about your order! Handoffs between processes are very sensitive and may need as much watching, as the process itself!

Process measurement is where these process towers can make sure that these handoffs are handled properly. Upstream Metrics and Downstream Metrics may need as much watching as the Process Metrics themselves. Each process may need to make sure that the handoffs from the previous process tower is done according to expectations - For an order management process to work properly, the previous order taking process tower may need to make sure that they handoff things properly. This is done by the Upstream Metrics. Once the order management process tower is done, handoffs to the production process tower may need to be handled properly. These can be ensured by the Downstream Metrics.

Process Towers are scary. Handoffs between them could wipe out any gains you may effect in any one single process tower. However, with proper management of metrics, Upstream and Downstream, from a proper process tower can ensure that handoffs are managed properly.

There’s many a slip twixt cup and lip - Proverb

One of the most critical processes during an outsource project is vendor selection. There are numerous companies that will take at least part of that responsibility away from you but ultimately it is you signing the contract. When deciding on a vendor here are some points to keep in mind.

Firstly, what are your requirements? If you don’t know what you want then it is impossible to find the right vendor for you. Whatever your requirements, however niche, there will someone out there that fits. It is a bit like the world of online dating. You are the prize catch and every outsource company wants a piece of you, or your money at least. Being specific, and I mean really specific, will weed out the majority of the potential unsuitable suitors. Vendor companies will claim they are an expert at everything as they are trying to win your business. Being specific will reduce the number to manageable proportions.

What is your risk profile? By that I mean, how adverse to risk are you when trying to outsource this function. If you can determine that, then this will futher reduce the number of potential vendors you have to wade through. As an example, a company approached me recently wanting to outsource printing of documentation for their customers. This is not a core function of the company and, although key, they could tolerate quite a high risk profile with their vendor. This meant that we could advise vendors in countries that we would not normally have suggested. The client was happy to sign with one of the suggestions, and they have set up a successful relationship, that has its ups and down but delivers to their expectations and saves them quite a bit of money. Going for a tier 1 vendor in a tier 1 market doesn’t always make sense.

Use a predefined score card and be consistent. It is easy to be dazzled by an excellent vendor sales team and not follow through with due diligence and select based on criteria not personality, although personality might be one of your elements in your score card.

Chose the right size of vendor for your company. As stated above, always going for a tier 1 vendor doesn’t mean you will get the best service. If your outsource agreement is $300M you will nearly always get a good service, but if you are considerably smaller, chose an outsource vendor that will value your business and tailor their relationship to you rather than tell you how you should run the relationship.

Verify the references. Make sure you are not speaking to the brother in law of the vendor companies sales manager. It is surprising how often this sort of thing happens so you should be aware and ask the right questions along with getting confirmation from other people within the reference company.

Perform a site visit. Plan it in, but just the threat of a site visit can change the information that a vendor provides. If they claim to have 3000 employees, and you are planning a site visit you may find that this, among other things is a fabrication. Be sure to set an agenda for the visit that provides you with enough information to aid in the selection process. A visit that consists of a tour plus a few ad hoc meetings may not provide you with enough opportunity to discover the truth. The vendor will usually set the agenda so don’t be afraid to question and amend anything they have proposed.

Take it in stages. Work on a long list of a about 10 then narrow it down to a short list of about 3. Ensure you have RFI documents for all 10 and use these to move to RFP. Once you have this level of detail you can start planning how to whittle it down to the final vendor.

Try a pilot. If you have a few vendors, and they are all looking like potential candidates you might want to think about running a pilot project with each of them. This can also be done with the chosen vendor as it will shake out any significant issues and provide insights into their day to day running. There will of course a be a separate contract, usually much smaller, for the pilot and once completed, it is easier to progress to full contract signing with greater confidence.

Lastly, don’t be afraid to ask for help. There are a number of companies who can help you through this process. Try not to let them take over as you will have to deal with the vendor in the long term so get involved. Any search in Google can throw up plenty of companies who specialise in this area, although personal recommendations are always preferable.

The Futility of Call Center Coaching is a very interesting and thought-provoking article on how spending more money on coaching individual agents in a Call Center or on other business process participants is essentially a waste of money!

The author says that spending more money in improving the entire system, adding money to improve systems and call flows (or process flows) for the entire set of agents is actually money better spent than on coaching individual agents.

The author starts out quoting Edward Deming and his famous experiments that proved that perturbing a system introduces more variation and does not help in achieving greater quality. The same thing happens when one of the Call Center agents gets trained individually.

Improving the entire system with expenditures on better systems, automation of call flows (in case of Contact Center processes) or processes help improve the end result, for example, Customer Satisfaction Scores (CSAT).

The author also does a mathematical simulation of the above situation and proves his point. In practice, it is quite impossible to avoid individual coaching of agents, given employee turnover and new employees joining. The author makes a point that even in organizations with very modest turnover, the situation is still the same! So no point in wasting time on training individual agents if you want to improve end results!

The above study also makes a subtle case for applying lean techniques to business processes first. These improve the flows at the process level and improves things for all agents, eliminating complete steps in some cases.

Elimination of waste is always a better idea than training agents on skills that are spent on wasteful activities,anyway.

A system is a network of interdependent components that work together to try to accomplish the aim of the system. A system must have an aim. Without an aim, there is no system. ...A system must be managed. The secret is cooperation between components toward the aim of the organization. We cannot afford the destructive effect of competition." - W. Edwards Demindg, The New Economics

In a recent article from Computer Weekly it discussed the recent trend of clients demanding large discounts from providers, sometimes up to 24%. This is counter productive as large discounts will only harm the relationship in the long term. Negotiating a ‘win, win’ scenario means that everyone is happy in a relationship. If your vendor is actually loosing money throughout the lifespan of the agreement they will attempt to re-coup losses. This may mean a reduction of quality in the form of staffing levels or technology improvements. No company acts as a charity for another. Thinking you have screwed the vendor in the negotiation only means you loose in the long term.