Bill On Business

How warnings of being “Penalized by Google” can sometimes be the online equivilent of the big bad wolf.

Clients are often be told by SEO consultants that a wide range of technical issues will lead to their website being “penalized by search engines”, notably Google.

The fundamental message this sends is inaccurate. This message can lead to dogged adherence to change without good business reason or SEO advice being disregarded as scaremongering.

There is only a small range of activities that search engines actively penalize, yet many reasons for poor performance that could appear to be penalties if cruedely interpreted.

This is not to say that search engines do not heavily favour certain sites for reasons other than pure query-document match relevancy but thats another post altogether.

Being organized and conforming to expected standards = Good SEO (blazing a trail is to be encouraged but be aware search engines take a while to catch up, and only follow the trends that have significant human user traffic already)

So here is my offline metaphor to explain why the penalties are not always penaties. Lets compare SEO and crossing international borders.

Event: Imagine arriving at JFK to change flights to San Francisco, on a last minute bargain single ticket & forgetting the address of your hotel. Casually mention your girlfriend lives in SF and it’s a romantic surprise visit. Several hours later, as you leave secondary interviews, you’ll have missed your onward flight with no recourse, face overnight hotels at your expense, a 24 hour delay and the costs of an airport issued ticket home to reassure US Immigration. Your Intentions? Good, Your Outcomes?, Bad.

But it’s important to realise none of the inconvenience “caused by US Immigration” was motivated by a desire to punish.

Analysis: The passenger was not trusted to perform as expected, had failed to conform to system requirements and had to be processed in a secondary manner requiring multiple research activities from a resource constrained team and systems.

Unlikely as it seems this scenario happens every day at airports around the world.

Event: The online equivalent of this story is a website re-launching without a URL cutover strategy, integrating large volumes of third party feed into it’s pages, not publishing a valid sitemap at an easily found location, with hard to interpret dynamic URLs and no inbound links as years of link equity is lost to 404 error pages. Site traffic drops precipitously for 90 days while the programme sponsor rides out the turbulence spending valuable time reassuring senior stakeholders everything is going to be ok and there will be a positive ROI. Of course it normally does, but it’s been stressful.

None of the inconvenience “caused by Google” was motivated by a desire to punish.

Analysis: The website was not trusted to perform as expected, had failed to conform to expected standards and had to be processed in a secondary manner requiring multiple research activities from a resource constrained team and systems.

Unlikely as it seems this scenario happens every day at huge online companies around the world.

Detect Duplicate Content – Think Like a Search Engine

A client’s copywriter recently went to lengths to show how he could reword copy to avoid duplicate content challenges by merely moving paragraphs and sentences around.

The copywriter was certain any search system could be outsmarted by moving text positions and syntax around because that would be theoretically so complex to monitor on a large scale that surely a search engine would not have the resources to cope.

Without going into the depths of duplicate content systems, content , scalable search systems or if he is right that you can blag it… it is worth thinking like a search engine for a minute but keeping things simple.

Rather than use search engine results or search systems to examine his plan, I resorted to the suitable unassuming MS Word.

By using a MS Word macro to calculate word frequency and total unique word count you can quickly see shortcuts to calculating document uniqueness without stopping by MIT for 4 years.

I compared two identical documents, with word orders shifted, paragraphs moved and sentences inverted etc. I crazied the format up until mr copywriter was good and happy.

The output of the macro was identical results for both documents.

1171 unique words per document and the 18 keywords listed below occuring in exactly the same frequency order in both documents.

To have exact unique keyword volumes and identical counts on keywords between two 3854 word documents is statistically very unlikely. So without a single Search system to hand, it can be seen how easy it is to compare document duplication.

Count   Keyword

83         search

65         seo

55         company-name-removed

51         in

49         project

48         company-brand-removed

41         we

33         team

32         at

29         as

29         will

28         delivery

28         this

27         on

26         our

26         site

25         with

21         performance

The Macro Used: Thanks to Allen Wyatt’s Word Tips – Count Unique Word Occurrences

http://word.tips.net/Pages/T001833_Generating_a_Count_of_Word_Occurrences.html

Sub FindWords()
    Dim sResponse As String
    Dim iCount As Integer

    ‘ Input different words until the user clicks cancel
    Do
        ‘ Identify the word to count
        sResponse = InputBox( _
          Prompt:=”What word do you want to count?”, _
          Title:=”Count Words”, Default:=”")
   
        If sResponse > “” Then
            ‘ Set the counter to zero for each loop
            iCount = 0
            Application.ScreenUpdating = False
            With Selection
                .HomeKey Unit:=wdStory
                With .Find
                    .ClearFormatting
                    .Text = sResponse
                    ‘ Loop until Word can no longer
                    ‘ find the search string and
                    ‘ count each instance
                    Do While .Execute
                        iCount = iCount + 1
                        Selection.MoveRight
                    Loop
                End With
                ‘ show the number of occurences
                MsgBox sResponse & ” appears ” & iCount & ” times”
            End With
            Application.ScreenUpdating = True
        End If
    Loop While sResponse <> “”
End Sub
</pre><p>If you want to determine all the unique words in a document, along with how many times each of them appears in the document, then a different approach is needed. The following VBA macro will do just that.</p><pre>
Sub WordFrequency()
    Const maxwords = 9000          ‘Maximum unique words allowed
    Dim SingleWord As String       ‘Raw word pulled from doc
    Dim Words(maxwords) As String  ‘Array to hold unique words
    Dim Freq(maxwords) As Integer  ‘Frequency counter for unique words
    Dim WordNum As Integer         ‘Number of unique words
    Dim ByFreq As Boolean          ‘Flag for sorting order
    Dim ttlwds As Long             ‘Total words in the document
    Dim Excludes As String         ‘Words to be excluded
    Dim Found As Boolean           ‘Temporary flag
    Dim j, k, l, Temp As Integer   ‘Temporary variables
    Dim ans As String              ‘How user wants to sort results
    Dim tword As String            ‘

    ‘ Set up excluded words
    Excludes = “[the][a][of][is][to][for][by][be][and][are]“

    ‘ Find out how to sort
    ByFreq = True
    ans = InputBox(”Sort by WORD or by FREQ?”, “Sort order”, “WORD”)
    If ans = “” Then End
    If UCase(ans) = “WORD” Then
        ByFreq = False
    End If
   
    Selection.HomeKey Unit:=wdStory
    System.Cursor = wdCursorWait
    WordNum = 0
    ttlwds = ActiveDocument.Words.Count

    ‘ Control the repeat
    For Each aword In ActiveDocument.Words
        SingleWord = Trim(LCase(aword))
        ‘Out of range?
        If SingleWord < “a” Or SingleWord > “z” Then
            SingleWord = “”
        End If
        ‘On exclude list?
        If InStr(Excludes, “[" & SingleWord & "]“) Then
            SingleWord = “”
        End If
        If Len(SingleWord) > 0 Then
            Found = False
            For j = 1 To WordNum
                If Words(j) = SingleWord Then
                    Freq(j) = Freq(j) + 1
                    Found = True
                    Exit For
                End If
            Next j
            If Not Found Then
                WordNum = WordNum + 1
                Words(WordNum) = SingleWord
                Freq(WordNum) = 1
            End If
            If WordNum > maxwords – 1 Then
                j = MsgBox(”Too many words.”, vbOKOnly)
                Exit For
            End If
        End If
        ttlwds = ttlwds – 1
        StatusBar = “Remaining: ” & ttlwds & “, Unique: ” & WordNum
    Next aword

    ‘ Now sort it into word order
    For j = 1 To WordNum – 1
        k = j
        For l = j + 1 To WordNum
            If (Not ByFreq And Words(l) < Words(k)) _
              Or (ByFreq And Freq(l) > Freq(k)) Then k = l
        Next l
        If k <> j Then
            tword = Words(j)
            Words(j) = Words(k)
            Words(k) = tword
            Temp = Freq(j)
            Freq(j) = Freq(k)
            Freq(k) = Temp
        End If
        StatusBar = “Sorting: ” & WordNum – j
    Next j

    ‘ Now write out the results
    tmpName = ActiveDocument.AttachedTemplate.FullName
    Documents.Add Template:=tmpName, NewTemplate:=False
    Selection.ParagraphFormat.TabStops.ClearAll
    With Selection
        For j = 1 To WordNum
            .TypeText Text:=Trim(Str(Freq(j))) _
              & vbTab & Words(j) & vbCrLf
        Next j
    End With
    System.Cursor = wdCursorNormal
    j = MsgBox(”There were ” & Trim(Str(WordNum)) & _
      ” different words “, vbOKOnly, “Finished”)
End Sub