Some pages don't have a title set in GSA data. This is a slowwww work around while that is remedied.

gistfile1.cs

this.PageName = node.SelectSingleNode("T") != null ? EPiServer.Core.Html.TextIndexer.StripHtml(node.SelectSingleNode("T").InnerXml.Trim().HtmlDecode(),200) : string.Empty;
if (this.PageName.IsNullOrWhiteSpace())
{
    this.PageName = "( No Title )";
    var UseWebRequestToResolveSupportTitle =
        WebConfigurationManager.AppSettings["UseWebRequestToResolveSupportTitle"] != null && Convert.ToBoolean(WebConfigurationManager.AppSettings["UseWebRequestToResolveSupportTitle"]);
    if (UseWebRequestToResolveSupportTitle)
    {
        var cachedTitle =
            (string)EPiServer.CacheManager.Get(SupportCacheKey);
        if (string.IsNullOrEmpty(cachedTitle))
        {
            // Make the WebClient we'll be using able to handle support's bad cert
            var sslFailureCallback = new RemoteCertificateValidationCallback(delegate { return true; });
            ServicePointManager.ServerCertificateValidationCallback += sslFailureCallback;
                        
            var client = new WebClient();
            var html = client.DownloadString(this.PageURL);
                        
            // Retrieve just the title
            var title = Regex.Match(html, @"\<title\b[^>]*\>\s*(?<Title>[\s\S]*?)\</title\>", RegexOptions.IgnoreCase).Groups["Title"].Value;
            EPiServer.CacheManager.Add(SupportCacheKey,title);
            cachedTitle = title;
        }
        this.PageName = cachedTitle;
    }
}