Onsite service. Automated order distribution

April, 2011
dia$par distributes orders among your service personnel according to their competences, how busy they are, and their own wishes. All the orders will be snapped up voluntarily in a matter of minutes, and all human administrators will be excluded from the process.

Business processes of any enterprise, which provides an on-site service, are more or less similarly organized.
It does not matter whether it is a delivery from an Internet-shop, DHL delivery service or home installation of washing machines.

Starring: a contact-centre receiving orders from the customers in all possible ways, a dispatching group that distributes orders, a warehouse to give away the orders, deliveries or spare parts, a cash-desk or its analogue where the cash from sales is checked to and where the orders fulfilled are closed. For simplicity sake  we will consider a single case, which is, nevertheless, typical for the model of an on-site service. Particularly, it is so called "computer assistance" that is a repair of computers and office equipment on-site.

So, the call-centre receives a call, the request is issued and forwarded to the dispatchers. It is important that the operator of the call-centre receiving a call promises that a service-engineer (SE) will arrive within 2 hours maximum.
From the description of the order distribution (below) it is clear that this process is slow, and in some cases it takes dozens of minutes. It is useless to hold the customer on during all this time (moreover, the customer will not do it himself), but to mumble something like "Thanks you for your call, we will find a proper engineer and within half an hour call you back and arrange the time of his visit" is also not an option — the customer will simply refuse and will call the competitors, which won’t take long as there really are plenty of them.

The dispatchers, according to their opinion, choose a proper service-engineer, call him and pass the order. It is quite typical when the service-engineer is out of reach by phone due to some personal or objective reasons, or he does not want to accept this particular order (it is late, the location is not convenient to get to, today is Friday evening, etc), so several calls are usually to be made to distribute one order.

Note: This case is a business-model of an urgent service — customers of such computer aid services, as a rule, are not willing to wait for a service-engineer longer than two hours. If well-organized an ordinary retail Internet-shop typically delivers the next day and the deliveries routing is quite different, since there is a opportunity to collect the requests for the day and in the evening/ at night/ early in the morning  distribute them at once topouj couriers/drivers with respect to the territorial sectors as well as mass and size of the load. However, it is a subject for a separate case.

Let us get back to the requests distribution among SEs. It can result in any of the following:

  • ideal- when a suitable service engineer is found and expected to come on time or be insignificantly late.
  • a service engineer is found, the order is distributed, but it is by no means possible to come on time within the two hours promised. It is necessary to call the customer back and to re-arrange the time. The case scenarios are clear: the customer may refuse for the time does not suit him or whatsoever — the range of problems is endless. After that, there goes the reward chain of adventures on re-arranging the order with the service engineer. Nevertheless, no positive result (the order is carried out and the customer is happy) is guaranteed. — For some reasons it is impossible to find a service engineer. It is necessary to call back and refuse the order and apologize. No comments.
  • for some reasons it is impossible to find a service engineer. It is necessary to call back and refuse the order and apologize. No comments.

It is already clear from the extremely tiring description that the last thing, which the business working under the above model can die of, is the excess of efficiency.
Endless number of calls and calls back, permanent human factor: "Tom did not say", "Harry forgot". Due to a huge amount of problems with re-arrangements, it is simpler for the dispatcher to assign, in a semi-forced way, the order to the service engineer who isn’t keen to take it or is just incompetent. Match the time — that’s the thing.

Needless to say how this practice influences the quality of works and the satisfaction of the customers. It also influences the profitability of the business due to the excessiveness of the staff and the abundance of non-productive transactions (and here we reach the famous muda). It is sad as well.

Well, the order is somehow distributed. The service engineer gets to the customer. Let us assume that the technology of the orders distribution worked well and the service-engineer is competent in the characteristics of the customer’s problems. Bingo!?
Not necessarily.
Despite each service engineer has an emergency set of tools and the most frequently used spare parts, it can happen (and it regularly does) than the one not included in the emergency set is required. What’s next? Back to the office to get it if in stock. Dear Reader keep the record of the events.

The needed part is received, the engineer comes back and fixes the problem. Next step? Next order? No way!
​The SE is to go to the office to deliver the money. Otherwise he won’t get another order — it is kinda peculiarity of the Russian market for poorly-paid labor and high staff turnover.
Do keep the record.

As a result, if we look at this ultimate woe in terms of the process approach and the formation of the added value chain, the only stages from the list of oddities mentioned and the things which really create the value are: a) receiving a call b) a service-engineer’s efforts to fix the problem. The END.

The business-process efficiency averaged with regard to the package of orders will be equal to about 3%. Similarly to a steamtrain in the second part of the XIX century. Mind we haven’t considered the difficult cases when it’s impossible to repair the item on-site and thus it has to be taken to the office and, naturally, brought back afterwards.


Dear Friends, the picture above describes how business processes worked in "RukiIzPlech" BEFORE. One might easily guess that it was before transit to dia$par accompanied by equally inevitable and wholesome re-engineering of the company’s business–processes carried out by Humanless business analysts.

Well now the "after-picture".
When receiving a call, a dia$par call-center operator fills in the request form with important details of the order in the program, presses the button "Save" and forgets. Fire& Forget.
The order goes into the automated distribution system.

An Engineer's Settings

dia$par matches the client’s demand and the competences of all available SEs and sends out info texts with all the details of the order. The first to text back gets the order. Before the order is taken, any addressee is able to listen to the conversation recording between the client and operator. The proud receiver of the order can also do it before the order is closed. They just need to dial the service number on their cell phone and enter the order number.

If an order is not taken during the fixed period (calculated according to the preset business logic dependent on time of day, day of week, number of SEs available, etc.), dia$par sends SMS 2. If a services engineer takes an order through SMS 2, he will get an extra percent of the order. So, a sort of the inverse logic of Dutch auctions is implemented — if an order is not immediately taken, the conditions are not favorable enough, so you can raise the rate.

Any number of such cycles may pass, but in reality if services engineers refuse to take the order through SMS 2, it is sent to the dispatcher responsible to solve the problem.

Addendum, 13.09.2013. SMS-dispatching of orders has given way to a special dia$par app for Android smartphones of services engineers.

This picture is Now, and this one is Before: dispatchers used to have to handle 100% of orders, 30-40% out of those passed through their hands twice or more. Now they only need to deal with 5-8%, a repeated processing is exceptional.

This means that services engineers take up around 95% of job offers on their own — quickly, too, or someone faster will scramble first to the food bowl. You can imagine what this change did for quality and customer satisfaction.

Moving on.
​Suppose that a services engineer needs a spare part not included in the portable emergency kit. He can go to any of Moscow’s numerous computer stores and buy what he needs, WITHOUT having to come to the office.

A webpage of a services engineer

The job is done, the money is in. The engineer completes the order on his own through the web interface of dia$par (from the customer’s computer or from the mobile version of the web interface on his own smartphone). The independently bought parts are counted up and added to the order on the spot. There is NO need to come to the office and submit paperwork for the purchase.

Order Page

Now all that remains is to hand over the money. The services engineer comes to any office of Alfa Bank, puts the money in an ATM, sends an SMS with the receipt’s numbers to a service number. dia$par will recognize the payment and, running in the background, put the money in the engineer’s account. NO need to stop by the office to unload the cash.


What does it all give in the end?
Productivity of service engineers, per number of orders a day or per customer, grows by a factor of 2.5. Engineers turnover drops to one-third of the former figure — obvious and hidden costs of hiring and training green workers are greatly reduced, overall competency of engineers skyrockets. Customers make 8 times fewer complaints about workers being late or taking too long to fix the problem. The number of dispatchers drops from 5 to 1, a last man standing.

The call center now receives practically no off-purpose calls (when customers call to yell where-are-the-technicians-you-promised). The staff’s morale is much higher, it is nice to work in a cutting-edge enterprise, and it is nice to hear a customer’s "thank you."

Being inside dia$par. Some stories
© 2020
U.S. Humanless Enterprises
contact us
U.S. customers:
Call me back
humanless service
We do NOT collect any visitor's
information, except given obviously,
and do NOT share it with third parties.