GoTo Part One • GoTo Part Three • GoTo Part Four •
The Swimming Pool Story
Building a ValueWeb
WITH a ValueWeb
 

 

The Model

 

 

The way that I accomplished this was by applying the relationships and methods that I had created in collaboration with the pool companies and subcontractors. I produced a drawing (plan) of the back yard. On it was the entire design. The price of each phase of work was stated along with the schedule (date and time of day) and a list of materials required and crew and equipment necessary. The client (technically acting as owner-builder) wrote me a set of checks (in advance) for each phase of work. Each was in an envelope. The contractors had learned that when I said “be here, at this time with these materials, equipment and people” that they could rely on a clean site, proper layout and supervision and INSTANT payment. They had no sales cost, engineering time, bidding costs, scheduling complexity, supervision requirements, and collection complexities. All they had to do was divert a crew for an hour (or a few) as specified and immediate contribution to their bottom line was accomplished. Added benefit: training for their crew by Matt Taylor. I was on site each day, for all crew hours, managed my other crews by wireless and then went on to other work when that days sequences were complete.

 

On the way to other projects, I did a couple of pool layouts myself. This kept me in good physical shape, connected to the business and paid for my salary and truck. The money made from the rare custom and fully featured back yard environments “dropped” to the bottom line of our business. In this 7 year process, work that typically was time consuming, frustrating, expensive and risky because simple, rewarding and profitable.

 

It should be noted, that when I started this process - 7 years before the example given - I executed a custom pool and landscape of the same scope and complexity (triangles instead of circles) and it cost $21,000 and took three months - all considered a financial bargain by the owner and a construction miracle by the contractors at the time. This project totally consumed me for four months. This last project employed about 50 hours of my time (about half of which I was able to do other tasks in my portable office on site).

 

THIS IS WHY I SAY THAT 75% OF THE TIME AND HALF THE COST OF ANY CONSTRUCTION PROJECT IS TYPICALLY WASTED - victim of the burden imposed by the organizational structures, the contracts, the build process, design ignorance and owner interference. This is a statement of huge significance. It is also something that virtually no one has wanted to hear in over 30 years.

 

The cost and impact of building can be radically reduced. This makes room for better design, higher quality, more sustainable materials and energy solutions. When we are told that these values cannot be “afforded” this is within a game that imposes upon itself a huge inefficiency tax. This requires, however, that several circumstances be brought together over a period of time: A steady volume of work. Total understanding and rationalization of the building possesses. Trust built up over a long period. Elimination of all possible overhead redundancies. A knowledgeable systems integrator on the site. Strict Rules-of-Engagement totally enforced. Also, appropriate use of technology. In this case, as will be elaborated below, the first “NavCenter” and Motorola wireless in every crew truck in the system. The complete system has to be engineered - not just the build part; this means, the engineering of the pool as a basic product, the sales process, the financing process, the drawing and contract documentation process, the scheduling process, the field supervision process, the field layout process, the build process of each specific trade, the feedback process between pool builder and trades AND individual workers, the pool startup and turnover to owner process, the relationship with the owner process. Over a 7 year period, each of these processes were refined through multiple iterations of work until risk, mistakes and sloppiness of every kind was eliminated.

 

This was an eloquent system of building.

 

A system that can be applied to many different business and production circumstances.

 

When I encountered the swimming pool industry in Arizona in the mid 60’s it was already groping it’s way to a primitive value web model. Over a six year period we were able to develop this impulse into an explicit system of work that today would be called a “value-added, mass-customized-product, just-in-time, lean-construction (production), extensively out-sourced, supply chain system.”

 

Every step from the customer meeting and sale through financing, design, construction, clean-up, first customer us and maintenance was developed into a single system that cut more than 20% off of the product cost and guaranteed delivery of the pool (from sale to swimming) in 10 days at the height of the season.

 

It also resulted in a significant increase in quality - aesthetic and mechanical. This completely refuted the often stated principle that between reducing time, better quality and lower cost you can have, at best, two of the the three.

 

In this system, as many as a thousand pool projects could be tracked at once and the entire network of construction crews coordinated minute-to-minute over a twenty mile radius to accomplish multiple work sequences on each pool in a single day. A fully equipped swimming pool construction requires 23 crew (size of 1 to 6) hours over 9 tasks. When we “practiced” by building a new display pool we would typically start early Saturday morning and be swimming Sunday evening.

 
 

Why swimming pools? Partly it was opportunity. I had designed some custom pools that the builders were having trouble building. When I showed them how easy it was, I was pulled into their industry. Why I stayed with it was threefold. First, the swimming pool as a piece of construction technology is interesting. It involves most of the trades in discreet combinations. It is “heavy” enough to require significant equipment and organization. There is a high variety of finishes and and many, many different conditions that generate different design requirements. By nature, the units are spread out, geographically, generating significant logistical problems. Second, they are “production” items in that, despite a great variety of designs, the underlying structure allows for the leveraging of volume components and techniques. Third, the customer is intimately involved but not dogmatic about the design and techniques of building as is, traditionally, the circumstance of house or office building.

 

These are ideal conditions for testing new construction and organizational methods. It should also be noted that by the time I started this work, I had ten years experience in designing, building and developing housing, apartments and high rise structures in 4 states. I had personally performed every task involved: R&D, sales, design, construction layout and management, enterprise management and every trade involved in the production of a pool.

 

I knew every step and sequence. I had already discovered that merely changing the sequencing of a project could yield significant decrease in construction time (motel for Royal construction, 6 months to six weeks, as example). Also, I had a passion. When I entered architecture, I was told that any change in design drove unacceptable construction costs. It was clear to me that the conventional way of building was intrinsically more complex than many of the suggested alternatives. Over the ten year period, I both learned the trade and accomplished significant cost/time reductions in conventional building. I did this in ways that allowed for better design to be accomplished. I was not satisfied as were many architects of my generation to regale builders, throw their hands up in the air and accept high costs, poor construction and long delays to completion as normal. I wanted to know why and I wanted to know HOW to build better. Imagine an artist who knows nothing about paint and “directs” the creation of a work via sketches and descriptions from a distance - that is the condition of an architect in a traditional practice model.

 

This is important to would-be system integrators. DETAILS are significant. Knowledge of the entire process is a requirement. On very large complex challenges - the F-15, as example - this means a sophisticated System Integration Team and appropriate participation of the entire ValueWeb membership. The F-15, involved over 900 units distributed worldwide, operated by several air forces. Each plane has 500,000 parts, 50,000 of which are unique. The system has been maintained and evolved (upgraded) over a 35 year period with the expectation of remaining viable for another 10 to 15 years. Up to an many as 40,000 people, are involved in designing, engineering, building, fielding, maintaining and using the system. These circumstances result in a huge configuration management problem. Complexity of this sort requires GROUP GENIUS.

 

The swimming pool system we put in place was simpler. However, at it’s full expression, about 40 people were involved in the design and maintenance of the process at any given time. Most of these were the workers who built the pools.

 

Because the swimming pool is complex enough to be nontrivial and small enough to support what today we would call rapid-prototying - and because there is clear competition between contractors - this is an ideal environment to demonstrate the efficacy of alternative technical and process techniques. Feedback and innovation was accomplished on a weekly basis.

 

Another advantage is that, in the 60’s, this industry was not made up of sophisticated players. Generally, most builders had learned by doing and were repeating mistakes learned from those who taught them. This was an “industry” ripe for learning. I found out that it is easier to teach uneducated people how to do it well than sophisticated folks who are “educated” with every reason why it cannot be done. In the end, we established a system that had a true economic measure: at the end of the week, everyone’s check told them how well we had done. This was the only time in 44 years of work experience that I have seen a compensation system put in place that actually worked, was fair and encouraged both individual and group performance.

 

The system grew organically. We worked with four competitor pool companies over a 7 year period. One of the companies we worked with, over a two year period, through their entire startup process. We supplied, at various times and in various combinations, the following services: system integration, schedule management, research in components, product design, field supervision, subcontracting. At the peak of the process we employed, ourselves, the entire ValueWeb in the production of our own custom pools. We also manufactured some low volume custom components. Typically, we negotiated a new contract each year and then provided our services exclusively to one company for that period. This was also the pattern for all of the subcontractors in the system. What happened, then, was competition moved from between companies to competition between networks. Networks that resorted themselves every year. Thus, techniques and methods migrated to all of the competing pool companies. Competitive advantage was seen as a yearly thing. Each subcontractor was very aware that they negotiated, annually, for both revenues per unit and for volume. The pool builders and the subs were very conscious, each negotiating season, of trying to attract the BEST TEAM with with to play the season. They also worked, together, with intensive collaboration to reduce, time, waste and cost for everyone in the system - not just for themselves. Competitive advantage meant how many pools could be delivered, in a short season, at what price point to the customer, with what features and for what level of quality. Competition was fierce in season. Sharing was ubiquitous between seasons. It was when this completion exceeded the rules-of-engagement and turned ugly that this nascent ValueWeb was shattered. I will discuss this sad aspect of the story in Part Four. A ValueWeb architecture is extremely vulnerable in today’s world of mixed-economy “Shock Capitalism.”

 

The customer truly gained in the process (which is a rare thing despite all the rhetoric). Construction workers achieved incomes as high as 30 and 40 thousand dollars (in the 1960s) and the pool builders maintained a gross profit of 25% unheard of in this industry. The price of the pool dropped every year, the feature set increased, time to swim declined radically and quality (horrible when we started) soured.

 

At the end, a 20 ft. by 40 ft. pool with automatic cleaning, heating, copious amounts of decking, trimmed out in a first class way sold for $4,000, was delivered in 10 days (sale to swim) and was seamlessly coordinated with other work such as landscaping, outdoor furnishings, building additions and other objects: walls, planters, gates, etc. The technique by which this was accomplished is decribed below (The Model).

 

At the end, I was able to build totally custom pools at the same per-unit cost, using the same contractors in the same time frame. Imagine a back yard with a pool made up of four intersecting circles, of different levels, a fountain rising up from the intersection of three of them, raised, cantilevered circular decks of concrete and laminated (4x6) redwood, circular trellises offering shade, extensive landscaping and lighting, custom enclosure walls (of complex shape) and ornamental steel inserts and gates - all build in 10 DAYS, in season for a complete price of $7,500 (1970 money). This was the peak of my capability as a designer builder). I have never matched it since - the closest being the Cambridge facility nearly 30 years later.

 

 

(insert multi-circle pool design)

 

The swimming pool industry, in the 60s, in Phoenix, was characterized by frequent bankruptcies of financially unstable small contractors, sloppy work, a bland product, time delays, high costs and a myth-based selling process.

 

Any attempt at customization from the standard rectangle or kidney drove the costs through the roof. Coordination with the many other suppliers and contractors necessary to complete a back yard environment was almost totally nonexistent. Cooperation was almost totally nonexistent - and, were there was cooperation it was mostly corrupt with back door payments, finders fees and locked in deals all to the buyer’s detriment. Owners were lied to, manipulated and ignored once a contract was signed. Quality was unbelievably bad. Often the pool was built in an easement leading to law suites and expensive fixes. There was little discipline in the system and the actual structural integrity of the units were often compromised by improper placement of materials. No one was making much money. Typically, owners who signed their contacts at the beginning of the warm season got to swim in the fall. From end to end it was a sleazy process dominated by suede shoe salesman - charisma and high pressure substituted for engineering performance and quality.

 

It was not that people did not want to do better -they did not know HOW. It was a circumstance driven by ignorance. The BELIEF was that this was intrinsic and things could not be approved unless unbelievable amounts of money were spent.

As example, when my first custom pool design - which had 16 corners, three levels and masses of custom tile was bid for ten thousand dollars, I questioned the builder. He said that it would take days to lay it out and form it, that the lines could never be maintained after excavation therefore the design would look terrible and the no one could do the tile work the way I had it designed (coming out of the pool and over the bond beam). I pointed out the the actual materials of the pool were the same as a typical 20 x 40 and he agreed but sighted outrageous labor costs. I asked him what would be the costs if I did the layout, formed the pool, established both grades and lines with offsets so the various trade could re-string the lines at will when building, and, if I took care of the time myself - what would be the cost? He gave me a cost 10% over a standard pool (the 10%, he admitted, was for “security.” “Done” I said. He asked me when I was going to lay it out and set batten boards. I told me the date. He said he would drop by “sometime” to see the process. I told him he had better be there between 8 and 10 am because I would be long gone afterward. He refused to believe it but said he would be there at 8 to watch my crew do the “impossible.” Needless to say owner, pool builder with his entire supervisory staff showed up at 8 and were surprised to see the “crew” called Matt Taylor. In one and a half hours the the job was done and I had the layout contract for 2,000 pools. My sojurn had begun.

They simply did not know how to layout a swimming pool accurately. Two to three inches deviation was considered normal - straight lines impossible (see Layout). Similarly, when it was time to do the decking (which did not follow the pool but “wove” in and out of it and the landscape), I started an hour ahead of the crew pouring and finished by the time they got the first section in (see: pouring concrete). Easy. It is a METHOD or it is a disaster - no middle ground in things like this.

 

They wanted to know how all of this was achieved with no dimensions on the drawing and I had to explain what a module was.

 

As for the tile, I went to the tile store, bought the tile (about three times as much as for a normal pool and custom (a total of 300 extra dollars). The pool builder sent his time contractor over, he looked at the pool, saw what I wanted and started to leave the job. Fortunately, his truck was blocked by mine. An more fortunately, I had a couple of six packs of beer in my truck. And, lucky for me, I knew how to lay tile (see: American Pool Building). We had a great time, got really drunk and produced a tile effect never seen before. The tile man became my best advocate, worked out a fair price and became the tile king of the NEW fabulous way to add accent and color to your pool. Mass customization had begun! He made a fortune of it and within 3 years refused to do “those puny bathtub” installations that once were the standard.

 

The shape of this pool was not the way it was for only esthetic reasons - although the result was spectacular. The yard was narrow. Even a small kidney-shaped pool could not fit in. That is why I was hired in the first place. The conventional would not work. I designed a “zoned” pool where the basic functions of diving, swimming laps, playing and sitting could all be done simultaneously without interfering with one another (each has flat bottoms at the best debth required for the function.. This intent drove the shaping (which provided great variety for design play and problem solving) and also allow me to fit the pool into the narrow yard with out removing all the trees (only one had to go) - another owner requirement.

 

In one stroke a new pool type was born. It became the “Tri-Level” Pool, dozens, maybe several hundred, were built and one pool salesman made a career of doing only this kind of pool by copying my custom pools and adding his own versions. He got quite good at it. He also leaned when it was important to bring the work to me (about one out of ten as it turned out).

 

There are several discreet lessens from this part of the story:

 

Find out exactly what the problem is. The “trade” person knows what s/hehas been taught. If not an innovative personality, s/hewill most likely not challenge these (hidden) assumptions. This does not mean that there is not skill, talent, nor the desire to do better. The PRIDE is usually there beneath the surface beat down by years of UpSideDown Economics. Ignorance (not a pejorative term) is ruling the situation.

Don't teach. EXTRACT the critical part from the system that runs by unforgiving protocols with attending negative economic consequences. DO the new procedure on a contractual, arms length, Input/Output of the existing system. Take the downside risks away from those in the system. Plug your result back into the system with minimum interference.

Put the new solution into the system (transfer) when people want it.

Let everyone celebrate in the success. The tile man; the salesman adapted the methods. The Pool builder ran full page ads within 10 days of finish. The perception of the possible shifted. The new became the standard. First, as an option, ultimately as a main line offering.

Understand the specific technologies and skills required to make the new process work - and work economically. The old way may seem stupid but it lives in a complex web of interactions. The change can radiate out in a myriad of unknown ways and negatively impact a bunch of people. KNOW these connections. Understand the end-to-end consequences. Midigate negative results.

Know when to innovate to what degree. This was not my first pool. It was the fifth. I wanted to do the Tri-Level design from the beginning but waited until I knew enough of the industry (in detail) and a client’s requirements demanded something other than the standard schema. This presented the window for innovation. My first pools were innovative designs in many ways, but the bottom profile was a standard slope.

 
 

This one pool started a process that changed an industry. There were several innovations for the time and place: Cantilevered decking over the pool edge (no coping - a new way of forming). Flat multilevel pool bottoms (structural consequences, steel tying changes, impacts on the automatic cleaning systems and water circulation). Multi-sided shape of the pool and “syncopated” decking (layout and forming “complexities”). Tile instead of decking over part of the pool (bond beam) perimeter (requiring exact bond beam geometry and workmanship; replacing large tile pieces with mosaic tile - not a standard at the time). Integration of the pool with the house and landscape (issues of grading, water run off, leaves in the pool, access from the house, safety concerns).

 

Each of these potentail impacts had to be addressed.

 

It turned out that there were only a (real) few blocks between a mundane design and an exciting and far more functional one. Those blocks could be removed, by removing them from the traditional process, demonstrating how easy it was to do the work and then transferring the process back into the common protocol. An architect that never gets out of the drafting room cannot do this. When THINK work is separated from DO work all kinds of unnecessary barriers result and cannot be resolved. I was able to do this because we were a design office AND a contracting firm. We were also willing, when necessary, to do R&D by absorbing these R&D costs and guaranteeing to the owner we would replace any thing he did not like (or did not work) after he had USED it awhile.

 

 

(insert Tri-Level Pool plan)

 

The organizational transformation process is accomplished on the level of the organization’s products - not management. Extract, do, demonstrate, transfer, explain the principles at work - in that order.

 

The MG Taylor Enterprise Application

 

From 1979 to 1982, Taylor Associates (now MG Taylor) created the basis for the work it is doing today. The Weak Signal Research and philosophical base had been laid down 1974 to 1979. In the first three years of it’s existence, MG Taylor built it’s first environment, developed the basis of the Modeling Language, created and refined the DesignShop process, fielded an RDS experiment and operated the first client NavCenter (Mentor Corporation). We formed a corporation, created a Board of Directors, received investment and sold the company. The beginnings of a ValueWeb was established. We serviced a small, national client base of medium sized organizations.

 

From 1982 to 1985, the Acacia years, the information factory concept was refined. We concentrated on one client and it’s customer base enjoying, for the first time, steady revenues and a budget sufficient to staff beyond what was just required to produce client work. What is now AI stayed in Boulder as a network member organization and was not reintegrated until the early 1990s. It was in this period that we gained the operational excellence required to deliver our service products at volume and profitably. On an event basis, we have been profitable ever since. In terms of an organization, overhead to operating profit, we have been profitable, break-even and at a loss, annually, in approximately equal shares. In total, we have accumulated a loss of about three million (out of 42 million revenues). With a total capital contribution of about 500 thousand dollars this means the entire cost of creating our products, service and the four business units that deliver them has been less that four million dollars. In 1999, we had a valuation of 10 million.

 

The sale of the business to Acacia involved some breaking up of the organization but most of it stayed intact. The piece that “spun out” later came back in. When the enterprise as a whole left Acacia this was the first time it was somewhat shattered. Most KnowledgeWorkers, the archives and physical facilities stayed inside Acacia. The principle gain of the Acacia years, besides financial stability and the refinement of operations, was the opportunity to work with a moderately large organization though a protracted change process.

 

1985 to 1990 were the ad-hoc years. The enterprise functioned as a network, had no fixed assets, did not market and worked predominately inside large organizations such as Capital Holding and GM. We used the Acacia Center (that we had designed for them) as a place to do infrequent events. We later took over this Center and operated it for a couple of years. The Capital Holding relationship exemplified our pattern of work that lasted until 1995: we did a DesignShop process focused on some specific problem (in this case, a merger), build them an environment (in this case two, one in 1997 and a larger facility in 1992) and facilitated their process through a multi-year transformation process (1987-1994). there were usually three of us as overhead and a network of about 15 to 20 KnowledgeWorker that performed work on a contract basis.

 

It was 1996 to 2001 that MG Taylor laid the foundations of a “standing” business. In this brief period, MG Taylor (and now, knOwhere, Yolke, AI, iterations) has gone through several “classical” phases of development, de-evolution and recreation. This was highlighted by the establishment of the knOwhere Stores (Hilton Head 1996, Cambridge 1997, Palo Alto 1997), the relationship with EY (1995 - 1998 and the sale of the Cambridge facility to them), the integration of AI (1996) and the establishment of our own production facility (1998), the peak of our employee base (over 30 in 1998, 20 in 2001 across all Business Units), the creation of separate Business Units (1999) and revenues fluctuating between 5.5 and 9.9 million per year. Through out this period our structure was a hybrid of traditional corporate structure, extreme outsourcing to the network, extensive partnering with a larger company and extended virtuality. This has been a high performance organization in product/service delivery and unstable in traditional management.

 

It should be noted, however, that we have been profitable - and not - in each of the eras (other than 1979 to 1982) and that there are simple correlations to be found, among the many variables, that drove these results.

 

Chief among them, of course is the ratio of fixed overhead to net revenues. This is determine by volume and our ability to scale up and down quickly enough to keep our ratios in line. This issue is made complex because our product/service delivery shifts rapidly in both location and in what our customer/clients buy. This last factor is caused partially our lack of critical mass but more from the nature of our market. For example in 1995, our revenue was mostly generated by DesignShop events delivered by RDS deployments to several clients all over the US. In 1996, the majority came from one client (EY) delivered primarily in the Hilton Head knOwhere Store. In 1997 and 98 mostly DesignShop events from one client (EY and their clients) delivered in our three knOwhere Stores, two EY environments and one client NavCenter. In 1999 and 2000, a number of different services including three NavCenter transfers to a diverse client base delivered through the Palo Alto knOwhere store and from within the client NavCenters in several states. During this period, direct AI WorkFurniture sales became one third of the business involving large environment sales one year to a few clients and many small environment sales the next to many clients all over the world.

 

In this same period, we entertained three serious merger/purchase options and two large joint ventures. While some of this is the consequence of being a small business in start up most of it is the consequence of a highly volatile market. A true ValueWeb architecture will mitigate some of the impacts caused by this volatility. This will move the impacts away from the operating business level to the web-enterprise level were traditional market volatility is better dealt with.

 

The Swimming Pool Model evolved to deal with certain circumstances: seasonal fluctuations, economic cycles, wide geographic distribution, time-demanding buyers, a design-focused product (mass customization), a new product (low cost consumer pools), fast growth, untrained workers, quality issues, brand competition and the integration of multiple products/services from multiple organizations. These circumstances match up nicely with the circumstances of the MG Taylor Corporation and affiliate corporations. These circumstances are widely shared among industries and organizations of all kinds all over the world.

 

Matt Taylor
Palo Alto
March 8, 2001

 

SolutionBox voice of this document:
INSIGHT • POLICY • PROGRAM

 
click on graphic for explanation of SolutionBox

 

posted March 8, 2001

revised February 18, 2009
• 20010308.451948.mt • 20010313.293455.mt •
• 20010324.452876.mt • 20010330.208844.mt •
• 20010331.987723.mt • 20020217.298700.mt •
• 20090218.232310.mt •

 
Copyright© Matt Taylor 2001, 2002, 2009
 
ipt src="http://www.google-analytics.com/urchi
Search For:
Match:  Any word All words Exact phrase
Sound-alike matching
Dated:
From: ,
To: ,
Within: 
Show:   results   summaries
Sort by: