58.3, August 2011

The Web Portfolios of Independent Technical Communicators … and the Documents of Their Clients

John B. Killoran

Abstract

Purpose: This empirical study examines whether independent technical communicators' Web portfolios are worthwhile and how technical communicators manage to post work for clients on their own Web sites.

Method: The study surveyed 240 independent contractors, consultants, and principals of small businesses that maintain Web sites to market their technical communication services, briefly interviewed half of them, and analyzed posted samples of their work.

Results: Results reveal the extent to which posted work samples are useful in marketing technical communication services, and how the numbers of such samples are associated with their overall usefulness. Results also reveal how independent technical communicators overcome such challenges as clients' concerns about proprietary information, confidentiality, and intellectual property in order to post their work.

Conclusion: These results carry implications not only for how independent technical communicators disseminate a portfolio of their work but also for how any technical communicators would develop portfolios of their work for employment, academic credentials, or professional certification.

Keywords: independent contractors and consultants, intellectual property, portfolios, Web sites

Practitioner's Takeaway

  • Technical communicators should be aware of the intellectual property conditions that can influence how they may legally, ethically, and practically disseminate portfolios of their work.
  • Independent technical communicators can market their services more effectively by posting a portfolio of their work on their business Web site.
  • To overcome intellectual property conditions that hinder them from posting their work on the Web, technical communicators can seek to elicit their clients' consent, carefully select or modify the work itself, and demonstrate their discretion to prospective clients.

Digital Portfolios

Digital portfolios, in their progress toward widespread adoption, have reached a milestone of sorts, having been the subject of several user-oriented books from both academic and mass market publishers (Baron, 2004; Claywell, 2001; DiMarco, 2006; Kimball, 2003; Reynolds & Rice, 2006). But whereas educational uses of digital portfolios have received considerable impetus from academics, including their use for students transitioning from college to the workforce, somewhat less has been heard about actual digital portfolio use among practitioners in the workforce itself.

Within the field of technical communication specifically, digital portfolios have been discussed occasionally but encouragingly in such venues as STC conferences and STC's Intercom. There, practitioners and educators have encouraged technical communicators on the job market to develop portfolios (e.g., Scott, 2000; Smith, 2002; St. Amant, 2002), and several have encouraged technical communicators to post their portfolios on the Web (Barry & Wesolowski, 2001; Frick, 2002, 2003; Hunt, 1996; Kendus, 2002; Kimball, 2003; Smith, 2003).

Apart from the advantages that Web portfolios offer to practitioners in any field—they cost relatively little to construct and publish, are easy to maintain and update, reach geographically diverse audiences, and enable timely and convenient access for those audiences—they are especially suitable for showcasing the verbal, visual, and multimedia work in which technical communicators specialize, as well as the technical skills to create such work (Barry & Wesolowski, 2001; Kendus, 2002; Kimball, 2003). Anecdotal evidence is emerging that employers, particularly those in industries invested in digital communication, are using the Web to seek out information about prospective employees (Finder, 2006).

As useful as Web portfolios might be for technical communication employees and aspiring employees, they might be even more so for technical communication independent contractors and consultants. Unlike regular employees, such independent practitioners must continuously be on the market for their next client and so would have a continuous need for a Web portfolio. DiMarco recommended a Web portfolio for independent practitioners in any information-intensive field (2006, p. 10), in part because of the personal connection it can establish in an often-impersonal business environment (p. 20). A Web portfolio could occupy the heart of an independent's small business Web site, and technical communication practitioners have recommended that their independent colleagues maintain such a site for the convenience and credibility it offers (Frick, 2002) and for its utility in enhancing their networking activities (Broach, Gallagher, & Lockwood, 2006).

Though this literature about Web portfolios is quite upbeat, little empirical evidence has yet emerged to confirm their efficacy for technical communicators in particular. Barry and Wesolowski (2001) surveyed 35 technical communicators about Web portfolios, but as their purpose was to develop a style guide, they did not inquire into the portfolios' efficacy. They characterized their inquiry as preliminary and called for further research. Empirical studies across occupational fields offer somewhat promising evidence of digital portfolios' efficacy (Killoran, 2011; Ward & Moser, 2008), but as employment practices in each field can be quite distinct, it is unclear how such general studies would bear upon the field of technical communication specifically.

Without more direct evidence, technical communication practitioners might be wary of developing a Web portfolio, for at least a couple of reasons. First, before practitioners can reap any return on a Web portfolio, they must first invest the time, labor, and expense to construct, maintain, and promote a Web site, a considerable disincentive. Second, even if Web portfolios are shown to be worthwhile, technical communicators may have relatively little work that they can post openly on the Web. Unlike much of the portfolio work created by students or by creative workers in the arts, the work of technical communication practitioners is typically controlled by their employers or clients. Often employed by leading-edge industries to document advanced technologies, technical communicators must frequently operate under nondisclosure agreements to protect their employers' or clients' proprietary and confidential information. Even when they do not, copyright over their work for hire would fall to their employers or would often be assigned to their clients. And even when not so restricted by their employers or clients, technical communicators might want to avoid the appearance of indiscretion that posted work samples might create with the intended audiences of a Web portfolio—future employers or prospective clients—who may be anticipating their own proprietary, confidentiality, and copyright concerns.

This study addresses both of these challenges. It focuses in particular on the Web portfolios of technical communication independent contractors, consultants, and principals of small businesses (hereafter independent technical communicators or independent practitioners), who, as mentioned above, have the most at stake in Web portfolios' potential. Independent practitioners constitute a sizable portion of the technical communication profession—roughly one quarter of STC's membership (STC, 2004), for instance—a proportion that has probably not diminished over the past decade of outsourcing, offshoring, and economic constraints on full-time employment. Their prospects for remaining employed in the field depend on how successfully they market their technical communication services, and Web portfolios might play a significant role in such marketing.

In response to the first challenge—an ongoing lack of empirical evidence demonstrating the efficacy of Web portfolios—this study aims to answer the following research question:

  • Is it worthwhile for independent technical communicators to post samples of their work for clients on their own Web site?

Even if this first question is answered affirmatively, independent practitioners face the second challenge of marshaling samples of work they can safely post, a situation that prompts the second research question:

  • How do independent technical communicators manage to safely post work for clients on their own Web site?

To address this second question in particular, this article will need to refer to the legal entanglements that affect how technical communicators disseminate their work, or do not. These entanglements are many:

  • The U.S. Constitution's intellectual property clause and the First Amendment's protection of free speech (Herrington, 2011)
  • Various national laws, such as the work for hire doctrine in the U.S. Copyright Act of 1976 (Herrington, 1999b), the fair use doctrine in the same Act (Herrington, 2011), U.S. agency-partnership law (Herrington, 1999a), and a set of 1997–98 laws including the No Electronic Theft (NET) Act, the Digital Millennium Copyright Act, and the Sonny Bono Copyright Extension Act (Logie, 2005)
  • Differences among various nations' laws for work disseminated internationally, and international agreements, such as the Berne Convention for the Protection of Literary and Artistic Works and the Trade Related Aspects of Intellectual Property Rights agreement (Herrington, 2010; McKee & Porter, 2010)
  • Trademarks, patents, contracts, nondisclosure agreements, right of privacy, rights of publicity, fair use, labor law, works for hire, permissions, licenses, license agreements, waivers, releases, defamation, and moral rights (Helyar & Doudnikoff, 1994/2003)
  • Case law and legal precedents (Herrington, 1999b)
  • Litigation and threats of litigation (Porter, 2005)

For the sake of conciseness, this article will refer collectively to such legal entanglements as problems of intellectual property (IP), as their proliferation in recent years seems to be driven by struggles to control the “property” inherent in IP, especially digital property.

In this article, I first review how these IP entanglements can affect how technical communicators disseminate their work through digital media. Then I describe how I surveyed and interviewed a large sample of independent practitioners who maintained business Web sites to market their technical communication services, and how I analyzed the samples of work posted on their sites. In response to the first research question, the survey together with the Web site analysis will reveal the extent to which participants found posted samples of their work to be useful, and how the numbers of such samples are associated with their usefulness. In response to the second research question, the interviews as well as the Web site analysis will reveal how participants tried to overcome IP problems to post their work. The article closes by discussing the implications of these results not only for independent technical communicators themselves but also for any technical communicators assembling portfolios for employment, academic credentials, or professional certification.

Intellectual Property and Technical Communication in Digital Media

Digital media are widely recognized to have increased the public's opportunities to access and publish IP, but in so doing, they have also raised the stakes in potential IP disputes. In response, new IP laws, policies, and practices have been emerging and have been seen as reducing the public's recently won opportunities. This apparent rollback of opportunity may account for the urgency with which IP has been treated by scholars across various fields, such as law professors Lawrence Lessig (2001, 2004, 2008) and Jessica Litman (2006), and media scholar Siva Vaidhyanathan (2001, 2004). These three have argued that powerful commercial interests are constraining the freedom of action of cultural creators and consumers. In the same vein, technical communication scholarship has warned about the perils that IP laws, policies, and practices pose for the freedom of action of technical communication practitioners in particular, as well as teachers, researchers, and the public in general, especially from corporate interests (Gurak, 1997; Herrington, 2010, 2011; Howard, 1996; Logie, 2005; McKee & Porter, 2010; Porter, 2005).

In contrast with a tendency in such IP scholarship to side with ordinary individuals and the public against corporations and governments, some scholarship has seemed to side somewhat more with employers, stressing that technical communication employees should exercise their rights responsibly. For instance, Smith Diaz (2007) described why and how technical communicators should preserve the IP rights of their employers beyond U.S. borders by carefully maintaining documentation and observing confidentiality. Rife (2007), citing several cases of employees fired because of their online postings, advised technical communicators that the unfettered liberties expected in such Web venues as blogs are not as unfettered as they might have thought when weighed against employers' legitimate concerns for their organization's reputation, privacy, and security of proprietary information. Rife argued that technical communicators should share these concerns and restrict their online writing's range of purposes, topics, and language.

In an empirical study of copyright law's potential chilling effect on Internet discourse, Rife's (2010) survey of technical communication students, teachers, and practitioners failed to find evidence of a chill as severe as the literature would have envisaged. Nevertheless, results of individual survey questions revealed some degree of chill in certain Web composing activities. For instance, more than half of her respondents felt that some of their Web material was constrained because they could not incorporate others' material into it. Almost half affirmed that, because of copyright concerns, they had not posted certain material to the Web in the first place. Almost one fifth affirmed that, for similar reasons, they had subsequently removed material they had previously posted to the Web even though they had received no request to do so. Notwithstanding Rife's serene overall conclusions, such results indicate that, for some Web composition activities, concerns about copyright liability are indeed chilling technical communicators' communication.

However, counterbalancing and in some cases overriding technical communicators' concerns with copyright liability were the imperatives of their employers, their own consciences, their critical attitudes toward copyright law, their assessments of the low risk of being caught, and, if caught, their expectations of negligible consequences (Rife, 2010). Rife also reported how some who actually observed copyright law did so in a backhanded way by disguising their use of copyrighted material so that their transgressions would remain undetected (p. 56). In documenting these countervailing attitudes and practices, Rife's study is among the few in the literature to report how technical communicators, faced with potential IP problems, persist rather than back off. At several points, Rife called for further research on these kinds of issues and activities, a call that I respond to here. She concluded that technical communicators were “inadequately educated … on how to exploit their own intellectual creations for profit” (p. 63; italics in original), a condition that, in part, spurs my study into the practices of independent practitioners who are maintaining business Web sites in attempts to profit from their technical communication services.

Technical Communication Creators vs. Corporations vs. Consumers

The IP literature all but ignores the plight of independent practitioners who attempt to post their own work to the Web, as distinct from accessing or posting someone else's. Indeed, against the typically corporate copyright holders of digital work, the IP literature focuses largely on the plight of would-be consumers of such work, not on its creators. However, all work obviously has creators, and their attempts to control the dissemination of the IP they created deserve attention. Their position and capacity in the struggle over IP should not be conflated with those of the other IP stakeholders, for legal, ethical, and practical reasons.

First, as Herrington (1999b) has explained, the work of creators who maintain a status as independent practitioners, as distinct from employees, is treated differently under the Work for Hire statute in U.S. copyright law. Whereas work created by employees is deemed to have been created by the employer, work commissioned from nonemployees, such as independent contractors and consultants, tends to remain under their own authorship, even though a client would be paying for the work. Indeed, in contrast with the dystopian visions of increasing corporate control over IP promulgated in some IP literature, Herrington observed a trend in legal cases to increasingly recognize the copyright claims of independent creators over those of organizations that contracted their work. Thus, the onus would be on prospective clients, when negotiating with independent practitioners, to request that copyright be assigned to them, that nondisclosure agreements be signed, and that other IP terms and conditions be recognized.

Second, in contrast with the consumers who, in the IP literature, typically seek to access or use someone else's work—appropriations that, to some, may seem parasitical—cultural creators would seem to have a greater ethical claim to what they labored to create. Likewise, they may seem to have a greater ethical claim than do corporations that purchased the copyright over their works, such as the clients who commission them. For instance, some countries recognize creators' moral rights over their creations (Helyar & Doudnikoff, 1994/2003, p. 500; Herrington, 2010; McKee & Porter, 2010). Depriving a creator of the evidence of her skills and accomplishments for display within her portfolio could constrain her livelihood—not an easily dismissed ethical dilemma.

Third, in the typical IP situation, the stakeholder with the greatest capacity to exercise creative control over the work is obviously its original creator; corporate clients and consumers become involved usually because they lack the skills and resources to create a comparable work of their own. Indeed, the creator is also probably more familiar with the final product than her corporate client who may hold the copyright to it, having herself gone through the experience of writing, editing, and/or designing it. The creator probably retains digital copies of the work, as well as the software and other resources used to create it—software that could easily be reopened to create an excerpt, to disguise the client's name and other identifying information, or to revive an earlier draft. If such adaptations of the client's final product were posted to a Web portfolio, would the client even notice?

For these legal, ethical, and practical reasons, independent practitioners would seem to retain a much greater authorial license over the IP they create than do the consumers who are typically the focus of IP literature or, indeed, the corporate clients who contract such IP. However, in practice, independent practitioners may seldom be able to exercise that license to display much of their work on their own Web site, for a corresponding set of legal, ethical, and practical reasons.

First, technical communicators are specialists in writing, editing, and/or design but typically not in law. Rife's (2010) study demonstrated the degree to which technical communicators misunderstand or are ignorant of copyright law. To preclude IP entanglements, technical communicators have been advised to seek out legal counsel (Glick-Smith & Stephenson, 1998, pp. 90–91; Helyar & Doudnikoff, 1994/2003, p. 501). By contrast, technical communicators' corporate clients are more likely to already have legal counsel, as well as legal policies, legal experience, and the financial resources to prevail in legal disputes. They would typically seek to preclude legal entanglements by imposing contracts that explicitly spell out how the contracted work may or may not be disseminated. Indeed, Herrington (1999b) recommended that technical communicators themselves avoid legal entanglements by negotiating such contracts. Clients paying for work would likely insist on terms and conditions covering copyright, nondisclosure agreements, and other IP issues that serve their own organizational needs, not those of the independent practitioner.

Second, regardless of their clients' consent that freed independent practitioners to post their clients' work to their Web portfolios, practitioners may have to impress upon prospective clients that they maintained ethical practices in obtaining that work. Technical communication work often involves documenting new IP itself, and many practitioners specialize in leading-edge industries that generate lots of new IP. Prospective clients in those industries might have their own IP concerns in mind as they evaluate independent practitioners and would note a practitioner's apparent liberties with her clients' IP. Accordingly, independent practitioners would have to maintain the appearance, if not the reality, of high ethical standards, and they may therefore conclude that the best skill to display on their Web site would be discretion.

Third, for practical reasons, the kinds of alleged IP transgressions debated in the IP literature are often never actually disputed outside the literature because the parties involved are ignorant of the alleged transgressions and indeed of each other—say, a student using a graphic found on the Web unbeknownst to the graphic's copyright holder. By contrast, an independent practitioner and her client are already united in a relationship based largely on the work whose dissemination could potentially be at the heart of an IP dispute between them. The independent practitioner might wish not to jeopardize that relationship with a client from whom she might get repeat business or referrals. Many organizations monitor their Web reputation by regularly conducting searches on keywords matching their brand names, trademarks, and other identifying information—searches that would easily unearth the kinds of terminologically dense work in which technical communicators specialize. A Web portfolio would be a very exposed forum in which to take liberties with a client's work, and accordingly independent technical communicators may avoid even attempting to do so.

Thus, for these legal, ethical, and practical reasons—collectively, rhetorical reasons—independent practitioners, with whatever authorial license they may have over their work, may seldom be able to display it on their own Web sites. To recognize more clearly in rhetorical terms the challenges independent practitioners face, it will be useful to briefly analyze the typical provenance of posted samples of work through the familiar rhetorical triangle of author, text, and audience. If, through this analytical triangle, we view both work originally delivered to clients and such work subsequently exhibited on independent practitioners' Web sites, we can see how such work would need to shift between two somewhat distinct triangles:

  • Authorship, which is frequently assigned to clients under the guise of copyright, nondisclosure agreements, and other IP terms and conditions, must be transferred at least informally back to practitioners in order that the work be republished under their own byline on their own Web site.
  • Text, created for clients' specialized purposes, might have to be recast to showcase not only practitioners' technical communication skills but also their compliance with clients' proprietary and confidentiality concerns and their own discretion.
  • Audience, originally clients and their stakeholders, would obviously be expanded to include prospective clients whose vicarious IP concerns may have to be appeased once the work is posted on practitioners' Web sites.

Hence, to post on their own Web site work originally done for clients would in principle require independent practitioners to intervene with these three rhetorical stakeholders to effect the shift. It is the challenge inherent in this intervention that prompts this study's second research question about how independent technical communicators manage to post samples of their work for clients on their own Web sites. We will return to this rhetorical triangle below where this study presents its findings in response to this question.

Methods

This study adopted a mixed methods approach that included a survey, interviews, and Web site analysis. Surveys are commonly used to research workplace writing in general (Anderson, 1985a, 1985b) and independent technical communicators' marketing practices in particular (STC, 2004; STC Consulting and Independent Contracting SIG, 2005). However, survey questionnaires inevitably constrain the kinds of information respondents can offer (Anderson, 1985b, p. 494), and so to complement the survey responses, willing respondents were asked open-ended questions in brief follow-up interviews, a dual method applied in other studies of technical communicators (Conklin, 2007; Dayton, 2004; Rife, 2010). Content analyses in varying degrees of methodological purity have frequently been used to explore large samples of individuals' Web sites (Döring, 2002) and organizational Web sites (McMillan, 2000; Thayer, Evans, McBride, Queen, & Spyridakis, 2007). As many technical communication business sites share elements of these two—in many cases representing one individual but representing him or her as a corporate entity—this study's large sample of Web sites was similarly subjected to systematic analysis.

Using search engines and a variety of other sources (described in Killoran, 2009, 2010a, 2010b), I collected an international pool of more than a thousand Web sites representing businesses and individuals offering technical communication services. Sites that did not show evidence of ongoing maintenance were removed (see Killoran, 2009, for details of this culling process), leaving a pool of 638 sites. After a brief pilot study (described in Killoran, 2009), proprietors of these 638 were e-mailed a solicitation to participate in a Web-based survey, and nonrespondents were e-mailed two follow-up solicitations. A total of 240 usable questionnaires from respondents in 15 countries was received.

When factoring in six unusable questionnaires and factoring out 17 sets of undeliverable e-mails to sites posting defunct e-mail addresses, the response rate was 39.6%. Seen in light of the response rates in other surveys of workplace writing, which in pre-Internet studies typically lingered under 50% (Anderson, 1985b, p. 468) and which more recently have been falling (e.g., Eaton, Brewer, Portewig, & Davidson, 2008, pp. 115–116), this response rate is within a normal range. It compares favorably with response rates achieved in other surveys of small businesses about their Web sites (Flanagin, 2000; Pflughoeft, Ramamurthy, Soofi, Yasai-Ardekani, & Zahedi, 2003) and of independent technical communicators (STC, 2004; STC Consulting and Independent Contracting SIG, 2005).

Survey

The survey instrument, as well as the follow-up interviews, inquired into many facets of a participant's business Web site, not just the (absence of) samples, a diversity of purposes that is common in survey-based research (Anderson, 1985b, p. 457). Only two survey questions are pertinent to this article. The first, a multipart question, asked respondents to rate the helpfulness, to their business specifically, of each of several features frequently appearing on the business Web sites of independent technical communicators, including a “portfolio or samples of your work.” The main text of the question read as follows:

Listed below are some kinds of information featured on the business sites of many technical communication companies, consultants, and independent contractors. For those kinds of information featured on your site, how much has such information helped you market your technical communication services? (For information not featured on your site, answer “Not applicable.”)

A numerical response scale for each Web site feature ranged from 0, defined as “[helped] not at all,” to 3, defined as “[helped] a lot,” with two additional response options indicating “don't know” and “not applicable.”

The second question was a generalized version of the first, asking, “Overall, how useful has your Web site been to your business?” A comparable numerical response scale ranged again from 0, defined as “not at all useful,” to 3, defined as “very useful,” but unlike the previous question, no additional response options were offered.

Interviews

At the end of the survey questionnaire, respondents were asked if they would be willing to participate in a follow-up e-mail interview, and slightly more than half the survey respondents went on to complete the interview. To induce their consent, they were promised that the interviews would be brief, a condition that limited the interviews to about a half-dozen open-ended questions.

Interview questions were based on each participant's survey responses and Web site features, and so varied from participant to participant. Typically, survey respondents who had rated their posted work samples as relatively helpful in marketing their technical communication services were asked to explain how the samples were helpful, though in many cases, other Web site features that had been similarly rated were bundled into the same question. Usually, participants whose sites featured on-site samples were asked about how they managed to post such samples in the face of potential concerns about copyright, client confidentiality, nondisclosure agreements, and so forth. However, it was not feasible to inquire into the background of specific samples. In a few cases, participants whose sites did not feature samples but instead other detailed information about work for clients, such as case studies, were asked why they had not posted actual samples as well. In many cases, questions about samples could not be asked at all, in particular when a participant's site did not feature samples, as such questions had to compete with questions about other survey responses or Web site features.

Web Site Analysis

After the survey, participants' Web sites were downloaded to secure a stable corpus for analysis. One participant's site could not be accessed for downloading, and another participant's technically sophisticated site could be only partially downloaded, so both these sites were excluded from the remaining 238 that were analyzed. Note that only participants' business sites were downloaded, not other sites that they linked to. Because some participants linked to samples of their work posted on their clients' sites or indeed constituting their clients' sites, and because these samples or clients' entire sites have sometimes subsequently been changed or are no longer online, such off-site samples are not reliably available for analysis.

By definition, all Web portfolios include work samples, but not all samples posted on Web sites are included within portfolios; some samples are presented in isolation from other samples or without much accompanying introductory, explanatory, or reflective meta-text that is typical of portfolios. Accordingly, this analysis focused primarily on samples, and that term, rather than portfolio, will usually be favored below. This study operationally defined work samples as texts (e.g., documents or excerpts thereof, graphics, multimedia, Web sites, help systems) created by an independent practitioner (or business) for real or hypothetical clients and exemplifying the technical communication services that the independent practitioner (or business) is marketing to prospective clients. “Hypothetical” clients were included so as not to exclude the coursework or practice work of novice technical communicators.

Each Web site was carefully examined to locate work samples or links to samples. In practice, samples were easy to locate, as these typically small Web sites, often geared to showcasing their owner's work, typically went out of their way to explicitly identify such work. Even off-site samples that could have subsequently been changed or are no longer accessible could easily be recognized by the HTML code of the outbound hyperlinks leading to them. But as off-site samples were not reliably available for analysis, and as some sites added very little meta-text about any of their samples, a detailed content analysis of the samples was not feasible. Instead, it was decided to simply tally the on-site and off-site samples. Tallies offered a straightforward, objective measure that could be easily and consistently applied across the diverse pool of 238 Web sites.

Results

This section opens with results of the tallies of Web-posted samples, along with a brief overview of how the Web sites presented those samples. This overview can serve to ground the quantitative and qualitative results that follow. In response to the first research question about the efficacy of posted samples, I present the results of the survey questions and then explore potential associations between these and the tallies of samples. In response to the second research question about how participants circumvented IP obstacles to post their samples, I draw on participants' interview responses, their Web-posted samples, and accompanying Web meta-text to summarize the challenges participants faced and then to explore participants' tactics for surmounting those challenges.

Posted Work Samples

Examination of the 238 Web sites revealed that more than half, 138, featured samples of their owners' work (see Table 1). Almost twice as many sites featured on-site samples as off-site samples, including many sites that featured both. Of the total pool of samples, the vast majority, almost four fifths, were hosted on participants' own sites, whereas the remaining one fifth were hosted typically on clients' sites (or constituted clients' sites) and accessed through hyperlinks. The typical site with samples averaged 14 samples, though this average was inflated somewhat by some sites that featured several dozen each; the median number of samples was only 10.

Table 1: Tallies and percentages of Web sites with on-site, off-site, and any samples, and tallies of samples, in totals, averages, and medians

Types of samples

Number of sites

Numbers of samples

With samples (% out of 238 sites)

Total

Average among sites with such samples a

Median among sites with such samples a

On-site samples

120 (50%)

1514 (78%)

12.6

8

Off-site samples

68 (29%)

421 (22%)

6.2

4.5

All samples

138 (58%)

1,935 (100%)

14.0

10

a Averages and medians of on-site and off-site samples exclude sites with 0 samples of their type.

Though the architecture of these small Web sites was quite diverse, samples were typically presented in one of two locations:

  • in a section devoted explicitly to the samples, with its own link in the site navigation most commonly called “portfolio”;
  • scattered across pages devoted to the business's various communication service specialties, in which case a page describing a service specialty would typically follow up with a brief list of links to samples of work in that specialty.

In parallel with the reflective statements that educators recommend their students write to accompany their samples, Dimarco (2006, p. 103) and Kimball (2003, p. 154) recommended that practitioners introduce and orient viewers to their samples, but sites varied widely in the extent to which they included such meta-text. A few sites posted elaborate page-length introductions to each sample, typically describing elements of the rhetorical situation; the independent's work process, digital tools, and/or “solution” to that situation; perhaps the outcome for the client or the document; and perhaps a legal notice about the status of the sample itself, such as its copyright (such notices are discussed in greater detail below). However, many sites said little about their samples, often no more than a few phrases that perhaps identified the genre, document title, client, and/or IP status (e.g., copyright), but in general letting the samples speak for themselves.

The Efficacy of Posted Samples

Results of the survey question asking about how helpful participants' Web portfolio or samples were in marketing their technical communication services reveal them to be fairly helpful. Table 2 (column 3) shows a clear trend toward the higher numerical ratings: More than a third of respondents chose the highest rating, 3, signifying that their samples helped “a lot.” The percentages drop for subsequently lower ratings, with 10% or fewer of the respondents choosing each of the two lowest numerical ratings, signifying helping “a little” or “not at all.”

Table 2: Survey respondents' ratings of the usefulness of their Web site and their Web portfolio or work samples, in percentages

Response options a

Work samples' helpfulness

Web site's usefulness

All respondents n = 240

Only respondents whose sites featured samples n = 138

n = 240

3 – Very / a lot

35.4

50.0

41.7

2 – Moderately

19.2

24.6

33.8

1 – Slightly / a little

9.6

8.0

19.6

0 – Not at all

10.0

4.3

4.6

Don't know

6.7

9.4

N/A (Not applicable)

19.2

3.6

No response

0

0

0.4

Note: Because of rounding, percentages may not add up to 100.

a Response options for these two survey questions used a similar 0–3 numerical scale but defined slightly differently these ratings of a Web site's overall usefulness and a portfolio's or samples' marketing helpfulness. Where they are different, both sets of terminology are listed.

These results bear closer scrutiny, however, as almost a third of respondents who chose a numerical rating, 57 of 177, did not actually feature samples on, or linked from, their Web site at the time the site was downloaded for analysis (and five who chose the not applicable response had sites that did in fact feature samples). Along with hastiness and inattentiveness that are common to many surveys, several possible explanations might account for such discrepancies:

  • Some participants may have previously posted portfolios or samples on their sites only to have removed them before I downloaded their sites. Indeed, ongoing informal comparisons of the live sites with their downloaded versions have often shown that samples and entire portfolios have subsequently been added or removed.
  • Some participants appeared to have held a more generous understanding of what constitutes a work sample than that indicated in the operational definition, listing among their samples texts created not for real or hypothetical clients but for themselves, for instance, such as their own Web site itself. By such a generous understanding, 100% of participants' sites posted samples of their owner's work. (The operational definition was not included in the questionnaire.)
  • Some participants may have conflated a portfolio or samples posted publically on a Web site with those that they would show to prospective clients by more private means. Indeed, a few who responded to the survey question about a portfolio or samples with a numerical rating actually explained on their sites why they did not post such samples and invited prospective clients instead to contact them to view samples.
  • Some participants appear to have held a different understanding of the response options than that indicated in the questionnaire. In particular, even though the not applicable (N/A) response option was explained in the question itself, more than a third of respondents avoided it in their response not only to this question but (somewhat implausibly) to the other 18 questions in which it appeared; in its place, many seemed to favor the 0 response option, which the questionnaire always listed first. Among the 57 participants presenting such discrepancies with the question about their portfolio or samples, the majority never chose the N/A response option throughout the questionnaire, and the plurality chose the rating 0 in response to this question, suggesting that many may have been similarly substituting 0 for N/A.

Among the 138 participants whose sites actually featured samples (as operationally defined) at the time their sites were downloaded, half rated such samples with the highest numerical rating of 3, signifying helping “a lot,” and close to a quarter chose the second highest rating, helping “moderately” (see Table 2, column 3). By contrast, only a combined eighth of respondents chose the two lower numerical ratings. Compared with the corresponding ratings from all 240 respondents (in column 2), these generally high ratings from this subset of respondents with the freshest experience of a Web portfolio or samples provide an even stronger affirmation of their efficacy.

To answer the first research question with more precision, we can determine whether the quantity of samples is associated with their efficacy by testing for possible associations between the tallies of samples on participants' Web sites and participants' ratings of those samples. The survey's four numerical ratings from 0–3 could provide a scale of four efficacy levels, but as relatively few participants chose the lower ratings, a comparison among the four lopsided groups would likely be skewed. Accordingly, sets of samples receiving ratings of 0, 1, or 2 were combined into one “less helpful” group and contrasted with the “very helpful” sets of samples receiving ratings of 3. Bisecting the range of survey ratings in this way produces two roughly equal halves, ensuring an optimally robust comparison.

As shown in Table 3, a t-test revealed that the sets of samples rated as very helpful featured significantly more samples than did those rated as less helpful. This positive association raises the possibility that one factor making samples into helpful marketing resources may be their sheer number.

Table 3: Average numbers of samples on Web sites whose samples rated as very helpful and less helpful, and on Web sites rated as very useful and less useful, and t-test statistics of the differences

Survey question

Very helpful/useful a

Less helpful/useful a

Comparison

n

M

SD

n

M

SD

df

t

Helpfulness of portfolio or samples b

69

17.9

17.1

51

10.1

8.2

103

3.31 **

Usefulness of Web site

100

11.5

16.6

137

5.7

8.3

135

3.21 **

Note: The n columns list the numbers of Web sites, whereas the M columns list the average numbers of samples on those sites.

a Very helpful/useful sets of samples or Web sites includes those rated as 3, and less helpful/useful sets of samples or Web sites includes those rated as 2, 1, or 0.

b Includes only sites with samples.

** p < 0.01

The other survey question asked about the Web site's overall usefulness (its results are shown in Table 2) and hence might be expected to be less sensitive to the numbers of samples, as samples were just one of many potential features on independent technical communicators' Web sites and were absent from a large minority of participants' sites. Nevertheless, a similar analysis performed with the comparable “very useful” and “less useful” groups formed from ratings in this survey question produced similar results: As shown in Table 3, Web sites rated as very useful featured significantly more samples than did those rated as less useful. Note that, unlike the previous t-test, this one included data from all 237 participants who answered the question (one did not) and whose Web sites could be analyzed (two could not), not just those with samples. This association between Web sites' samples, despite their absence from many sites, and Web sites' usefulness ratings raises the possibility that samples, where present, may be contributing significantly to Web sites' efficacy.

Indeed, when asked to explain how useful their site was to their business, participants frequently mentioned their Web portfolio or posted samples of their work. Some indicated that they created and used their sites primarily to display such samples, and several mentioned that they would direct prospects explicitly to the samples, or that the samples were the part of the site that prospects most frequently examined. For instance, one participant who rated his portfolio as helping a lot in marketing his technical communication services explained this high rating:

My web site is primarily a vehicle to show a portfolio of work that I have performed for clients. As such, it provides me with tremendous credibility when I reach out to new prospects. By featuring the names of clients that I know other web site visitors will recognize, along with the work I have done for them, prospects are able to overcome a significant barrier to doing business with me: the fact that I am otherwise unknown to them. I think my web site is able to achieve this better than I could ever convey through alternative communication channels.

His site featured one of the largest portfolios in this study: 54 on-site samples mostly of white papers, his specialty, plus one link to an off-site sample, all featured in one long list organized mostly by client. In response to another interview question, he described one way he could confirm that his portfolio was reaching its audience:

When I reach out to new prospects, through email, I usually include a link to the portfolio section of my web site in the body of the email message. I have noticed that this does lead to more visits to my site – from analyzing web site logs kept by my ISP.

Challenges of Posting Samples

As useful as Web-posted samples appear to be for such marketing purposes, it becomes clear when examining participants' Web sites and interview responses that participants faced challenges in exhibiting samples of their work. The quantitative analysis above revealed that a large minority of Web sites featured no samples at all, and many that did mustered only a handful (see Table 1). Of the sites featuring samples, almost half linked to off-site samples, typically on or constituting clients' sites, indicating that many participants were quite willing to let other sites host their samples. Yet such off-site samples amounted to only slightly more than a fifth of all the samples tallied. Given how easy it is to post a link to an off-site sample, this relatively small portion of such links could indicate that clients seldom post the work that independent practitioners do for them. Accordingly, practitioners would usually have to take the initiative to post samples of their work on their own sites.

The challenges participants faced in doing so emerged in their interview responses. For instance, the white paper specialist quoted above also recalled an episode early in his business when, because of IP constraints, he refrained from posting work that could have added much-needed credibility:

I have worked on material that was for my client's internal use only…. On one project I wrote material for the largest vendor in the market [in my area of specialty]. At the time I did not have many clients and posting the work on my site would definitely have given my services substantial credibility. However, I knew the material was for internal consumption only and could not be used for my personal marketing purposes.

Several participants likewise reported that they had not bothered asking their clients for permission to post work when they knew the answer would be no, or had been denied permission when they did ask. Typical among the reasons offered was the confidential or competitive nature of their clients' industries or their own work in particular, as well as the cumbersome permissions process itself. Among the industries mentioned as particularly sensitive to posting any kind of material was national defense. For instance, one participant specializing in the defense industry explained,

Our work efforts are proprietary, and client confidential – no show and tell. We are not writing instructions for the proper operat[ion] of a can opener; we're creating data that keeps aircraft flying and devices that keep civilians out of harm's way…. There are client intellectual property rights and global security issues we need to protect!

Clients in less sensitive industries were sometimes concerned about revealing commercially valuable proprietary information to their competitors or potential paying customers. For instance, one participant offered as an example a project she worked on—

for a client who produces online content management software for organizations. The only thing I could potentially post from that project is the instructions for logging in on the site. Everything else shows what the [content management] interface looks like and would potentially reveal too much to their competition.

Perhaps as a consequence of such commercial pressures, many of the posted samples seemed to be texts originally designed for wide-ranging audiences and hence not devalued by exposure on participants' Web sites.

Even in cases in which the work might not have been so sensitive or commercially valuable as to preclude Web publication, a few participants mentioned that the process of pushing a permission request up a company's bureaucratic ladder was so time-consuming or intimidating that it discouraged them from trying. A participant whose site featured fictionalized case studies explained that she did not use real case studies or real samples of her work or even name some of her clients because “this would require obtaining permissions—often from folks several levels above the actual clients. I am not sure the effort to do this is worth it.”

Finally, some participants describing IP challenges pointed not just to clients for whom they did work but to prospective clients who would be viewing such work on their Web site. Anticipating the concerns of security-conscious prospects about the dissemination of their own IP, some participants refrained from posting any samples at all or at least carefully considered the selection of samples they did post. For instance, one participant whose site featured only four samples explained:

I am very careful to avoid placing what looks like sensitive information on my site because I would not like to create the wrong impression with potential customers with whom I am not yet working, and who might fear that I am liable to place their sensitive material on the Internet.

His page of samples was introduced with a notice explaining that, “[i]n the interests of confidentiality,” identifying details had been removed, and indeed, within the four samples themselves, no company names or other identifying information was featured.

Clearly, the work independent practitioners do for clients does not readily reappear as samples on practitioners' Web sites. In this light, let's examine how about half the participants managed to post at least some of their work on their sites. Adopting the rhetorical triangle of author, text, and audience that was discussed above, the following three sections report on the variety of ways participants intervened with each of these three stakeholders of their work's situation:

  • Authorship, in particular how participants communicated with, or did not communicate with, their clients for consent to post work for clients as their own on their own Web site
  • Text, in particular how participants selected, excerpted, redacted, and modified their work for clients to post as safe samples on their own Web site
  • Audience, in particular how participants communicated to their Web site audiences, especially prospective clients, the status of the IP displayed or withheld from display on their site

Authorship

Interview responses revealed how participants would communicate, or in some cases not communicate, with their clients for consent to post participants' work on their own Web sites. Such communication sometimes started even before the work itself was started, in the form of IP terms and conditions explicitly spelled out in the contracts that practitioners negotiate with their clients. For instance, one participant quoted a clause that he used in his standard contract: “The Client gives the Writer permission to publish the documents he has produced under this contract on his own Web site, either as extracts or in full.” Of course, some clients might not agree to such a clause, or would seek to impose conditions on it. This participant added, “In practice, I would only publish material that the client intends to be in the public domain.” His site featured only four on-site samples.

Participants more often mentioned seeking clients' consent to post their work after a project had been completed. Waiting enabled them to specify precisely what document or excerpt they were interested in posting and, according to one participant, also avoided creating an impression before or during a client's project that the independent practitioner was preoccupied with pursuing other clients. One participant explained the process by which she would seek clients' consent—a process that anticipated her clients' concerns, respected their prerogatives, and built trust:

I identified elements I would like to use, bearing in mind the issues of copyright and commercial sensitivity, asked about the theoretical possibility, created the extracts and approached clients to obtain permission. This was freely granted in every case – I would not publish anything without this permission. By showing exactly what I want to use and in what way when asking, this has always been granted and I always acknowledge the copyright. Most see it as a (very minor) way of promoting themselves as well.

A couple of participants mentioned that they preferred to make such requests by e-mail, especially with larger clients, so that they would receive permissions in written form and thereby avert potential IP disputes.

However, participants also reported that it was not always possible, or desirable, to ask former clients for such permission. Many independent technical communicators develop their Web sites only after several years in the profession, at which point they would recognize the opportunity to post samples of work done for clients or employers years earlier. Over that interval, those companies may have been bought out or ceased to operate. Accordingly, some participants mentioned posting without consent work they did for clients who had since become defunct. For instance, a participant who specialized in writing user documentation for technology companies explained that, because both a client for whom he wrote documentation and the products about which he wrote the documentation had become defunct, he did not bother to seek consent from the subsequent copyright holder to post a couple of the samples on his site:

I posted work for a company that was bought out by another company that was in turn bought out by still another company. I didn't have permission to post the samples, but I figured that since the products and company no longer existed, no one was likely to care. I tried to say nice things about the company on the pages with the samples. If the current owner of the copyright asks me to take the samples down, obviously I will.

Other participants likewise posted samples of work for clients without soliciting clients' consent, for other reasons. For instance, one participant confessed that he had not sought his clients' consent for posting two of his site's six samples because of the inconveniences of doing so and a likely denial, but nevertheless went ahead and posted them behind passwords: “Well, strictly speaking I should ask my client for permission, but it's a huge corporation, and I know that it would take forever and [I] would probably (if I'm honest) be turned down just because that's company policy…. Should my client(s) find out, I'll argue the case….”

A few participants who similarly neglected to solicit their clients' consent pointed instead to customary practice within the outsourcing culture. For instance, one participant justified his posting of clients' work by pointing to—

an unwritten rule that small-scale availability of most internally developed technology documentation for that purpose [of showcasing as work samples] is a quietly accepted practice. My general rule of thumb is to post only dated material (at least five years old) or material that is decidedly non-sensitive or not covered by a nondisclosure agreement. If asked by a material owner to remove any material considered sensitive by that company, I would remove the material immediately.

Among his site's 15 samples was one announcing on its title page, “For [company name] Internal Use Only.”

Similarly, another participant who, at the time of the interview, had only recently constructed her site acknowledged that IP rights for her site's ten samples resided with her clients but justified her use of them by pointing to analogous practices among independent practitioners:

I didn't advise anyone that I was using the material. I know that copyright passes to my clients, but I also know that Web developers routinely post thumbnails of the work they themselves have done, sometimes showing that copyright is shared, and sometimes not. In a similar way, my photographer gave me copyright of the photo she took, but expressly retained the right to use the photo for her own promotion. I looked upon the building of my portfolio with this knowledge and experience in mind and easily concluded: why rock the boat when what I really, really needed was an online portfolio. Why give clients the opportunity to object when it's not entirely clear that they have grounds for objection, and when I'm not at all well placed to deal with any objection?

Text

A second way participants managed to post work samples on their site was by carefully selecting, excerpting, redacting, or modifying the work they did for clients. Anticipating their clients' IP concerns, some participants indicated that they had selected only samples that their clients would consent to, such as documents revealing only safe information, old documents that had outlived their confidentiality, or safe excerpts from otherwise sensitive documents. For instance, a participant who claimed never to have been refused permission to post work she had done for clients explained that she considered her clients' concerns when selecting work for her Web portfolio:

I show [clients] the piece I intend publishing, and ALWAYS choose something that's innocuous and definitely not confidential – that makes it much easier! For example, for the retail jeweller, I used a page about diamonds, NOT about their security alarm policies!

Her 27 on-site samples were mostly screen shots of single-page excerpts, rather than full documents. Obvious excerpts like these were quite common across this study's pool of samples. For instance, many samples included a full table of contents for a long document but not the complete contents themselves. One participant explained that the table of contents “showed that [she] had produced complex documents.”

Participants also created publishable samples by redacting clients' confidential or proprietary information. Even a cursory examination of the 1,500-plus on-site samples revealed more than two dozen sites featuring samples with obvious redactions. Compiled into a composite document, such redactions could constitute a sample entitled “[Document Title]” for “Company XXX” located in “Anytown,” with the phone number “(xxx) xxx-xxxx,” an e-mail address “john@xyzcompany.co.uk” and the corresponding Web address “www.xyzcompany.co.uk,” about “the XYZ Product” trademarked as the “XYZ brand” and featuring “Company Logo.” It would also feature passages of greeked text and blacked-out text. The illustrations would be blacked out or replaced with placeholder images, and the verbal content accompanying illustrations would be left blank or replaced with filler such as sample text. In their interview responses, a few participants referred to samples with such redactions as “sanitized” or “obfuscated.” Participants also described other sanitizing modifications that would be less obvious to an outside observer, as they went beyond hiding passages to actually modify them, such as by inventing fictitious names for a client's products, rewriting passages at a more general level, or creating new graphics.

In some cases, participants admitted to sanitizing samples not to assuage their clients' confidentiality concerns but to evade their clients' discovery entirely. For instance, one participant admitted, “Some of these samples went up after the project was long over, and if I thought I could disguise the sample quite well I didn't always get permission.” Several of her 32 on-site samples were clearly sanitized. Sometimes, however, a sanitized sample might still be recognizable to the client. Such was the experience of a participant who claimed to regularly sanitize his samples before posting them:

Once or twice I wasn't careful enough about my “sanitizing.” The client emailed and asked me to take it down. I did immediately with profuse apologies. Had I gotten it right the first time, I would have been OK; but after that, I don't use that sample at all.

His portfolio page featured six extant on-site samples, two of which were openly introduced as having been “sanitized” and featured redactions like the ones described above. Perhaps revealingly, another item listed among his samples, described as “sanitized selections” from a client's document, linked instead to a page that announced, “So sorry. The sample you selected is currently unavailable.”

Audience

Finally, though many Web portfolios and samples were accompanied by relatively little meta-text, many participants did communicate to their Web site audiences the status of the IP displayed or withheld from display on their sites. Some portfolios and individual samples were accompanied by legal or quasilegal notices acknowledging the clients who held the copyright to the samples, the trademarks mentioned therein, and so forth. Apart from acknowledging clients' IP, such familiarity with legal discourse signaled participants' own regard for and compliance with IP policies, or at least the appearance thereof. One participant admitted to posting legal notices acknowledging his samples' copyrights and registered trademarks as a preemptive attempt to fend off any complaints about work he had posted without securing his clients' consent:

I know I'm running a fine line here. I don't own the rights to this material. How do I prove that I can write procedure manuals, user manuals, online help if I cannot show samples of my work? I got around that issue by displaying the legal notices. Besides, I never signed any legal document that said that I could not use these documents for self promotion.

Counterbalancing the legal and quasilegal notices acknowledging clients' IP were a few samples that retained markings indicating that they should not be disseminated indiscriminately on the Web. For instance, one site's sample request for proposals, an uncommon genre in the pool of samples, featured a long proprietary statement that concluded, “Individuals with access to this information shall be bound by the nondisclosure confidentiality agreement.” However, across the pool of Web sites, such apparent indiscretions appeared to be outshone by messages reassuring Web audiences that clients' consent had been obtained to post the samples. Some sites posting such notices also went on to explicitly reassure prospective clients of the independent practitioner's policies regarding clients' IP. For instance, one portfolio's quasilegal notice included a statement announcing that the samples' proprietary information might be redacted, followed by an uncharacteristically expressive promise: “We would do the same for you!” Another site specified that the technical communication business maintained the right to keep copies of clients' work for its portfolio, but in the next line promised “not [to] disclose or display any confidential material without written consent.”

Among participants whose sites featured few or no samples, or whose samples were excerpted or redacted, many were not quiet about it. On their sites, they explained absences by pointing to clients' copyright over the work, client confidentiality, the samples' proprietary content, and nondisclosure agreements. In this way, they could make a virtue of the absences by showcasing, if not their work per se, then at least their discretion.

Many sites, including both those that did and those that did not exhibit samples, invited prospective clients to contact the independent practitioner to see (more) samples. The invitation alone gave participants the opportunity to advertise that they had lots to show, and in a few cases, the Web sites would list the genres or even individual projects for which samples were available, or would invite prospects to describe the kind of genre they wanted to see. So whereas their work's proprietary or confidential character prevented participants from posting such work on the Web, it did not prevent them from using the Web to arrange a more discreet exhibition of their work, such as by posting samples behind logins or passwords, e-mailing samples, mailing a CD-ROM portfolio or hard copy samples, or showing samples in person.

Above, we saw an example of a participant who, without his clients' consent, posted two work samples behind passwords. His Web site nevertheless maintained the appearance of discretion, announcing that the password would be made available only to qualified parties upon request. In the interview, he added that he would also include the password in cover letters for prospective jobs. Another participant followed a similar practice:

I have a number of samples that I would always be uncomfortable about making available on a public web site. However, for potential clients who are seriously negotiating a contract, I have set up password-protected ftp sites to which only I can grant access and in those situations I would direct the client to the password-protected site and supply the client with the login name and the password.

Though the home page of this participant's Web site did link to one public FTP page listing 15 accessible samples, there were no links to these other password-protected FTP sites, and so any samples they listed were not included in the tally. It is possible that other participants similarly posted samples on their servers without posting links to them, making them accessible only to prospects.

Discussion

Against a backdrop of concerns about IP problems mushrooming throughout digital media, this study has explored the work samples that independent contractors, consultants, and companies post on their business Web sites to market their technical communication services. In response to this study's first research question about the efficacy of such Web-posted samples, results from both the survey and the accompanying Web site analysis tend to affirm that they are effective. Independent practitioners with Web sites that exhibit work samples tend to find those samples helpful in marketing their services. Notwithstanding the obvious importance of the quality of samples, the quantity of samples is directly associated with practitioners' experience both of their samples' marketing helpfulness