The Truth About Star Trek and the Ratings

Still from "Court Martial" (1967)

Introduction: Success or Failure?

It’s part of the popular understanding of Star Trek that the original series was a ratings disappointment during its first run, which was cancelled by NBC due to low ratings. This view has been reiterated in memoirs, newspapers, documentaries, and academic studies. As Herb Solow and Bob Justman put it in their book about the making of the series:

From the premiere of 'Man Trap' to the finale 'Turnabout Intruder,' despite all the letter-writing campaigns, marches on and harassment of the network, after all the petitions and phone calls and everything else, Star Trek’s Nielsen ratings had dropped by well over fifty percent from birth to death. 

- Herbert F. Solow and Robert H. Justman, Inside Star Trek: The Real Story (1996), p.415

Recently, however, author Marc Cushman has been challenging this account in a series of self-published books and a flurry of interviews promoting them (my review of Cushman’s first volume, These Are The Voyages: TOS – Season One, can be found here). In one of those interviews, at Trek Core, Cushman said:

Star Trek was not the [ratings] failure that we had been led to believe. 

It was NBC's top rated Thursday night series and, on many occasions, won its time slot against formidable competition, including Bewitched, ABC's most popular show. And when they banished it to Friday nights, as Book Two will reveal, it was the network's top rated Friday night show. Yet NBC wanted to cancel it! Even when they tried to hide it from the fans at 10 p.m., during Season Three, it's [sic] numbers were not as bad as reported. So, once I made this discovery, then, of course, I needed to find out the real reason for the way the network treated Star Trek, and the documents regarding that, which build as we go from Book One to Two and then Three, are quite fascinating.

Cushman elaborates upon his argument near the end of his first volume, These Are The Voyages: TOS – Season One:

One must wonder why a network would even consider cancelling a Top 40 series that was almost always a solid second place in the ratings -- often hitting the No. 1 spot in its timeslot -- against formidable competition, pulling in, on average, just under 30% of the TVs in use across America. (On the few occasions when it slipped to third place, it was always in a close race for the number two spot.) 

- Marc Cushman with Susan Osborn, These Are The Voyages – TOS: Season One (2013), p. 541

The views expressed in These Are The Voyages about Star Trek's ratings performance are, needless to say, irreconcilable with previous accounts. Either the series was a ratings failure -- as has been so often understood -- or it was, as Cushman argues, a ratings success. In order to determine what is fact and what is fiction, I must first lay out the terrain of television audience measurement in the 1960s, and from there examine the methodology, claims, and reasoning of Cushman's argument in detail.

Still from "Assignment: Earth" (1968)

Television Ratings in the 1960s

For a variety of reasons, the landscape of television ratings has changed dramatically in the past fifty years. In the 1960s, when Star Trek first aired, ninety percent of the television audience was tuned in to one of the three broadcast networks (CBS, NBC, and ABC). Today, with the proliferation of hundreds of cable channels, only about twenty five percent of the television audience watches one of the broadcast networks – of which there are now five – and that percentage continues to decline. In 1960, a program watched by thirty percent of that night's television audience (or, in ratings parlance, a "thirty share") might have been cancelled due to low ratings. Today, a thirty share would indicate a monster hit. Thus, it is nearly impossible to usefully compare ratings from the three network era with those from today.

Additionally, in the 1960s, the A.C. Nielsen company wasn't the only ratings game in town. Although Nielsen was the largest ratings service at that time, it had three notable competitors – American Research Bureau (ARB), Trendex, and Pulse – which published their own ratings reports based on their own research methodologies. ARB and Nielsen largely derived their ratings through the use of an automatic recorder, although at the local level they still used the diary method, or:

...a form on which one household member recorded, in prescribed manner, information on television viewing. It typically asked for such information as program name, channel, and sex of listeners by quarter-hours. Diaries provided total audience ratings, computed by quarter-hours, and so did not yield average minute ratings. To calculate total audience ratings, the number of households counted in fifteen-minute intervals was expressed as a percentage of a specified base, usually the potential television audience.

- Katherine Buzzard, Chains of Gold: Marketing the Ratings and Rating the Markets (1990), p.49

In contrast, Trendex and Pulse derived their ratings reports based on two different types of personal interviews:

...the telephone coincidental...used by Trendex, and the in-home interview used by Pulse. The telephone coincidental questioned in-home respondents about what they were viewing when the phone rang, and secured information as to others watching at that time. In addition to being subject to problems of representiveness [sic] (only telephone homes could be reached), the telephone coincidental was expensive.

The personal in-house interview, by contrast, represented television viewing during the preceding twenty-four hours by individual household members.  It was a recall method and provided a measure of total audience. Details of viewers and household characteristics were collected. Pulse, the firm most identified with this method, used a program schedule to reduce the problem of memory loss. The personal interview method was also criticized for contributing to human error by interviewing one family representative for the entire family’s viewing. Its big advantage was the qualitative information it provided about the purchases of TV audiences. Its high cost limited it mostly to metropolitan areas.

- Katherine Buzzard, Chains of Gold: Marketing the Ratings and Rating the Markets (1990), p.50

The differences between these varying methodologies make it somewhat difficult to directly compare the ratings measured by one ratings service to those of another, and their conclusions were sometimes dramatically different. The September 14, 1966 issue of Daily Variety, for example, pointed out that, "ABC contends it is penalized from 10 to 15% in the national Arbitrons, a contention challenged by the other two networks. CBS and NBC, on the other hand, maintain that Trendex inflates ABC ratings."

Finally, it is important to understand that Nielsen, which was already the dominant ratings service in the 1960s, published several different kinds of television ratings while Star Trek was on the air. The backbone of Nielsen's ratings service was the National Television Index (NTI), also known as the "national pocketpiece" or the Nielsen national ratings. The NTI measured ratings based on a two week period using the Nielsen Audimeter:

...an unobtrusive little device which can hide in a closet, yet it records all video set usage – is the set off or on, to what channel is it tuned, what switches are made to other channels, is a second or third set also on, or even a portable one in the backyard? 

- Clay Gowran, "TV Today: How Nielsen Arrives at Those TV Ratings," Chicago Tribune (May 26, 1968)

Because of the sample size (1,190 homes), the technology being used (the Audimeter used a film cartridge, which the participant had to remove and mail to the A.C. Nielsen company at the end of each reporting period), and the time it took to generate a ratings report, the NTI took Nielsen two weeks to create once the reporting period was finished. In order to furnish the three networks with more immediate ratings information, however, Nielsen provided two other notable ratings services – the Multi-Network Area ratings (MNAs) and the "overnights."

The MNAs were based on a subset of the homes sampled for the NTI, and were focused on the thirty largest television markets in the country. Nielsen provided the MNAs on a weekly basis, and at a faster pace than the NTI (it took about a week for Nielsen to process the MNAs, compared to two weeks for the National Nielsens). The most immediate Nielsen ratings, however, were the overnights, which were released within twenty-four hours of being measured. The Nielsen overnights were based on a sample taken from the New York area market (measuring about ten percent of the national television audience). The ratings information found in the overnights, which measured a more urban audience, often painted a different picture than the ratings information found in the National Nielsens, which measured more rural television viewers. The Multi-Network Area ratings painted a picture that was somewhat in the middle.

Ratings comparison for Nielsen ratings 9/12/66 to 9/25/66

Consider, for example, the Nielsen ratings measured during the two week period that elapsed from September 12, 1966 to September 25, 1966. During those two weeks, Star Trek broadcast its second and third episodes, "Charlie X" and "Where No Man Has Gone Before." Note that a simple average of thirty market MNA ratings produces different figures than the NTI, which measured a larger audience. Thus, in the MNAs, Star Trek was competitive against My Three Sons during this period, nearly tying it one week and beating it by more than a full ratings point the next. In the National Television Index, however, My Three Sons pulled ahead of Star Trek by nearly 3.5 ratings points, finishing 12th overall in the National Nielsens (Star Trek placed 33rd).

(Note: Nielsen NTI data found in the October 17, 1966 issue of Broadcasting Magazine. Nielsen MNA data found in These Are The Voyages - TOS: Season One.)

Screencapture from the These Are The Voyages website (2014)

Evaluating Cushman's Ratings Thesis

Having set the table in regards to how television ratings worked when Star Trek was on the air, I'll now delve into the arguments laid out in These Are The Voyages. To begin with, here's Marc Cushman's basic understanding of the way ratings worked in the 1960s, from an interview at Trek Movie:

Here’s how it work [sic] back in the 1960s and even the 1970s: There were two ratings services. One was A.C. Nielsen. The other was Home Testing Institute that did TVQ – competitors. Nielsen would send the network the ratings – a page for each night so it was a seven-page report for all three networks, all the prime time shows.

As already established, there weren't two ratings services in the 1960s – there were four – A.C. Nielsen, American Research Bureau, Trendex, and Pulse. Home Testing Institute was not, strictly speaking, a direct competitor of any of the four ratings services, because it measured totally different things (which I will explain in a moment).

Secondly, as previously established, Nielsen didn't just send the networks "a page for each night." In fact, Nielsen sent the networks several different ratings reports – the overnights, the multi-network area ratings, and the National Television Index ratings. The overnights and the MNAs were broken down nightly, but the NTI was an average of a two-week sample.

Home Testing Institute logo (1963)

On the subject of Home Testing Institute's TVQ, Cushman is continually mistaken about what it actually measures. For example, in his chapter about “Miri,” Cushman writes:

As in past airings, Nielsen’s National survey, factoring in rural communities, gave Star Trek a couple of percentage points less than the “overnights” conducted only in metropolitan areas. But Nielsen wasn’t the only service counting noses.

Home Testing Institute, A.C. Nielsen’s competitor, had a survey of its own called TVQ. For the month of October, which “Miri” closed out, TVQ prepared a Top 10 list and ranked Star Trek as being in a three-way tie for the fifth most popular series on TV, under Bonanza, I Spy, Walt Disney and Red Skelton, and tied with Mission: Impossible, Family Affair and the NBC Saturday Night Movie. The Time Tunnel and Gomer Pyle were at nine and ten, respectively.

- Marc Cushman with Susan Osborn, These Are The Voyages – TOS: Season One (2013), p. 260

Cushman makes several errors in this passage. First of all, the ratings he identifies as Nielsen overnights were actually conducted by Trendex, and the ratings he identifies as "Nielsen's National survey" are actually the thirty market MNAs. These sort of mistakes are rampant in These Are The Voyages, which juxtaposes various Trendex, Nielsen, and Arbitron ratings with no explanation as to the different ways these ratings were measured or their various biases. From the revisions made to the second edition of These Are The Voyages – TOS: Season One, it appears that Cushman had trouble keeping the different ratings straight. Eight reports identified as “Trendex 26-city ratings” in the first edition of the book are relabeled as “Nielsen National ratings” in the second edition, and nineteen other reports initially identified as “Nielsen National ratings” have been relabeled “Nielsen 30-Market ratings” as well.

Perhaps a larger error is the characterization of TVQ as a nose-counting service, which is simply false. For reference, here is the TVQ report brought up in These Are The Voyages (printed in the December 5, 1966 issue of Broadcasting Magazine):

TVQ list from Broadcasting Magazine (1966)

To explain what TVQ actually measures, here is a selection from “TV’s Vast Grey Belt,” an article written by Walter Spencer, which appeared in the August 1967 issue of Television Magazine. Note the sentence I have placed in bold:

Another major grey-area yardstick for Klein [the vice president of audience measurement for NBC] is the “Q Number,” a service of TVQ. It is found by taking the number of people who consider a show among their favorites and dividing it by the total number of people who have seen the show. Thus a “high-Q show” has a dedicated following among people who have watched, although it may not have attracted a large audience.

Such a high-Q situation can occur when a good new show is put on the air against an established popular show; it may get a high-Q number as it picks up an interested audience from among those who tune in, while the majority of viewers are so busy watching their old favorite that they don’t soon get around to trying the high-Q show.

In other words, TVQ measures the dedication of a show’s audience, not the size of it. In spite of this fact, Cushman uses TVQ as an indicator of the size of Star Trek’s audience more than once, including in the following passage:

For its fourth week on the air, with “The Naked Time,” according to TVQ Star Trek won its time slot for the entire hour.

- Marc Cushman with Susan Osborn, These Are The Voyages – TOS: Season One (2013), p. 280

Another component of Cushman’s argument is the notion that the three networks were very secretive about ratings information in the 1960s, because they were afraid that if certain stars or producers got a hold of it, they would use the information as leverage to negotiate a bigger payday:

In the 1960s, A.C. Nielsen delivered the gospel that the networks swore by. But there was an air of secrecy surrounding the gospels -- the ratings reports were not for public consumption. Nielsen would “loan” the survey documents to its customers -- NBC, CBS and ABC, who were very selective with whom the information was shared. Unlike today, those all-important life and death numbers for a television series were confidential. The theory was that if an actor, or producer for that matter, knew exactly how popular his show was, he would be all the more difficult to deal with. Time has proven this thinking correct. Consider how much more a star of a popular series is paid today compared to the 1960s. Shatner was a top-dollar star in 1966, but was only making $5,000 per episode. That would be comparable to around $35,000 now, a paycheck that most TV stars wouldn't even get out of bed for.

- Marc Cushman with Susan Osborn, These Are The Voyages – TOS: Season One (2013), p. 279

Many of Cushman's assumptions in this passage are incorrect, and much of his reasoning simply doesn't hold water. The premise that television producers were denied access to Nielsen ratings information when Star Trek was on the air is incorrect. One of the ratings report sent to Roddenberry during the run of the series has been posted online, and others are available in Roddenberry's archival papers at UCLA (the source of most of Cushman's research material). The papers of writer/producer Bruce Geller, who created Star Trek's sister shows at Desilu (Mission: Impossible and Mannix) also include Nielsen ratings reports that he received during the same era.

The premise that the Nielsen ratings "were not for public consumption" is true, but the suggestion that the three networks were the only clients who used the ratings services is false. According to an article in the May 26, 1968 edition of The Chicago Tribune:

The Nielsen company has something like 600 clients–advertisers, advertising agencies, networks, stations, and program producers – who pay from a minimum of perhaps $15,000 a year up to a beautiful maximum of hundreds of thousands of dollars each 12 months for those reports.

The suggestion that Shatner's $5,000 a week salary was chump change doesn't stand up to much scrutiny, either. The cast of Bonanza, which was the number one show on television during Star Trek’s first season, earned no more than $1,000 during their first season of production and, after seven annual raises, were still only earning $12,000 an episode during the 1966-67 season. The complete National Nielsen ratings may have not been printed every week, but the dominance of Bonanza in the top ten was well documented at the time. If access to ratings information was the key to negotiating for a much bigger payday, it raises the question – why did the cast of Bonanza settle for what Cushman seems to think was so little?

The truth of the matter is that for a mostly untested leading man – his previous series, For The People, lasted just thirteen episodes before it was cancelled – Shatner was earning good money (and, contractually, his salary went up each year). Cushman's premise that $35,000 a week would be on the low end for a leading actor of a major series today is certainly true. However, his conclusion that Shatner's salary must have therefore been on the low end is based on the incorrect assumption that television production has kept pace with inflation. In point of fact, the cost of television production has far exceed inflation since the late 1960s.

Star Trek and the Ratings: The First Season (1966-1967)

If you’ve made it this far, you’re probably wondering, how did Star Trek do in the ratings? The way Cushman sees it, Star Trek was a huge success right out of the gate:

“The Man Trap” hit big in the ratings, drawing 46.7% of the TVs in use throughout America. The rating was a triumphant 25.2, compared to the 14.1 attributed to The Tammy Grimes Show and the 9.4 to My Three Sons. (Ratings reflect the total percentage of TVs in use that evening, tuned to a particular show.) Star Trek remained the clear winner at 9 p.m., as well. ABC’s most popular series, Bewitched, drew a 15.8 rating. On The CBS Thursday Night Movie was The Ladies Man, starring Jerry Lewis. It only managed a 10.7. Star Trek towered above them with a 24.2 rating and 42.2% of the TV audience.

- Marc Cushman with Susan Osborn, These Are The Voyages – TOS: Season One (2013), p. 279-280

Speaking broadly, Cushman is absolutely right – “The Man Trap” debuted to monster ratings. If Star Trek had been able to maintain these numbers, it would have finished squarely in the middle of the top ten for the 1966-67 broadcast season.

TV Guide advertisement for NBC's "advance premiere" week (1966)

There are a number of factors, however, that this gushing analysis of Star Trek’s debut numbers ignores. Chief among them is the fact that NBC decided to air three of their new shows for the 1966-67 season – including Star Trek – as “advanced premieres,” one week before the rest of the broadcast season began. This meant that My Three Sons, Bewitched, and even The CBS Thursday Night Movie were all reruns. Only The Tammy Grimes Show, which made its debut on ABC, would broadcast a new episode against Star Trek. As it turned out, The Tammy Grimes Show wasn't much in terms of competition. In fact, it was such a ratings disaster that ABC pulled The Tammy Grimes Show from its schedule after just four weeks, making it the first show of the season to be cancelled.

A secondary factor tempering this analysis is the fact that the numbers These Are The Voyages prints for "The Man Trap" were the multi-network area ratings, which drew from major metropolitan areas and favored Star Trek over its competition.

A third factor is that the ratings printed in These Are The Voyages are incomplete, since they only indicate how the show rated on the half hour. Luckily, when it comes to "The Man Trap," a more detailed ratings report exists in the UCLA archive to help fill in the blanks. Courtesy of the Gene Roddenberry papers at UCLA, here are the Nielsen MNAs for "The Man Trap," based on thirty television markets:

(Network – Show – Share)

8:30
NBC – STAR TREK – 46.7
ABC – TAMMY GRIMES – 26.1
CBS – MY THREE SONS – 17.4

8:45
NBC – STAR TREK – 43.3
ABC – TAMMY GRIMES – 27.1
CBS – MY THREE SONS – 19.6

9:00
NBC – STAR TREK – 42.2
ABC – BEWITCHED – 27.6
CBS – THURSDAY NIGHT MOVIE – 18.7

9:15
NBC – STAR TREK – 39.8
ABC – BEWITCHED – 29.8
CBS – THURSDAY NIGHT MOVIE – 19.0

These Are The Voyages reports the numbers from 8:30 and 9:00, but not the numbers from 8:45 and 9:15. An analysis of these figures shows that although Star Trek premiered to large numbers, it was shedding viewers every fifteen minutes, with an audience share that dropped from 46.7 at 8:30 to 39.8 by 9:15. By and large, these viewers weren't turning off their television sets when they gave up on Star Trek, but tuning into the competition on ABC and CBS.

When the multi-network area ratings from UCLA for Star Trek's second broadcast episode, "Charlie X," are added to the mix, the downward ratings trend continues:

(Network – Show – Share)

8:30
NBC – STAR TREK – 32.0
ABC – TAMMY GRIMES – 21.4
CBS – MY THREE SONS – 33.4

8:45
NBC – STAR TREK – 31.5
ABC – TAMMY GRIMES – 20.0
CBS – MY THREE SONS – 35.1

9:00
NBC – STAR TREK – 29.2
ABC – BEWITCHED – 25.0
CBS – THURSDAY NIGHT MOVIE – 36.0

9:15
NBC – STAR TREK – 26.6
ABC – BEWITCHED – 28.9
CBS – THURSDAY NIGHT MOVIE – 36.2

To be fair, these numbers represent the series’ ratings performance for only two weeks – specifically, on September 8 and 15, 1966. Unfortunately, the archival record at UCLA is incomplete – picked over by unscrupulous visitors when the library's reading room wasn't as well-monitored as it is today – but this data helps fill in a few blanks without having to absorb the cost of licensing ratings information form Nielsen.

Star Trek in the Top 40?

To truly get a sense of Star Trek’s ratings, you have to look at the numbers over time. As previously quoted, here is These are The Voyages' conclusion as to how the series rated, overall, during its first season:

One must wonder why a network would even consider cancelling a Top 40 series that was almost always a solid second place in the ratings -- often hitting the No. 1 spot in its timeslot -- against formidable competition, pulling in, on average, just under 30% of the TVs in use across America. (On the few occasions when it slipped to third place, it was always in a close race for the number two spot.)

- Marc Cushman with Susan Osborn, These Are The Voyages – TOS: Season One (2013), p. 541

There are a number of claims that could be examined here, but the most eye-catching is the assertion that Star Trek was “a top 40 series” during the 1966-67 broadcast season. As far as I can tell, Cushman comes to this conclusion based on a single Nielsen NTI report from early in the season, covering the two week period of September 12-25, 1966 when "Charlie X"  and "Where No Man Has Gone Before" were first shown. This report came from the October 16, 1966 issue of Broadcasting Magazine, which can be viewed online here (the NTI ratings report is found on pages 68-69 of the PDF). For ease of use, I have reproduced the Nielsen NTI report below:

Nielsen NTI rankings for the second and third episodes of Star Trek's first season (1966)

Eagle-eyed viewers will notice a few things about this report. First of all, the version printed in These Are The Voyages (on page 281 of the first edition) omits several programs (mostly news and talk shows, although some remain on the list) beginning with NBC’s Huntley-Brinkley Report at number 81. This doesn't impact Cushman's argument pertaining to Star Trek, although it is rather sloppy. Secondly, as I pointed out in my comparison of multi-network area ratings with the National Television Index, My Three Sons is a full 3.4 ratings points ahead of Star Trek in this report, despite the earlier MNA numbers Cushman printed for those episodes showing Star Trek barely coming in second against My Three Sons with “Charlie X” and beating it with “Where No Man Has Gone Before.” The revised, national numbers evident in the NTI report reflect people in rural communities, who watched Star Trek far less than people in metropolitan areas.

Nielsen NTI rankings for the fourth and fifth episodes of Star Trek's first season (1966)

These Are The Voyages is right about one thing. In the NTI report covering September 12-25, Star Trek was in the top 40. It achieved this position, however, against the extremely weak competition of The Tammy Grimes Show, which was removed from the schedule after only four weeks and replaced with The Dating Game, which did far better in the 8:30-9:00pm timeslot on Thursday nights. Indeed, the very next NTI report, published (in part) in the October 25, 1966 edition of The Chicago Tribune, shows Star Trek plummeting from the 33rd spot to the 51st position (see above).

1966-67 programs rated 30-70 in the National Nielsens (1967)

This drop in position makes sense. Even if you only examine the Nielsen MNAs that are presented in These Are The Voyages, it is evident that following the cancellation of The Tammy Grimes Show, Star Trek's ratings position began to decline. Indeed, after ABC pulled The Tammy Grimes Show from its schedule, Star Trek only reached first place in its timeslot with four first run episodes, and only once held the first place position for the entire hour. By the end of its first season, Star Trek’s average ratings position was 52nd place, according to Television Magazine’s August 1967 issue. The show was no longer in the top 40 – it wasn't even in the top 50. At 52nd place, Star Trek was in the middle of the road, ratings-wise, and in the same issue of Television Magazine, it and Mission: Impossible were "cited as examples of marginal shows that got tapped for a second year."

Still from "Bread and Circuses" (1968)

Conclusion: Why Was Star Trek Renewed?

Marc Cushman closes These Are The Voyages – TOS: Season One by asking why NBC would even consider cancelling Star Trek at the end of its first broadcast season. This question, however, is predicated on the assumption that Mr. Cushman's argument about the ratings is correct. I believe I have pointed out enough flaws in his reasoning and presented enough counter-evidence that such claims should be held in considerable doubt. 

Therefore, I believe a more appropriate question to ask would be this: why was Star Trek renewed for a second season? After all, the show was an expensive one to produce, and following an initial flash of success, its ratings had dropped to a level that was nothing to shout about. I can think of three reasons which may have been the tipping point convincing NBC to go forward with the program – although I hope my readers will be able to come up with others that I haven't considered.

First, Star Trek had garnered some awards recognition at the close of its first season, with five Emmy nominations (including the Emmy for Outstanding Drama Series) and a Hugo Award (for "The City on the Edge of Forever"). NBC may have hoped the publicity surrounding this recognition would have translated into increased viewership.

RCA ad for Star Trek and color television (1967)

Second, as argued by Solow and Justman in their book, Inside Star Trek: The Real Story, at the time the series was produced, RCA was the parent company of NBC, and Star Trek helped sell color television sets for RCA:

In 1966, NBC, at the behest of RCA, commissioned the A.C. Nielsen Company to do a study on the popularity of color television series as opposed to all television series. The results were expected–and very unexpected.

Favorite series were popular whether or not they were viewed in color. For example, NBC's Bonanza series was a top-rated series on the overall national ratings list as well as on the color ratings list.

However, in December 1966, with Star Trek having been on the air only three months, an NBC executive called with some news. The Nielsen research indicated that Star Trek was the highest-rated color series on television. I distributed the information to the Star Trek staff. We thought it was all very interesting, nothing to write home about, and went back to work. We were wrong; we failed to see the importance of the research

Perhaps those initial and subsequent Nielsen color series ratings contributed to giving Star Trek a second year of life. Putting aside low national ratings and lack of sponsors, perhaps a reason for renewing Star Trek, other than all the phone calls, letters, and demonstrations at NBC, was its position as the top-rated color series on the 'full color network.' NBC's parent company was RCA. Star Trek sold color television sets and made money for RCA. 

- Herbert F. Solow, Inside Star Trek: The Real Story (1996), p.305

Third, NBC may have simply had nothing better to replace the series with. Star Trek wasn't generating huge ratings, but the ratings weren't disastrous, either, at least not during its first season. According to Television Magazine in 1967:

Disaster...is the shock word in network programming. One of the best ways to avoid it is to put on even a weak grey-area show [a show ranked 30th-70th in the ratings] rather than take a chance with the least promising of the new batch of programs.

Fourth, renewing the series might have made sense because of the overall younger demographic it appealed to, which even in the late 1960s was becoming more important to advertisers. Paul Klein, the vice president of research for NBC, told Television Magazine in 1967 that "a quality audience – lots of young adult buyers – provides a high level that may make it worth holding onto a program despite low over-all [sic] ratings." He went on to tell the magazine that, "'quality audiences' are what helped both Mission Impossible and Star Trek survive another season." In a later TV Guide interview, Klein specifically mentioned Star Trek again, telling the magazine that the series was renewed in spite of weak ratings, "because it delivers a quality, salable audience...[in particular] upper-income, better-educated males."

Whatever NBC's reasons were for renewing the series, they made a commitment that Star Trek would be back for at least sixteen more episodes during the 1967-68 broadcast season. How the series performed ratings-wise in its second and third seasons may be the subject of a future post, but for now, I'll leave it at that.

Author's Note: Thanks to Dave T., Maurice M., and Kevin K. for reading an early version of this post and offering valuable feedback, which has improved it. Any remaining errors or logic gaps in the final version are entirely my own. If you've noticed any errors or have other feedback, please leave a comment or drop me a line using the contact form to the right. For more information about Star Trek's ratings performance, I can't recommend this piece at Television Obscurities enough. It certainly informed my approach to this post, and led to the discovery of a number of key sources.

Certain images courtesy of Trek Core.

Sources:

The Gene Roddenberry Star Trek Television Series Collection (1964-1969)

Chains of Gold: Marketing the Ratings and Rating the Markets (Katherine Buzzard, 1990)

Inside Star Trek : The Real Story (Herbert F. Solow and Robert H. Justman, 1996)

"Cult Television as Digital Television’s Cutting Edge," in Television as Digital Media (Roberta Pearson, p.105-131, 2010)

These Are The Voyages: TOS, Season One (Marc Cushman with Susan Osborn, 2013)