The HotLine Magazine Archives
For The Profession of Cloud/SaaS Product Support

Customer Support Management Magazine; Premier Issue — March/April 1997  “Engineering vs. Support: How To Play The Game”

The following article was published over ten years ago.  In the interim, how much has changed?  You be the judge.

In the world of software publishing, Support takes the heat for all those little glitches that Engineering misses. But here’s how managers can ease the development process — and dramatically reduce support call volume.


Not that long ago, software industry support managers like me were told that our only job was to answer the phones and to keep the customers happy. “Don’t bother the Engineers,” we were cautioned, “they’re busy writing the next revision/product.” In those early years, neither the customers/problems nor the support groups were taken seriously; the race to build the companies and to get new technologies and products to market was all. But while the importance of the customers can no longer be denied as the industry has matured, companies still have difficulty understanding the importance of Support in the product development process itself — and in the preservation of the bottom line.

“When I took over as Vice President of Support and Service, there was a wall between Support and Engineering” reflects Matt Doyle, AST’s VP of Support & Service. “Occasionally, a door in the wall would open, but it was only occasionally. That just wasn’t enough.” Similar comments have been voiced in virtually every high technology company in the industry, and in many organizations, there still are serious complaints about the chasm between the two departments. When asked to describe the current relationship between their group and Engineering/Development, most support managers and executives quickly reply that they’re somewhere on the scale from “we’re not really talking” to “open warfare.”

Even within more enlightened companies, few support managers are satisfied with their standing. “We’ve come a long way, but there’s still a lot more to do,” observes David Aune, Claris’ Senior Director of Support . Bill Strauss, Intuit’s VP of Support agrees. “We’re not fully where we want to be. We’ve moved towards a new model for support, but it’s not fully implemented yet.”

Surprisingly, at the less advanced end of the scale, inside companies that are seemingly locked into an outdated support models, managers have discovered that change is possible despite substantial odds. The key requires an understanding of industry realities, together with an appropriate repositioning of the role of support and its data resources within the company. It isn’t enough to just build the better mousetrap, you still have to market it in order to be successful.

Support and the Competitive Edge

In the early days of the software industry, a severe inter-departmental conflict between Support and Engineering was common to nearly all software manufacturers. “To begin with, there was an implicit assumption on the part of Engineering that once the programming was complete, their job was done. You didn’t have to listen to customers when what you were doing was so ground-breaking, so new and insanely great that people just couldn’t wait to get their hands on it,” Colleen Byrum, Director of Customer Service for Amazon.Com, recalls. “But that was a symptom of an immature industry. Where it still exists, (and it still does, unfortunately) it’s a sign of an immature company.” Eric Ornas, Borland’s Director of Support, concurs. “In the early 80’s, the customer didn’t have a real voice in the product. It was talked about, but companies didn’t want the customer telling them what to do. It was all technology-driven, and ‘Tech Support’ was definitely not welcome in Engineering’s ivory tower.”

Now industry and market realities are very different from what they once were, but the end result is still the same. “Development cycles are so short and the marketplace is so competitive, you literally don’t have a moment to spare before you get a product to market,” says Ornas. Curiously, while many believe this highly competitive climate is the reason that developers push product out the door before Support has the opportunity for input, others insist that it has actually paved the way for increased Support involvement. “The shortest route to the market intelligence that you must have in order to compete is the real-time flow of information from the customers via the support department” Ornas insists. “You’ve got to move quickly, and you can’t afford to be wrong” As Byrum succinctly states: “Companies have realized that they can’t let the competition understand what their customers want better than they do. They’re competing on features, performance, price and time to market all at once — and you have to win on every front in order to win at all.”

The significance of these market changes is also emphasized by Jerry Weltner, VP: Marketing at TriSource. “The playing field has been leveled, and Support is now an important part of the team. They’re needed. And when you need something from someone, you have to give them a little more respect. You can’t just take your ball and go home anymore, because in this case, it’s Support’s ball!” But Support cannot sit back and wait for the rest of the company to beat a path to its door; it must market the data resources and services to Senior Management and the other business units if it is to become a full professional partner in any company.

What roles can Support play?

As no market research can, the daily flood of calls from every sector of the customerbase provides a constantly updated understanding of how people are perceiving and using the product, and of what features are the most used and desired. While well-positioned to advise customers on ways to gain the maximum value from the product, the support team is also uniquely situated to make product design and upgrade recommendations to both Marketing and Engineering.

“It’s vital that the whole company recognizes the customer advocacy role that customer support and service are playing,” says Byrum. “When Engineering listens to the support group, they’re really listening to the voice of the customer.” Years ago, when she was at Aldus in the early days of the PageMaker product, Byrum saw the then-CEO, Paul Brainerd, set the tone for the company in this regard. “There were regular sessions between Support and Engineering, called ‘What We’re Hearing’ meetings. They came from Paul’s insistence that what the customer had to say was very important. The meetings fed an appetite within the company to know what the customer wanted, what the customers were thinking.”

“We can’t just take our ball and go home anymore, because in many cases, it’s Support’s ball!” Jerry Weltner, VP Marketing, TriSource

Other companies and groups have also begun to recognize the potential of Support. “I welcome Support’s input in the whole product design process, from initial specification through design review” says Debra Scharfetter, Vantive’s VP of Engineering. “They have a different perspective that’s very valuable. When it comes to supportability, there’s no one better than they are on this issue.” As the head of engineering for a company focused on producing support technology, Scharfetter is in an excellent position from which to comment. “Most support organizations have good people but often don’t get the opportunity to really show what they can do. That needs to change.”

The opening, however, may not come from Engineering or from management, Scharfetter notes. “Don’t wait to be asked. For example, Support came to me with a particular problem that was generating a lot of calls. It was an error message that they felt was unnecessarily confusing, and was resulting in a lot of known-explanation calls. They had the data to back up their point, and they took the initiative to propose a specific alternative, including an offer to rewrite the message to make it more understandable. I told them that if they’d write the message, I’d gladly change the product. They did, and I did — and the results were as predicted. The calls went down and the company saved time and money”

“When you boil it down, technical support and engineering/development are customers of each other,” Intuit’s Strauss notes. “But you have to give them what they need, or you don’t get the product improvements that you ultimately want out of it. Development wants actionable information that answers the question: How can we make the product better?”

Taking the Initiative

Obviously, some important prerequisites must be addressed before attempting to transform the role of Support within any organization. The first item of business is to gather the customer contact data that will back up your ideas and requests. This is where the modern support technology tools become invaluable. By knowing the details of call volumes, issue categories and severity levels as well as the specific technical aspects of the problems in advance, Support can effectively present a prioritized list of requested actions to Engineering that carries a lot of credibility.

If your support group doesn’t have a call management system, you may actually be able to enlist the aid of Engineering and Sales in the requisitioning of the technology that you need. By identifying the data that such a system could provide to their operations, the other groups will often lend weight to the request. “I’d put in requisition after requisition for an advanced phone system,” reports Julian White Eagle, Seagate’s Telephone Tech Support Manager. “They all were shot down. Finally, I went to see the VP of Sales and asked him if a faster telephone response time from support would help the sales team to sell more. He went with me to see the president, and fifteen minutes later I had my signed requisition in hand.”

The next step is to put the support team on a sound business footing. Even if you aren’t a profit center yet in the company’s current account structure, you definitely need to know your costs of operation to a fine level of detail. A verifiable accounting of the actual costs involved needs to underscore every presentation that the group makes. The formal approval of Accounting is critical to the credibility of the cost analysis. Just because you are not yet offering paid-support plans, don’t assume that you are not making significant contributions to the company’s bottom line; every call you prevent increases the company’s retained profits.

Once the data and the financial analysis in place, it’s time to design a campaign that will move the group from its current status to one closer to the ideal. As with any political campaign, however, a realistic evaluation of the total company environment is a vital pre-requisite for success.

The Politics of Support

Every company has a unique internal political landscape. As any political strategist will tell you, the attempt to deny that politics are a factor simply closes the door on the possibility of effective change. If you’re going to win, you have to play the game.

The creation of a good game plan begins with an honest appraisal: How much authority and influence do you have right now? What’s the nature and status of your relationship with the other departments of the company?

Once you have accurately determined where you stand, the next step is to identify where you want to be at the end of the campaign. Ideally, what would that position look like?

With the before and after pictures clearly drawn, you can then assess the players. Who are your potential allies and supporters? What motivates them to share your goals? How far does that motivation extend? Where is resistance likely, and what might cause it? Knowing the desires and motivations of all of the players will enable the savvy manager to propose deals and trade-offs that can work, and to avoid wasting time on things that are out of reach at the moment.

After the plan is complete, it’s time to evaluate the risks and to make a decision: Is it a go?

Effecting the Change

“We started very carefully at Borland,” Ornas recalls. “We selected areas where we knew we could add value, and where we could have an immediate effect. On some projects, we found areas where Engineering was short-handed where we could fill in, but we picked only sure things. We didn’t want to fail, and couldn’t afford to. Once we had some momentum going, then we expanded a bit.”

AST’s Doyle took a similar tack. “I set up a service quality group in my organization, staffed with some of our best people. We made a specific proposal to Engineering to do what we called ‘Service Verification Testing’ for a new release on a trial basis.” The proposal met with some initial resistance from Engineering due to the time required for the testing, but Doyle was able to make the case. “I told them that we would guarantee them we’d find things that they would never see, because Engineering tests to see if a product meets the specs while we would test it from the customer’s point of view, using the same applications that the customers would.”

“For a while, it was tough,” Doyle admitted. “But we were able to take each issue and run it through our database and tell them how many calls it would generate if it got out to the field. And since we knew what our costs per call were, we could also give them sound predictions as to what the overall costs for each issue would be. The Engineering management looked at all this and were overwhelmed.”

“The final point was that we had a greater than 50% reduction in the failure rate for the product that we tested once it was released. We were able to show the company that the reduction was due to the things that we found and fixed via the SVT that under the old system would have gotten out into the field. We could say to Engineering: ‘You guys designed this stuff out of the product, you prevented millions of dollars of support costs by fixing it before it went to the field.”

“Once we had delivered the promised results on the first projects, then we had a different problem,” noted Ornas. “But it was a good problem to have. We had to manage the demand for what we could do. And there is some cost to providing services to Engineering, for dedicating those resources to such projects can detract from your overall service level. But it’s worth it. Even if you take off a few performance points while getting a product ready, you’ll substantially improve your performance later when the product is released in better shape.”

Matt Doyle is already looking ahead to the next steps. “We haven’t tied our SVT successes into the incentive system yet, but I can see that coming.” Borland is also interested in the idea of linking support prevention efforts to compensation.

The Shape of Things to Come

“We know where we ultimately want to be,” acknowledges Bill Strauss. “And we know what we have to focus on to get there. We have to start seriously decreasing the cost to revenue picture in support. The call-intensity per copy has to come down; support prevention is the only way to ultimately win on that point. For those calls that we cannot prevent, we have to deliver the correct answer the first time, every time.” All of that requires ever closer integration and cooperation between Support and Engineering.

At Borland, Eric Ornas also recognizes the needs and the trends. “We’re starting to blur the lines between Support and Engineering here. We’ve got people working all over in Engineering, and they have their members working over here. We haven’t figured out how to manage that yet, but we’ll soon have to, for it’s where the market forces are driving us.” Companies can’t afford the costs of Engineering vs. Support squabbles or feuding anymore. TriSource’s Jerry Weltner predicts steadily increasing evolution of the relationship as the market gets hotter and ever more competitive. “It’s a matter of risk and return. If you’re betting the farm because you’ve only got one chance to hit the market, the product has to be as good as you can possibly make it. That takes everybody pulling together hard and fast. Whatever gets in the way of that all-company effort has to go, or you lose. It’s that simple.”

©1996, 2008 Mikael Blaisdell