A read-only scan of 57 web agency front pages across six compliance dimensions. Every site had at least one accessibility finding, 70% sent no Content-Security-Policy, and a quarter tracked visitors before consent. The full method is published so the study can be repeated.
Abstract
This paper reports the results of a single-page compliance scan of 57 web agency websites, carried out between 16 and 25 September 2026 with the Sitejar scanner. Each site's front page was rendered once in a real browser, read-only and without login, and evaluated across six dimensions: accessibility, security, privacy, deceptive design, technical quality and SEO. All 57 sites belong to third parties; none is named here, and no published figure describes fewer than five sites. On the median site the scanner recorded 14 findings (mean 13.7). Every one of the 57 sites (100%) fell short of full marks in accessibility, and 50 of 57 (87.7%) did so in security. Low text contrast was the most common single finding, present on 45 of 57 sites (79%), followed by a missing Content-Security-Policy header on 40 of 57 (70%). By contrast, all 57 sites scored 100 in deceptive design and 50 of 57 (87.7%) scored 100 in SEO. We argue, with reasons, that these two clean results say as much about what a front page can show an automated check as they do about the agencies themselves. The paper sets out the sample, the method, the two results tables, what a confirmed finding does and does not mean, and the claims this dataset cannot support. The full protocol is described in enough detail to be repeated.
Introduction: why measure the people who build the web
Web agencies occupy an unusual position in the compliance picture. They are not the end of the chain: they build for clients, and their own sites are the shop window in which those clients judge them. If any group of websites should be expected to meet the accessibility, security and privacy standards that professional buyers now ask about, it is the sites of the firms that sell that expertise. That expectation is what makes them a useful benchmark. If the professionals' own front pages carry the common failures, a reasonable inference is that the average client site does at least as badly, and probably worse.
There is also a practical reason. The large public studies of web accessibility, most prominently the annual WebAIM Million, measure the top one million home pages as a whole, and their headline numbers describe the entire web rather than any single trade. According to the WebAIM Million 2026 report, 95.9% of the top one million home pages had at least one detectable WCAG failure in February 2026, with an average of 56.1 errors per page. Those numbers are the essential context for anything smaller, but they do not tell an agency owner where their own sector stands, and they measure accessibility alone. This study is narrower on both axes: one professional segment, six dimensions.
The research questions were three. First, across the six dimensions, where do web agency front pages stand, and which dimension is weakest? Second, which individual findings are most widespread, and how many sites carry each? Third, and as important as the other two, what can a single read-only scan of a front page legitimately claim, and what must be left to human review or a fuller crawl? The third question shapes the whole paper, because the honest answer to it limits what the first two can be taken to mean.
Data and methodology
The scanner and what one scan does
Every measurement in this paper comes from the Sitejar scanner, run by the authors. A scan loads the target URL in a real, headless browser, waits for the page to render including its JavaScript, and then runs a fixed rule set over the rendered document, the network requests it made, the response headers it received and the consent behaviour it observed. The output for a page is a set of findings, each mapped to the standard or instrument it fails, each with a severity, and each with a confidence label. The scanner does not modify the page, submit any form, follow any link or attempt to log in. It reads what any first-time visitor's browser would receive.
For this study, exactly one scan was run per site, against the site's front page only. No crawl was attempted and no internal page was visited. Where a site had been scanned twice in the collection window, only one scan was counted so that the site appears once in every table; the rule for choosing which scan to keep is given in the reproducibility section. The scanner honoured each site's own robots directives, which matters for the sample and is discussed below.
The six dimensions
Accessibility checks the rendered page against WCAG-aligned rules: text contrast, text alternatives, form labelling, accessible names on controls, heading structure, landmark regions and keyboard reachability of scrollable areas, among others. Security examines the response headers and transport: the presence and quality of Content-Security-Policy and Strict-Transport-Security, and whether any form on the page submits sensitive data with a GET request. Privacy observes consent behaviour: whether third-party cookies are set and whether tracking scripts load before a consent choice has been made. Deceptive design looks for interface patterns of the kind regulators pursue, such as pre-checked consent and asymmetric prominence of accept and reject controls. Technical quality covers markup validity and broken references. SEO checks on-page fundamentals such as titles, descriptions and heading presence. Each dimension yields a score from 0 to 100 for the page, where 100 means the scanner found nothing to report in that dimension.
Sample, dates and inclusion
The sample consists of 57 web agency websites, all belonging to third parties. Scans were run on 16 September 2026 and 25 September 2026. Rows that were dry runs, meaning the authors' own findings replayed against another domain for testing, were identified and excluded; there were none in the final count, so 57 real scans were counted and 0 rows were left out. The mean number of findings per site was 13.7 and the median was 14, which indicates the distribution is not badly skewed by a few outliers.
Three publication gates were applied before any figure was written down. No site is named. No published row describes fewer than five sites; twenty individual findings were held back from Table 2 for that reason alone, because on a sample of this size a finding present on two or three sites is close to identifying them. And every number is a scan the authors ran, not a figure taken from elsewhere. Where this paper cites external work, it says so explicitly and links the source.
Confidence labels
The scanner attaches one of three confidence labels to every finding. Confirmed means the rule is deterministic and the failure is certain from the rendered page alone, for example a header that is absent from the response, or a form field with no associated label in the DOM. Likely means the rule fired but a person could reasonably conclude otherwise on inspection. Needs review means the scanner has flagged something it cannot decide, and is asking for a human. Tables in this paper report the number of sites on which a finding was present at all, and separately the number on which it was Confirmed, so that the reader can see how much of each figure rests on certainty.
Results
Table 1: the six dimensions, worst first
Fifty-seven sites were measured in each dimension. The figures below are the worst score observed, the median, the best, and the number of sites that did not reach full marks.
Accessibility was the weakest dimension by every measure. The worst site scored 56, the median 78 and the best 96. All 57 of 57 sites (100%) fell short of 100. No agency front page in the sample was free of accessibility findings.
Security was second. The worst score was 48, the median 74 and the best 100. Fifty of 57 sites (87.7%) fell short of full marks.
Privacy had the widest spread. The worst score was 26, the lowest single score recorded in any dimension, while the median was 78 and the best 100. Forty-eight of 57 sites (84.2%) fell short of 100.
Technical quality was better. The worst score was 59, the median 94 and the best 100, with 43 of 57 sites (75.4%) short of full marks.
SEO was close to clean. The worst score was 95, the median 100 and the best 100. Only 7 of 57 sites (12.3%) fell short of 100, and none by more than five points.
Deceptive design was entirely clean on this sample. Every one of the 57 sites scored 100. Zero sites fell short. The interpretation of this result, and of the SEO result, is taken up in the discussion, because it is not the same as saying these agencies never use deceptive patterns.
Table 2: findings by how many sites carried them
Fifteen findings were present on at least five sites and are reported here, ordered by prevalence. For each, the dimension, the scanner's severity rating, the number and share of sites, and the number on which the finding was Confirmed are given. Twenty further findings appeared on fewer than five sites each and are withheld.
Low text contrast, an accessibility finding rated high severity, was present on 45 of 57 sites (79%), Confirmed on 43. It was the most widespread finding in the study.
A missing Content-Security-Policy header, a security finding rated high, was present on 40 of 57 sites (70%), Confirmed on all 40.
A link without a discernible name, an accessibility finding rated high (the scanner's rule identifier is Link Name), was present on 23 of 57 sites (40%), Confirmed on all 23.
No accessibility statement link, an accessibility finding rated medium, was present on 20 of 57 sites (35%), Confirmed on all 20.
Page content outside any landmark region, an accessibility finding rated medium (rule identifier Region), was present on 19 of 57 sites (33%), Confirmed on all 19.
A missing Strict-Transport-Security header, a security finding rated high, was present on 18 of 57 sites (32%), Confirmed on all 18.
A third-party cookie set before consent, a privacy finding rated medium, was present on 16 of 57 sites (28%), Confirmed on all 16.
A form field missing a label, an accessibility finding rated high, was present on 14 of 57 sites (25%), Confirmed on all 14.
A button or control missing an accessible name, an accessibility finding rated high, was present on 10 of 57 sites (18%), Confirmed on all 10.
A tracking script loading before consent, a privacy finding rated high, was present on 9 of 57 sites (16%), Confirmed on all 9.
A scrollable region that is not keyboard-focusable, an accessibility finding rated high (rule identifier Scrollable Region Focusable), was present on 7 of 57 sites (12%), Confirmed on all 7.
Sensitive form data submitted with a GET request, a security finding rated high, was present on 7 of 57 sites (12%), Confirmed on all 7.
Missing alt text, an accessibility finding rated high, was present on 6 of 57 sites (11%), Confirmed on all 6.
A skipped heading level, an accessibility finding rated medium, was present on 5 of 57 sites (9%), Confirmed on all 5.
Duplicate landmark labels, an accessibility finding rated medium (rule identifier Landmark Unique), was present on 5 of 57 sites (9%), Confirmed on all 5.
Of the fifteen reported findings, ten are accessibility findings, three are security and two are privacy. No deceptive design, technical quality or SEO finding reached the five-site threshold. Of the fifteen, fourteen were Confirmed on every site where they appeared. The single exception is low text contrast, Confirmed on 43 of its 45 sites, with the remaining two carrying a lower confidence label.
Three findings worth a paragraph each
Low text contrast deserves the first paragraph because it is both the most common finding here and the most common finding on the web at large. Its 79% prevalence on agency front pages sits close to the figure the WebAIM Million reports for the whole web, where low contrast text is the single most detected failure. That an agency, whose trade is in part visual design, ships a front page with text that fails the 4.5:1 contrast threshold is not a sign of carelessness so much as of how the failure occurs: it is almost always a shared colour token, a light grey chosen for a muted caption or a footer, applied everywhere at once. The two sites where the finding was rated Likely rather than Confirmed are instructive too; they are cases where text sat over an image or gradient and the exact background colour could not be resolved deterministically, which is precisely the situation an automated check should decline to call with certainty. For anyone who wants to fix the underlying cause, the practical steps are laid out in our guide to fixing low colour contrast.
The missing Content-Security-Policy header is the second paragraph, and it is a different kind of finding. Contrast is a design decision that went slightly wrong. A missing CSP is the absence of a control that was never set up. Seventy percent of agency front pages sent no CSP at all, which means that on those sites nothing constrains where scripts may load from, the primary browser-side defence against injected script. Eighteen of 57 (32%) also lacked HSTS, the header that keeps a browser on HTTPS after its first visit. These two headers are a configuration change on the server or CDN, not a code change, and the fact that a majority of professional sites lack the first of them suggests that response headers sit in a gap between the people who write the pages and the people who run the infrastructure. The detail of what each header should contain is covered in our complete 2026 checklist for HTTP security headers.
The consent findings are the third paragraph, because they carry legal weight that the others do not. A third-party cookie set before consent (16 sites, 28%) and a tracking script loading before consent (9 sites, 16%) together indicate that roughly a quarter of agency front pages begin tracking a visitor before that visitor has made a choice. Under the consent frameworks now in force in the EU and in a growing number of US states, that timing is the fact on which enforcement and litigation turn. It is also invisible to the site owner, who sees a banner and assumes the banner is doing something. The scanner sees the network requests and can tell that it is not. The regional rules that make this matter are set out in our region-by-region guide to cookie consent, and the litigation exposure that follows from pre-consent tracking, particularly in California, in our piece on reducing CIPA risk from session replay and pixels.
Discussion
The clean dimensions, read carefully
The most striking rows in Table 1 are the two clean ones. Every site scored 100 in deceptive design, and 50 of 57 scored 100 in SEO. It would be easy, and wrong, to read those rows as a finding that web agencies do not use deceptive patterns and have excellent SEO.
The correct reading starts from what was scanned. This study measured front pages only. The deceptive patterns regulators actually pursue, pre-checked consent boxes, drip pricing at checkout, hard-to-find cancellation, confirmshaming decline buttons, live in flows: a checkout, a subscription form, an account settings page. A web agency's front page has none of those. It is a portfolio and a contact form. The scanner's deceptive design rules had, on these pages, very little surface to test, and a score of 100 in that situation means "nothing found where there was little to find," not "this organisation is free of deceptive design." The same caution applies, less strongly, to SEO: front-page fundamentals such as a title and a description are exactly the things a professional site gets right, and the check does not reach the deeper pages where SEO problems tend to accumulate.
The honest statement, then, is that on a front-page-only scan, deceptive design and SEO produce almost no findings on agency sites, and that this result should not be extrapolated to the agencies' client work or to their own deeper pages. A fuller crawl, including any transactional flow the agency operates, would be needed before any claim about deceptive design could be made in either direction.
Comparison with the web at large
The accessibility results are directionally consistent with the WebAIM Million. According to the WebAIM Million 2026 report, 95.9% of the top one million home pages had a detectable WCAG failure. In this sample the equivalent figure is 100% of 57. The two studies use different rule sets and different scanners, so the numbers are not directly comparable, and the difference between 95.9% and 100% on a sample of 57 is within the range one would expect from that alone. What the comparison does support is the weaker claim: web agency front pages are not measurably better on accessibility than the web as a whole. That is the finding, and it is enough.
On the specific failures, the overlap is also close. WebAIM reports low contrast text as the most common failure across the million; it is the most common here. Missing form labels, missing alt text and empty links all appear in WebAIM's top categories; all three appear in Table 2. The agency sample is a smaller, cleaner mirror of the general web, not an exception to it.
There is no comparable public benchmark for security headers, consent timing or the other non-accessibility dimensions at this scale, which is one reason a six-dimension study is worth publishing at all. The 70% figure for missing CSP is, to the authors' knowledge, the first such number for a professional segment, and it should be treated as a first estimate rather than a settled fact until it is replicated.
The pattern across dimensions
Read together, the tables describe a segment that is professional where professionalism is visible and inconsistent where it is not. SEO and technical quality, the two dimensions closest to what a client can see and a search engine can reward, score well. Accessibility, security headers and consent timing, the three that are invisible until a disabled user, an attacker or a regulator arrives, score poorly. That is not a moral observation about agencies; it is a description of incentives. The dimensions that get checked get fixed. The dimensions that no one checks accumulate the failures this paper records.
Limitations
This section is deliberately detailed, because the value of a benchmark depends on the reader knowing exactly what it can and cannot support.
The sample is 57 sites. That is enough to report prevalence for findings that appear on a substantial fraction of sites, and the five-site floor on every published row keeps individual sites unidentifiable, but it is not enough to estimate rare findings, to compare subgroups of agencies, or to attach tight confidence intervals to the percentages. A finding reported at 12% (7 of 57) could plausibly sit anywhere between roughly 5% and 23% in the underlying population, and the paper should be read with that width in mind. The twenty findings withheld for appearing on fewer than five sites are exactly the ones this sample cannot say anything reliable about.
The sample is not random and the paper does not claim it is. The sites were chosen as web agencies, and the selection method, described in the reproducibility section, introduces a bias toward agencies that are discoverable and that permit automated visits. Agencies whose robots directives disallow scanning were, by design, not scanned, and there is no reason to assume their sites resemble those that were. The results describe this sample, and are offered as a first estimate for the segment, not as a measurement of all web agencies.
Only the front page was scanned. This is the limitation the discussion returns to most, and it cuts in both directions. It means findings that live on deeper pages, in transactional flows or behind a login are entirely absent from the data, which is why the clean deceptive design row cannot be read as the absence of deceptive design. It also means that a site whose front page is unusually good or bad is represented by that one page. The front page is a reasonable proxy for the care a site receives, and it is the page every visitor sees first, but it is not the site.
Automated detection has a ceiling. Every finding here is machine-detectable by construction. According to Deque's automated accessibility coverage report, automated testing identified 57.38% of accessibility issues by volume across a large dataset of real audits, and the widely cited estimate by number of WCAG criteria is lower still. The accessibility scores in Table 1 are therefore a floor on the true number of barriers, not a measure of it. A site that scored 96 is not 96% accessible; it is a site on which the automated rules found little. Keyboard traps, meaningless alt text, incoherent heading structure and colour used as the sole carrier of meaning are examples of failures a person finds and a scanner does not. We set out this boundary at length in our analysis of why a perfect Lighthouse score is not WCAG compliance, and it applies with equal force to this study.
A scan is a snapshot. Every figure describes a page as it was served on one of two days in September 2026. Pages change; a header added or a tracker removed the following week is not reflected here. A repeat of this study after an interval would be more informative than the study alone, and that is the natural next step.
Finally, the scoring is the scanner's own. Scores are calibrated against reference sites as described on our transparency page, so that comparable pages produce comparable numbers, but a score of 78 is a Sitejar score, not a universal one, and should not be compared numerically with scores from other tools. The findings in Table 2, each tied to a specific rule and a specific standard, are more portable than the dimension scores in Table 1.
What a confirmed finding means, and what it does not
Because the tables lean on the word Confirmed, it is worth being precise about it. A Confirmed finding is one where the rule is deterministic and the rendered page satisfies its condition without ambiguity. A response that contains no Content-Security-Policy header has no CSP; there is nothing to interpret. A form input with no label element, no aria-label and no aria-labelledby has no accessible label in the document; that is a fact about the DOM. Fourteen of the fifteen findings in Table 2 are of this kind, and on those, the site count and the Confirmed count are identical.
Low text contrast is the one finding where the two counts differ, 45 present and 43 Confirmed, and the gap is a feature of honest measurement rather than a flaw. Contrast is computed from a foreground colour and a background colour. When the background is a flat colour, the ratio is exact and the finding is Confirmed. When text sits over an image, a gradient or a semi-transparent layer, the effective background is not a single colour, and the scanner reports the failure as Likely and asks for a person to look. Two of the 45 sites were in that situation. A less careful tool would have counted them as failures without the qualification, or ignored them; the correct behaviour is to report the uncertainty, which is what the label is for.
The general point is that a Confirmed finding is a statement about the page as served, and only that. It says a barrier exists; it does not say how many users it affects, whether the site is in breach of any particular law, or how the site compares with a competitor. Those are inferences a reader may draw, with care, from many findings taken together. No single row in this paper supports them on its own.
What this data cannot tell you
Several claims a reader might be tempted to make are not supported, and the paper states them plainly so that it is not cited for them.
It cannot tell you which agencies are good or bad. No site is named, no row describes fewer than five sites, and the study was designed so that no individual site can be recovered from it. It cannot tell you that any site is or is not compliant with any law; compliance is a legal conclusion that depends on jurisdiction, on the whole site and on facts a scan cannot see. It cannot tell you that web agencies are worse than any other segment, because no other segment was measured with the same method at the same time; it can tell you they are not measurably better than the web at large on accessibility, which is a narrower claim. It cannot tell you anything about deceptive design in agency work, for the reasons given in the discussion. It cannot tell you the total number of accessibility barriers on any site, only the number a machine could find. And it cannot tell you what these sites look like today, because each was measured once, on a single day.
What it can tell you is more modest and, the authors think, more useful: on a specific set of 57 professional sites, on a specific pair of days, a specific and repeatable method found a specific set of failures at the rates recorded, and the most common of those failures are the same ones the largest public study finds across the whole web.
Implications and recommendations
For agencies, the tables suggest an order of work. The two most prevalent findings, low text contrast and a missing CSP, are also two of the cheapest to fix at the source: contrast is usually one colour token, and CSP and HSTS are a header on the server or CDN. Together they account for the failures on 79% and 70% of sites respectively. An agency that fixed only those two on its own site would, on this evidence, move from the middle of the sample toward the top. The accessible-name findings (links, buttons, form fields) come next, and are the ones most directly felt by screen reader users. The consent findings are last in prevalence but first in legal exposure, and the fix is a matter of timing rather than design: nothing non-essential should load until the visitor has chosen. Because every finding in Table 2 is a source-level defect, none is addressed by adding a script on top of the page, a point on which the authors have a stated position and which the data here does nothing to weaken. Agencies that want the same findings for client sites can run website audits for agencies with the scanner used here.
For anyone using this paper as a benchmark, the recommendation is to compare findings rather than scores. A site's own list of Confirmed findings can be set against Table 2 directly, because each row is a named rule against a named standard. Dimension scores are the scanner's summary and travel less well.
For the field, the recommendation is replication. A second sample of the same segment, scanned the same way six or twelve months on, would turn a snapshot into a trend, and a sample of a different professional segment scanned the same way would allow the comparison this paper cannot make. The method below is written to make both straightforward.
Reproducibility
To repeat this study, the following is sufficient.
Target selection. Assemble a list of web agency websites. The authors' list was compiled from public directories and search results for web agency services; the exact list is not published because publishing it would identify the sites, but the criterion was that the site presents itself as a web design or development agency. Any comparable list will do for a replication; the point is to fix the criterion before scanning, not to reproduce the identical set.
Scanning. For each site, run one read-only scan of the front page with a scanner that renders the page in a real browser and applies a fixed rule set across the six dimensions named above. The Sitejar scanner does this and is available without an account for up to five pages through the free scan on the homepage, or programmatically for larger samples through the API described in the developer documentation. Do not crawl, submit forms or log in. Honour each site's robots directives and record which sites were excluded on those grounds.
Deduplication. If a site is scanned more than once in the collection window, keep the earliest complete scan and discard the rest, so that each site contributes exactly one row. Exclude any scan that is a dry run, meaning a test in which known findings were replayed against a domain for validation.
Counting. For each dimension, record the worst, median and best score across sites, and the number of sites scoring below 100. For each finding, record the number of sites on which it appears and the number on which it is Confirmed. Withhold any finding present on fewer than five sites.
Reporting. Name no site. Publish no row describing fewer than five sites. State the scan dates, the sample size, the number of dry runs excluded, and the rule set version. Link every external figure to its source and label every internal figure as the authors' own.
The authors ran the scans for this paper on 16 and 25 September 2026 against 57 sites with the production rule set current on those dates. Twenty findings were withheld under the five-site rule. Zero dry runs were excluded. All 57 sites belong to third parties.
Frequently asked questions
How many websites were scanned?
Fifty-seven web agency websites, each belonging to a third party, each scanned once on its front page only, between 16 and 25 September 2026.
What is the single most common problem on web agency sites?
Low text contrast, present on 45 of 57 sites (79%) and Confirmed on 43. It is also the most common failure the WebAIM Million reports across the whole web, so agency sites are not an exception.
Do web agencies have better accessibility than other websites?
Not measurably. All 57 sites (100%) had at least one accessibility finding, compared with 95.9% of the top one million home pages in the WebAIM Million 2026. The two studies use different methods, so the small gap should not be over-read; the supportable conclusion is that agency front pages are not better than the web at large on accessibility.
Why did every site score 100 on deceptive design?
Because only front pages were scanned, and the deceptive patterns regulators pursue live in checkout, subscription and cancellation flows that a front page does not contain. The clean score means the scanner found nothing where there was little to test. It does not mean these agencies never use deceptive patterns, and the paper does not claim that.
What does "Confirmed" mean in the tables?
That the rule is deterministic and the failure is certain from the rendered page, such as a header that is absent or a field with no label in the DOM. Fourteen of the fifteen reported findings were Confirmed on every site where they appeared. Low contrast was Confirmed on 43 of 45, with two sites flagged Likely because the text sat over an image or gradient.
Can I find out which agencies were scanned?
No. No site is named, no published row describes fewer than five sites, and twenty rare findings were withheld specifically so that no site can be identified from the paper.
Are these scores comparable with Lighthouse or other tools?
The dimension scores are Sitejar's own and should not be compared numerically with other tools. The individual findings in Table 2 are more portable, because each is a named rule against a named standard. All automated tools, this one included, detect only a portion of real accessibility barriers.
How can I compare my own site with this benchmark?
Scan your front page with the same method and compare your list of Confirmed findings against Table 2, row by row. Comparing findings is more meaningful than comparing scores.
Will this study be repeated?
The method is published so that it can be. A repeat of the same segment after an interval would convert this snapshot into a trend, and is the natural next step.
Conclusion
On a read-only scan of 57 web agency front pages in September 2026, every site carried at least one accessibility finding, 87.7% lacked at least one security control the scanner checks, and 84.2% fell short in privacy. The most widespread individual failures were low text contrast on 79% of sites and a missing Content-Security-Policy header on 70%, followed by a cluster of accessible-name failures and, on roughly a quarter of sites, tracking that begins before consent. Deceptive design and SEO were clean, a result the paper attributes primarily to the limits of what a front page can show an automated check rather than to the agencies' practices, and declines to extrapolate. The accessibility findings mirror those of the WebAIM Million closely enough to support one conclusion with confidence: the sites of the professionals who build the web are not measurably better on accessibility than the web they build. Every failure recorded here is a source-level defect, and the two most common are among the cheapest to correct. The method is published in full so that the study can be repeated, extended to other segments, and turned from a snapshot into a trend. A follow-up benchmark of 300+ sites has since extended this sample with the same scanner.
References
WebAIM, The WebAIM Million 2026
Deque, The Automated Accessibility Coverage Report
Sitejar, Transparency: how scores are calibrated
This paper is educational and reflects Sitejar's point of view. It is not legal advice. For your specific situation, consult a qualified attorney.