0.6 Focus groups  (Page 2/2)

Let me start with the pull string toy. They are notorious for breaking when children play with them. Therefore the Speak N Spell would not be reliable. We weren’t creating a mechanical toy but an electronic product with no moving parts - other than the keys on the keyboard. The product we were creating would obviously not have the same issues that the pull string toy would have. A idea of a noise maker that was suggested could have also come from the pull string toy.

The tape recorder accounted for many of the criticisms. It was well understood that a tape recorder played things in the same order they were recorded. Therefore there was little chance for any kind of randomness or variation. Yes, boring. There was a bit of adult iteration implied to rewind the tape, find the right location to start, etc. Further, most tape recorders were used to play loud music – at least it was probably the general experience of the participants. Particularly if they had older children who listened to pop music with the volume turned up loud. We actually did worry about the loudness of the Speak N Spell. But decided a volume control knob would not be a good solution (it was easily $0.10 extra, which we couldn't afford).

The vocabulary size would remain an issue. It seemed small to the focus groups and it seemed small to us. But we only had 256K bits of memory for all of the speech (vocabulary and overhead words) and the spelling for each vocabulary word. Even at 1,000 bits per second, there was a risk that we couldn't even get 250 vocabulary words in, along with all of the overhead words like "now spell", "wrong, try again", "the correct spelling of", and finally all of the letters of the alphabet and the numbers zero through ten. It was this concern that convinced us to have add on modules available for additional vocabulary words. Unfortunately, where we had to place the port for additional modules was in the battery compartment and not obvious to the child or parents that it existed or how to plug them in. Of course there were two reasons for being in the battery compartment:

• We needed to insure that the power was off before the modules were inserted or removed. The fact that the batteries had to be removed to insert a module guaranteed that the power was off.
• I had chosen, for cost reasons, to use a new, untested, tin/lead connector system rather than the proven, reliable gold plated system. The issue was that the tin/lead system was only good for about 100 insertions before it began to fail. It was obvious that by hiding the module port in the battery compartment we could be assured that the modules would not be inserted too many times.

One comment we heard from the focus groups irritated our speech research team. When listening to the various character voices, the focus groups were leaning toward a voice with monotonic robotic speech, rather than the natural speech the research team had struggled to make happen. It might be great for the market to have a robotic voice, but our research staff didn't want to have to explain to the technical community the reasoning for having created such bad speech synthesis.

The results, taken at face value, were a bit disappointing. But we took it with an optimistic view and continued on with the program only making a few minor changes to product specification. And, as we had done with the information we had gathered from the technical community and the educational expert, we didn't disclose the details to our management team. It was fortunate that our management trusted us to know to do the right things.

What we learned

Now to answer the second question of "what did we learn from the focus groups?" The obvious thing we learned was that focus groups are made up of normal people rather than wild eyed crazies, also known as "nerds". Therefore there is no way to properly explain a new concept other than the actual working model. Even with a working model it becomes difficult. This lesson was an important one to learn and put into practice. Many times since then I have had to explain a new concept using available combinations of products they knew about. But I knew form this experience that getting them to leap to the same understanding that we had about the new product concept was not going to happen.

The best way to explain what we learned is to embellish an old saying. It goes something like this: "A picture is worth a thousand words, and a demo is worth a thousand pictures." We realized that we needed to expect people invited into a focus group not to understand. And, then, to test the results later when we had an actual working prototype to demonstrate. It became obvious to us that waiting until we had the working prototype would delay the product introduction far too much. Remember, we had a real deadline to meet - Christmas!