Oh Dear....Is the UK version of Looksmart up to no good now?

Are you being charged for terms you haven't bought?

[Ross]

This is slightly confusing but here goes

Espotting sell their PPC results to BT Looksmart

NTLWorld and BT Openworld use BT Looksmart for their search results

My Espotting bill seemed much bigger than normal so checked further. It looks like one of the three companies above is doing partial matches on my PPC terms !

I've bid on "blue widgets" and I've been charged for searches from NTL & BTOpenworld for "free widgets" "blue films" "blue" "widgets" "blue-widgets.co.uk" and a host of others.

I've complained to Espotting (who have always been very helpful with any problems), and they're going to look into it. They've reassured me that I'll get a refund for any incorrect clicks. I doubt that they'll return ALL monies to ALL users unless you get on to them.

Check your referrer logs and search terms if you're using Espotting PPC, you could be in for a nasty surprise.

[IanTurner]

Please let us know the outcome

[NFFC]

Been happening for a long time, spotted I think by makemetop, Barry will know the dates.

[Ross]

Bloody Hell Notts....you're right. My keyword stats only go back as far as July and it was happening then.

So Espotting are charging us for clicks on terms that we haven't bid on, is that correct?. If that's so, what has anyone done about it?. Has everyone else stopped using Espotting, leaving me to turn the lights out on the way out?.

[Abrexa_UK]

As an Espotting partner, I can assure you that they were very clear on the rules of using their search results in combination with site searches. If someone is using partial matching, as far as I am aware, this is against the Espotting rules for partners.

However, if BTLooksmart hasn't passed on these requirements to their search partners then it is BTLooksmart at fault, not NTL and BTOPenworld.

Partial matching might well be a useful tool, as demonstrated by Overture and Google, but I would expect this to be an option rather than mandatory.

[Ross]

There's a BIG difference between Overture / Google partial matching and this Espotting / Looksmart version.

In Overture your search term must match exactly what the user types in, in other words your bid on "blue widgets" will match a search for that phrase but not for a search on "cheap blue widgets"

In the default mode of a Google search if you bid on "blue widgets" they will match searches for "blue widgets" "cheap blue widgets" "free green or blue widgets" etc.

Bidding on "blue widgets" with Espotting is producing a match from BT Looksmart on "widgets" "blue" "green widgets" "blue films" in other words they are producing a keyword match when ANY ONE of the words input by the user matches ANY word in your search term.

On Friday afternoon Espotting agreed that BT Looksmart were doing partial keyword matching. They are now going to stop my bids being passed to Looksmart, and they are going to attempt to work out how much is due to be refunded....a massive job when you consider that this seems to have been going on since April.

[Abrexa_UK]

Are Espotting looking at this same process for everyone, or is it on a request-only basis? I imagine that this is a fairly complicated job, especially if they have to refund everyone. I wonder who will be footing the bill - BTLooksmart or Espotting.

This does seem to be a major step away from proper keyphrase targeting - are Espotting happy with BTLooksmart doing this, or is it something that they were unaware of?

[makemetop]

Yes, this has been going on since the beginning of April. It has been brought to the attention of both Espotting and LookSmart by phone and in person several times since then. It is really an LS problem and if they haven't done anything about it by now - I think it is pretty unlikely that they will.

They have always said that they will refund anyone if they can isolate the phrases that they are paying for which have not been bid on - but how many people are going to do that? ;)

[Ross]

I'm sure that they know about it and I'm equally sure that they're happy to let it continue !. I didn't get the impression that they were about to give all of their customers a refund ;)

Here's a little transcript of part of our conversation :

Me: I've got clicks for blue widgets on a search for green widgets
Him: Don't you bid on blue widgets
Me: Yes but on a blue widgets search
Him: So can you tell what search term has been used in the searches
Me: Yes
Him: Oh !

[omegadm]

We have used espotting since the begining of the year and I haven't noticed anything like this.

All our bid terms use tracking url's and we analyse log files using WebTrends. Each month we compare the WebTrends reports with that of invoices from eSpotting, Overture etc and they are usually within 10%.

HTH.

Best regards, Brian

[Ross]

Brian,

In your logs you are looking for a referrer of "synd-uk.looksmart.co.uk". Look at the search terms entered for any entries using the Espotting tracking URL. I GUARANTEE that you've been paying for search terms that you didn't bid on.

My Espotting search terms should be blocked to Looksmart from tomorrow onwards and Espotting have offered to refund the costs of a SUBSTANTIAL proportion of my clicks to Looksmart since April of this year ;)

[omegadm]

Ross

I have looked through July and August's logfiles (thanks!) for a Spanish car hire client - I have only found 1 suspicious entry:

62.254.0.6 Malaga-car-hire.com - - [06/Aug/2002:20:51:56 +0100] "GET /?source=espotting HTTP/1.1" 200 4005 "http://synd-uk.looksmart.co.uk/synd-yal/SearchResults.jsp?key=car+hire+europe&lan=uk&cp=3&skip=30&se=0&sir=40&cs=2&st=0&rp=0" "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 5.1)" "-"

The client does not bid on "car hire europe" - way too general
But that's it. Total LS click throughs = 107...

Best regards, Brian

[Ross]

Brian,

You don't seem to have a problem there. I was getting a much higher percentage of bad clicks but I can't think of any reason for the discrepancy between our two accounts. I'm sorry to have put you to such trouble, it's only a 2 second job for me to get a report showing referrer, search term, number of clicks etc and I was forgetting how laborious it is to wade through raw logs.

I'd still advise you to keep an eye on that particular referrer.

Best Regards

Ross

[jamsy]

hmmm looks like time to remove the dust from the old log files this weekend and then call espotting on Monday :)

thanks for the info on this

[omegadm]

Ross - nyet problyem

We use WebTrends so it is also easy for me to look at the referrer/keywords reports per search engine. However what you were saying wasn't showing in the reports so I just did a grep on the log file. Didn't take that long :))

Keep us informed as to what happens. Just because I don't see it (3 clients using LS UK), doesn't mean it isn't happening....

Best regards, Brian