Have we shut ourselves up in an ivory tower?

Question

Writing up the bio of an Abbess of Shaftesbury I wanted to observe that her nephew Henry VII defeated Richard III and made himself King. As an extra I always link to other people named in the text if only to demonstrate that we have profiles for them. Henry was easy as I'd already stated the relationship. Richard was a bit harder but I knew enough about the relationships to find him. But, that's me, I already know this stuff. Try looking up Richard III from the position of someone with little historical and no Wikitree knowledge. I also wanted to link to Edward the Martyr, you could try that.

Google Richard III. Wikitree does not come up. It should. Surely we can and should do better than this.

Comments

Thanks for pointing to this problem again. IMHO it connects to some basic themes which go back years, for example that Wikitree invested itself so much into the idea of searching by LNAB, and thus defining people by LNABs. It will always be a dead-end but in the meantime I thinking proving more obvious alternative ways to search can help. In this particular case maybe a small Plantagenet focussed project or sub-project could also set-up information pages with lots of good linking.

---

Good SEO is about more than just linking these days, as I have found from creating my own website.

Unique content, meta descriptions, High quality pictures, Alt text, position of keyword, social network sharing- they are all incredibly important too.

---

Maybe the Jamie wikitech tag is appropriate here?

Answer 1

If other websites do not link to Richard’s WikiTree profile, that will be a big factor. There are loads of other webpages in competition with WikiTree.

Comments

I din't know there were loads of web pages in competition with WikiTree. I can only think of two others like it: WeRelate.org and FamilyPedia.org (I think that's right).  Oh, three: familysearch.org is trying to a do a single tree site, but they don't appear to have near the ability to manage all the dupes the way WT does.

---

Actually, geni.com is also putting together a connected world tree - and as of right now they claim to have connected 147 million people to their World Family Tree and a total of more than 200 million profiles (that is, about seven times the size as Wikitree). Given that they are connected with MyHeritage, which is the preferred site for European genealogists (as opposed to ancestry.com), I found that they have, for example, much better coverage of Norwegians than Wikitree.

---

I think Katie is exactly right. The biggest factor in whether we appear in search results for Richard III is whether other websites link to us when they talk about Richard III.
https://www.wikitree.com/g2g/885192/will-you-help-spread-the-word-about-wikitree

Geni has a size advantage, and that matters here. They're also owned by a website that spends millions on advertising and PR, some of which passes through to Geni.

---

I didn't mean only websites with information very similar to WikiTree's. When someone Googles "Richard III" or even something more specific like "Richard III family tree" there are lots of results, and they are all in competition with WikiTree.

---

I don't think google optimizing was the only point? For a normal user, with google it is EASIER to find medieval famous people on wikitree than with wikitree internal searching.

Answer 2

If there were a Space page, if people could find the Space page, it could have a list of: Kings of England and the Wiki surname for each king, plus ID#.

And for other kingdoms, as well, of course.

Comments

Several years ago, when I did a complete overhaul of the European Aristocrats Project (and sub-divided it into five sub-projects), I also did a complete overhaul of the categories.  I had been asked for a place where people could go to look for royals and nobles.  We now have the top-level category Aristocracy and Nobility with lots of sub-categories (such as Royalty) and subs of those subs.  If people want more lists, they can work with the categorization project to set up a new category.  You would also want to coordinate with the project overseeing the profile(s).  While in the past the EuroAristo project has overseen the royal, noble, and aristocratic profiles, we are slowly handing these over to the various country projects (so England handles English nobility, Ireland handles theirs, etc).  Ultimately EuroAristo will only oversee profiles where there is no other project to manage them.

---

Thanks for all your work, Darlene.

Lois's idea is a good one. The more pages that reference the profiles, the more easily they can be found, directly and via Google searches.

---

I agree.  We have the 'backbone' in place for it.  I had set up numerous categories such as British Nobility with pages for lists of  British, English, Irish, and Scottish monarchs.  I'll reach out to the England project about it...  

Edited to add (also Scotland, Ireland, Wales, Germany, France...)

---

Isabelle has already set up a space page for French monarchs here.  Scotland has a page, but no work has yet been done on it.  Ireland has several pages;  here's one.

---

Thanks to Darlene pointing out this discussion, the Scotland Project will make sure we have a page for our monarchs, as well.  This is a great idea for drawing attention to our Profiles!

---

Thanks Darlene for setting that up, It's been very useful for me.

If any one is interested in Irish Nobility, we have pages for High_Kings_of_Ireland and Irish_Kingdoms which has links to the pages to the Kings of those Kingdoms. Because there were over 150 Kingdoms at any one time for over 1,200 years, the list is obviously not completed yet. As someone has on his Emails, So much Wiki and so little time. 

---

Chris, you wrote:

The more pages that reference the profiles, the more easily they can be found, directly and via Google searches.

Why would people link a space (page) if this information exists already in Wikipedia (the list of Kings for a specific country)?

I understand the problem but I don't think that creating a summary page on WikiTree is solving the problem that WikiTree isn't linked in other websites.

Richard III should come up when someone searches for it, as it seems to do with Geni.com - I suggest you rather check their SEO to find out why that is and what is missing in the SEO of WikiTree.

That would have a much higher chance of success IMO.

Just my two cents ;-)

---

Andreas, you seem to be contradicting yourself. Yes, we need more external websites linking to WikiTree. More sites link to Geni than WikiTree. On-page SEO has little to do with it.

Answer 3

When I searched I didn't see ancestry.com come up and I'm sure there is at least one profile for Richard III on their site.

Answer 4

I'm afraid I don't quite see the problem. This is a site for genealogy, not for history. When I google "Richard III", among the top sites are the Wikipedia entry, the entry from the Encyclopedia Britannica, and the entry from the Richard III Society (as well as sites relating to Shakespeare's play) - which I believe is exactly as it should be, as those are written by historians and go into good detail (much more detail than the Wikitree entry).

Now, the fact that a search for "Richard III" "genealogy" brings up the geni.com page and a couple of geneanet pages but not the Wikitree page - that may indeed be a matter for concern.

Comments

The problem is that if you want to access the genealogy of Richard III, you can't do it without first knowing his Wiki-surname, which seems to be York now, although I think it used to be Plantagenet. You can't search for Richard III here without knowing this, so people often Google for him to find him here because the search function won't do it.

So if Google doesn't turn him up, he's rather unfindable.

---

Ah, I see, so the problem is finding him internally in Wikitree.

---

That is certainly a problem, yes. Because Wiki-surnames for people like kings and chieftans tend to be counter-intuitive or obscure.

---

I can think of a solution to the problem but it would probably be a more complex task than the team would be willing to take on because it requires a database architecture change.  For whatever it may be worth:

Add a new field for keywords that would work kind of like the other last names and nicknames fields do.  We could add any terms we think people might use when trying to look up the profile, separated by commas.  Then add keywords to the search engine.  If that were in place, you could enter Richard III in the keywords field of his profile and then that profile would come up as a search result for that term.

---

If I google "Richard III WikiTree" the first result corresponds to the correct WikiTree profile. Maybe the results are different when you search from a different place (I'm in France). - There are WikiTree profiles that won't come up at all, even with the WikiTree word in the search.

---

Gaile, that's an interesting idea, but I think Google would essentially ignore the keywords. Words in the text, in a natural language way, would probably have more weight. To really make a difference, it would have to be the headline.

Probably the most powerful thing that could be done: Other websites linking to the profile with "Richard III" as the link anchor text.

---

The Wikidata page for anyone who has a Wikipedia entry should also have the field completed for WikiTree ID. That could affect the Google index.

---

Chris, I didn't mean the new field as something google would necessarily pick up.  I intended it as a field to be added to WikiTree's own search algorithm - in addition to the name fields now included.  I suppose that, if you're interested in optimizing the profile title for google searching, you might consider adding the keywords field to the profile's title, perhaps parenthetically, or perhaps using a white font so that it does not display at the top of the profile page.

Actually, because of the magnitude of the developmental task involved in doing this, I really didn't expect you to take the idea seriously - I wouldn't think the value added would be even close to considering undertaking all the work needed to do it.

Answer 5

I went to Find - Search, then scrolled right down to the bottom where it says

4. Search for any text

Then put in "Richard III" (without the quotes) and, in the Google page which came up, WikiTree was the first entry (after the paid ads).  Took me all of .24 seconds.

Comments

Maybe everybody who read this post just searched to try it, and it weighed the algorithm.

---

Just did a bit of 'ego surfing' and did the same thing i.e. Find-Search-Search for Any Text and put Ros Haywood in there.  The first 25 results were me (and took .21 seconds), then the next few PAGES of results were Haywood on WikiTree.

---

Maybe if you search from a Google page, it won't appear.  But if you search from WikiTree's search page which *uses* Google, then it's fine.

I tried searching for Ros Haywood from a blank page at Google, and WikiTree was the second entry (the first was LinkedIn, which I haven't used for years

---

Yup, it worked. Clearly I'm not the only one who didn't know to do that.

---

Perhaps that free form search box should be at the top, and not the bottom (or both?)?

Answer 6

In some ways this is a technical problem - if you search in Google for the phrase "Richard (York) of England KG" you get Wikitree because that's the title of the page.

What needs to happen is we need some ability to change the label, which would power the title tag in the meta which is indexed by Goggle. But because this is generated from the name fields we can't make any changes to how the title is generated.

Andrew in a comment laments the LNAB focus which is part of the problem, but allowing profiles to change the page title seems to me surely something the tech team could work on. And like the PPP flag it could be enabled by a project since for most people you don't need to customize the label like this.

Comments

Well, you could propose having more fields for alternative names, but we have those, and all those fields together, with all their various rules based on the idea that LNAB is the key to the database, end up giving us very un-natural names for most medieval people. Best for googling are probably those awful profiles that try to fit everything into single fields.

I think as a general rule it might help a bit to have a policy to place natural looking names prominently within profiles.

Answer 7

I googled Richard III Genealogy and his wikitree page came up, albeit on the second page.

Answer 8

I did a private mode search using Safari and the search engine Duck Duck Go and no results for Richard III on WikiTree.

However, when I searched for Tommy Buch, the following WikiTree entry was found on the second page.

https://www.wikitree.com/g2g/1026687/tommy-buch-is-a-wonderful-wikitreer

So, WikiTree is showing up in search results with an April 29th post date.

On the third page of search results (listed as item #20) I found the following link:

https://www.wikitree.com/g2g/user/Buch-358

with the following description:

“User Tommy Buch Activity by Tommy Buch. Score: 36,570 points (ranked # 469): Title: G2G6 Mach 3: Questions: 39 (12 with best answer chosen): Answers:”

And then 9 entries down on the 3rd page I saw my dad’s profile (https://www.wikitree.com/wiki/Buch-359) with me listed as a son. It has the following as a description: “Husband of Charlotte Maurine (Gray) Buch — married 26 Jun 1954 in Bastrop, Morehouse, Louisiana, United States Descendants Father of [private son (1950s - unknown) ], [private son (1960s - unknown) ] and Tommy Buch”

And last I found my name listed on my business website on the fifth page of search results.

Duck Duck Go only gave 5 pages of search results.

Maybe more current dated entries are showing up over aged ones.

Answer 9

Great point C. This is about more than just using the website nicely. It also is about business (Wikitree is paid by ads for which SEO is crucial) and also about attracting people from other countries/in other languages.

I got the request by the team to post my ideas on G2G on improving the use of the AKA/Nickname in SEO (= improve the chance a search machine shows the pages in the first 5 hits), which is much more important than a formal name for many profiles, so here is the text from that e-mail.

---

An example to describe the problem is here https://www.wikitree.com/wiki/Beij-3
She is world famous in the Netherlands under her stage name, which is put under the nickname or alias. It is also present in the profile.
Hardly anybody knows her by her formal name, which of course is used to build her tree.

Looking at the rendered page, it starts with:
<link rel="canonical" itemprop="url" href="https://www.wikitree.com/wiki/Beij-3"/>
<title>Maria (Beij) Servaes (1919-1998) | WikiTree FREE Family Tree</title>
<meta name="description" content="Is this your ancestor? Explore genealogy for Mary (Beij) Servaes born 1919 Leiden, South Holland, Netherlands died 1998 Horn, Limburg, Netherlands including ancestors + 1 photos + more in the free family tree community."/>
<meta name="keywords" content="family tree of Mary Beij, Mary Servaes genealogy"/>

The description does not take the nick names into account. It is a rather general text asking people to help out.

I reckon most family members of non-notables will also use the nick name (I have examples from my own tree).

Is there a reason not to include the nicknames in the description? Or make the title of the page use the same as is shown under the vitals class div: <div class="VITALS"><span class="large">

Another thing is the apparent trouble Google has with the site map of Wikitree. Do you provide one for them? That would increase the chance that Wikitree is included in the smart page. See eg the google search result for 'Maarten Biesheuvel' (very famous writer in the Netherlands).

https://nl.wikipedia.org/wiki/J.M.A._Biesheuvel
generates the smart page
Google/startpage/Bing seaarch on "maarten biesheuvel genealogy" (in English or Dutch) does not show the wikitree page on the first two pages of results. It does show geniweb/openarch/werelate though...

---

Just my thoughts how Wikitree could be improved to attract more search hits and more non-USA inhabitant.