Proactive Chat

[Julie A Arendt]

Hello,

Is anyone out there using proactive chat (pop-up invitations that appear after a certain amount of time on the page) in LIbraryH3lp? 

My library is preparing to dip our toes into it, and I'm looking for any advice someone who has done it before can give. 

Sincerely,

Julie Arendt

Julie Arendt

Science and Engineering Research Librarian

121J Cabell Library | VCU Libraries

901 Park Ave | PO Box 842033

Richmond, VA 23284-2033

(804) 828-1539 | jaar...@vcu.edu

[Gwen Exner]

Hi everyone,

I wrote this up to answer Julie's question, but it occurred to me it might be helpful for other people too.

Cheers,

Gwen Exner

Chatstaff

---------- Forwarded message ----------
From: Gwen Exner <gwen...@gmail.com>
Date: Fri, Jun 3, 2016 at 8:01 PM
Subject: Re: [libraryh3lp] Proactive Chat
To: Julie A Arendt <jaar...@vcu.edu>

Hi Julie,

Expect a huge increase in traffic, pretty abruptly.

I'm with Chatstaff, and here's a few charts for one of our clients, showing what adding proactive in late April/early May 2014 did to their traffic.

Chats per month, divided by normal and proactive queues.

As you can see, the addition of proactive turned historically "slow" months up to traffic levels similar to historically "busy" ones, and shot busy ones through the roof.  This actually makes the start of summer a great time to try this, so your staff has time to see how busy it will end up being.

12-month moving average:

When viewed as an average over 12 months, traffic was fairly steady until they added proactive.  As you can see it not only brought a chunk of traffic through itself it also notably increased traffic on the regular queue as people became used to thinking about it at a resource.

Contact hours per month:

	2012	2013	2014	2015	2016
1	24	27	24	79	43
2	25	23	27	97	92
3	20	26	35	123	83
4	26	41	32	96	84
5	11	14	19	34
6	22	15	51	53
7	22	28	64	33
8	16	15	40	32
9	26	24	106	106
10	33	29	100	129
11	31	30	119	107
12	14	19	56	46

Last but not least, here's a table showing the shift in contact hours.  The increase here isn't in direct proportion to the number of chats because proactive chats also seem to take a little bit longer on average.  Over the last two years, the normal queue has averaged averaged 8.2 minutes per chat, and proactive has averaged 9.6 minutes per chat -- an 18% increase per chat on top of the . 

I hope this is helpful!  I'd also love to hear any information you get from other people.

Best regards,

Gwen Exner

--
You received this message because you are subscribed to the Google Groups "libraryh3lp" group.
To unsubscribe from this group and stop receiving emails from it, send an email to libraryh3lp...@googlegroups.com.
To post to this group, send email to libra...@googlegroups.com.
Visit this group at https://groups.google.com/group/libraryh3lp.
For more options, visit https://groups.google.com/d/optout.

[Michael Peper]

We had a number of concerns before we started our proactive service at Duke, but it hasn't had a dramatic impact on our service in any way.  We were worried about 1) annoying patrons with unwanted windows and 2) creating a surge of questions that our staff couldn't handle.  We've had a couple of patrons say that they didn't want to see a chat invitation, but most people who mentioned it via chat thought it was a nice service.  

As for the traffic, we have not seen the surge that Gwen describes.  I don't have a nice chart (but I can share details with Julie or others if you're interest), but here is the summary.  We started our proactive service in January 2015.  Before that time, we were getting around 600-700 chats per month during the fall and spring semesters.  During these same peak times since 2015, we get about 60 chats via proactive queue and chats to our non-proactive chats has decreased to about 500-600 per month. So, the proactive queue has about a 10% share of our total chats.  Our general service has been pretty busy, though, so we may have been saturated already, while if you had some untapped demand/need, proactive might generate a lot more business.

We went through some design/technical decisions before launching, but they were fairly easily managed by our IT staff.  Again, I can provide details if you're interested.

I hope that all helps!

Michael

[phil blank]

Great thread everyone. I'm also with Chatstaff and used to work at Duke way back when they first set up their first VR. Michael, I may be wrong but I think Duke's proactive chat is only triggered on search result pages? If so, that's a very likely explanation for the lack of similar traffic. In fact, it's an impressive # considering the other options for students to click on that are available on those pages. But it brings up an interesting facet of proactive- its most attractive feature may be that, because it doesn't compete for webpage space in the traditional way, it can be deployed on more top level pages. In our samples, it often out-performs the non-proactive VR options on the same page, but not always.

I actually don't have much data on non-top level proactive traffic, but I imagine placement would vary results just as it does for all other widgets (Duke's data would be interesting as a test case if my understanding of where it is triggered was correct). But, proactive, in general, does give libraries a bit more flexibility with placement, which may explain part of its impact. 

Phil Blank

[Michael Peper]

Yes, you're right.  We have the chat invitation appear only on our page with "bento" search results, which is where our default search box goes.  Where you implement the proactive service would have a big impact on how much traffic it gets.  The search results page gets plenty of traffic, though. 

[Jennifer Heise]

P.S. we use mostly embedded chat windows other than that proactive set; we find we get better results with embedded chat than 'click on icon to chat'.

On Mon, Jun 6, 2016 at 1:22 PM, Jennifer Heise <jhe...@drew.edu> wrote:

We've enabled Proactive chat on our ILLiad pages, and we don't get significant numbers of chats from it-- our ILL staff login and monitor the queue as well as the reference librarian. But we figure if it helps people who need it, that's good enough.

We may try proactive chat on one of our main pages next fall.

On Fri, Jun 3, 2016 at 10:58 AM, Julie A Arendt <jaar...@vcu.edu> wrote:

--

You received this message because you are subscribed to the Google Groups "libraryh3lp" group.
To unsubscribe from this group and stop receiving emails from it, send an email to libraryh3lp...@googlegroups.com.
To post to this group, send email to libra...@googlegroups.com.
Visit this group at https://groups.google.com/group/libraryh3lp.
For more options, visit https://groups.google.com/d/optout.

--

Jennifer Heise, jhe...@drew.edu

Reference/Web Manager, Drew University Library

973-408-3675

"Comments are free, but facts are on expenses." -- Tom Stoppard

--

Jennifer Heise, jhe...@drew.edu

Reference/Web Manager, Drew University Library

973-408-3675

"Comments are free, but facts are on expenses." -- Tom Stoppard

[Jennifer Heise]

We've enabled Proactive chat on our ILLiad pages, and we don't get significant numbers of chats from it-- our ILL staff login and monitor the queue as well as the reference librarian. But we figure if it helps people who need it, that's good enough.

We may try proactive chat on one of our main pages next fall.

On Fri, Jun 3, 2016 at 10:58 AM, Julie A Arendt <jaar...@vcu.edu> wrote:

--

You received this message because you are subscribed to the Google Groups "libraryh3lp" group.
To unsubscribe from this group and stop receiving emails from it, send an email to libraryh3lp...@googlegroups.com.
To post to this group, send email to libra...@googlegroups.com.
Visit this group at https://groups.google.com/group/libraryh3lp.
For more options, visit https://groups.google.com/d/optout.

[Julie A Arendt]

Thanks to everyone who responded. 

We are planning to put it on just a handful of pages and have decided against putting it on our search results, for fear of bringing in too many questions. Given Michael's experience at Duke, maybe search results would bring in less than regular site pages.  It will be interesting to see how much the handful of pages we're trying brings in. 

Thanks again,

Julie

Julie Arendt

Science and Engineering Research Librarian

121J Cabell Library | VCU Libraries

901 Park Ave | PO Box 842033

Richmond, VA 23284-2033

(804) 828-1539 | jaar...@vcu.edu