Most pay-per-call telephone applications are developed from a simple theme. I recall Nick Loder, co-founder of Lo-Ad Communications, as further categorizing the most successful pay-per-call applications to be founded upon either: fate, lust, or greed. There are many exceptions, of course, and one such exception was the Dial-A-Shuttle program introduced in 1984 by the National Space Society and AT&T 900 Services. Dial-A-Shuttle employed the AT&T product “Call Broadcaster”, one of the two AT&T “Dial-It” services then existent. (The other Dial-It product was “Call Counter”, most notably remembered for its introduction on the ABC television program Nightline during the Grenada invasion, and later for the ‘Larry the Lobster’ poll on the NBC TV show Saturday Night Live. Both employed ‘yes/no’ votes using two different 900 numbers.) Call Broadcaster employed group access bridging (GAB) technology to deliver the live feed from the shuttle control site in Houston to hundreds of simultaneous callers with a single 900 number.
Early in 1995, AT&T decided to eliminate both the ‘Call Counter’ and the ‘Call Broadcaster’ offerings. The services carried high daily minimums, and much of the functionality – if not the total capacity – had become available through service bureaus. When the National Space Society approached AT&T later in the year to again offer “Dial-A-Shuttle”, they were surprised to find the Call Broadcaster service they had previously employed was no longer available. AT&T suggested to the National Space Society that possibly I could come up with an alternative.
When the NSS officials explained that the 1984 edition had received 500,000 calls and the average stay time was 4.5 minutes, I became very interested. However, replacing what had been a part of the AT&T infrastructure for over 10 years in a matter of weeks was not only impossible, it was also impractical.
Meetings with the NSS ensued. Once they learned that not only could the program NOT operate exactly as it had before, but that it COULD include features not previously available, they became enthusiastic and active participants in developing a replacement program design. Completing the design for a system that would replace Call Broadcaster actually took very little time. That is not to say that the development of the program progressed effortlessly or smoothly, just that deciding WHAT to do was the easy part. The program had a 10 year legacy to uphold – live feed from space. Combining that history with the desire to make programming information available to rotary and pulse phone callers meant there was no choice but to employ some sort of GAB technology. The only remaining details to be addressed were what other features should we employ, and how could we keep this from costing more than the shuttle launch itself?
To the surprise of many, adding features and controlling costs were not antithetical. In fact, adding features became one way of attempting to control costs. We had begun working with a price of $.99 per minute, simply because that price range had worked before. As with any undertaking, the examination of primary, secondary, and even tertiary objectives became essential to arriving at a decision regarding both price AND features. The main objectives to consider in this program were:
- Increase membership in the NSS
- Reflect favorably upon a major sponsor/benefactor
- Provide useful information on the national space program
- Not lose money
By keeping the price at $.99, we could expect more callers. However, high call volumes alone would not assure achieving any of the listed objectives.
Increasing membership in the NSS would require more than just a ‘warm fuzzy feeling’ to be created by the program. Beyond increasing awareness, we wanted to provide a call to action. If we were to provide such a call to action by offering an opportunity to get more information about the NSS – while still on the same call – we would have to be prepared to respond to that request. Transferring callers between live GAB facilities and recorded information would add significant costs to the program. And how could we afford to offer GAB and Interactive Voice Response (IVR) for thousands of simultaneous callers, assuming we could locate the capacity outside the AT&T network.
We were able to meet the needs of each of the sponsor’s objectives through the same ‘modular’ approach we use in every program design. After identifying the component parts of the program design, we arranged for them to operate independently, yet be connected when and as required.
The heart of the design was dependent upon IVR acting as the ‘traffic cop’ that would allow callers to reach the component parts that appealed to their specific interest. Digitized voice message files occupy much more space than digitized data files. When IVRs operate in a networked arrangement (to accommodate high capacity), the movement of voice files between components of the network can become a problem. Likewise, recording and storing voice files can slow down that network, regardless of total capacity. To accommodate the desire by NSS to gain more direct information from those callers interested in joining, we transferred those callers out of the network to an isolated system that could record their requests without impacting on the overall network.
Once the calls were sent out, unless we made some modifications, they would have to hang up and re-dial. If we were truthful in the menu (as we felt compelled to be) many callers would NOT choose to request more information IF they were to be bumped out of the que for what they really called about – the live feed from space. The solution to this dilemma came from an unexpected quarter. The NSS had inquired about offering Prepaid Multi Quest 900 cards to sponsor/benefactor executives, who were kind enough to provide a grant to get the program ‘off the ground’ so to speak. There wasn’t enough time to print up the cards, so we ‘snapped in’ a previously developed module that allows sponsors to activate and deactivate Personal Identification Number (PIN) access to 900 programs via their 800 access number. The access path for the PIN callers also became our way for repatriating 900 callers who had chosen to leave a name and address. The executives had a way into the program, and our callers who wanted more information about the NSS would not be resistant to asking for it because they would have to re-dial.
The issue of the GAB capacity was a simple idea that took a long time to explain. Teleconferencing as a telecommunications feature is certainly not declining. One of the reasons AT&T scrapped dedicated capacity was that there are a lot of competing resources on the market. Once we were able to explain the concept to teleconferencing vendors, many were eager to participate. AT&T itself assigned significant capacity to the project for the entire duration of the flight, which was much appreciated.
The reason why vendors were willing to participate, and how we were able to keep costs down, has a lot to do with how we planned to distribute the calls. Each vendor was asked to supply capacity on an ‘as available’ basis. This allowed vendors to sell ‘excess’ capacity, without requiring a reservation and the costs associated with such a purchase. How could we be sure we had enough capacity? Sign up a lot of vendors and build a supervised network for distributing the calls. Simply put, the routing to the GAB facilities could be likened to a ‘rollover’ or ‘hunt group’ similar to what you might order from your local phone company – except for the capacity and some additional security features we added. Your local phone company doesn’t support most Centrex hunt groups past 16 lines. Obviously, we intended to go much further. Also, we didn’t want our calls getting confused with other scheduled conferences, nor did we want to entertain the possibility of ‘hackers’ getting into the live feed for free. Conferencing companies are used to such problems, and cooperation was the order of the day. We simply used what had worked before.
After testing, negotiating, and re-testing, we felt prepared on the scheduled day of the launch. The parts of this endeavor that were new had been tested and ready. The rest had all been done before. Run the full page ads. Launch the rocket. Handle the calls. Mail the check. We felt confident. Still one thing nagged at my subconscious. Being an engineer by training I often reduce my observations to formulas. The formula that haunted me in the days before take-off was S = 1/k C. Expressed in words, ‘the success of the endeavor is often inversely proportional to the complexity of the undertaking.’ Without exception, my 10 years in designing audiotext programs had taught me that the more time spent on developing intricate applications, the more likely attention will be diverted from what it takes to make a program offering a success – promotion. However, I had seen the client list of advertising placement. This had worked before. Our program was the ‘new and improved’ version. What could go wrong?
As it turns out, plenty. A full page ad in USA Today that was expected to include the 900 number wasn’t placed as planned. The placements at the premier of Apollo 13 resulted in lots of calls to the NSS 800 number requesting more information, but netted a disappointing number of calls to the 900 number. On the day of the launch, I watched Peter Jennings of ABC news give 10 times more coverage to the allegedly improper flight of an Air Force General costing $116,000, than to the historic 100th manned space flight, with a first ever list of accomplishments at a cost of over $116,000,000 ! When the shuttle with our crew and two Russian Cosmonauts and one American who had been in space for 115 days finally touched down on July 7, 1995, I had little beside my sense of pride as an American to feel good about. Tom Brokaw, on the NBC Nightly News called the STS-71 flight “The most successful mission in space, ever.” How did our part in this historic event – which was such a technological accomplishment – result in such a financial disappointment? Has America again become jaded regarding the marvels of space travel? (As they had become for the launch of Apollo 13, without the networks even providing live coverage of the launch.) Does it matter? The message seems to be that no matter how compelling you think the application might be, others have to think that too before the phone will ring. Any successful pay-per-call program doesn’t just depend upon successful advertising , it starts with advertising. If you want to make money from rocket science, you need to start by reading Scientific Advertising.
Epilogue
This article was prepared with the assistance and permission of the National Space Society. Although no one got rich, the program did meet its stated objectives, and Telecompute was proud to be involved in this project. Call Broadcaster, Call Counter and MulitQuest are registered marks of AT&T.