The slow-rise phenomenon

18 Apr 2008 - 2:25pm
5 years ago
5 replies
537 reads
Robert Hoekman, Jr.
2005

A friend shared a quick fact with me today that reminded me of an
interesting phenomenon. Two quick stories: 1. My friend told me that he made
a few small changes to his site about a week ago, and today, he noticed his
page views have doubled. The first few days that the new design was live,
nothing changed. Now, he has twice the page views. 2. A few months ago, I
worked on the WordPress.com homepage, using some small tweaks to make the
Sign-up method more prominent. The first full day the change went live,
conversions went up 12%, but a week later the conversion rate had increased
by 25%. It's stayed that way ever since. I've noticed other instances like
this as well, and I'm very interested in discovering the why behind the
delay.
I think it has something to do with the trust that comes from repeat visits.
There's always been a large percentage of people who will not buy products
from an eCommerce site, for example, until they've visited the site several
times. I'm sure this mentality affects other types of sites as well. As in,
people are less prone to explore a site until they've been there at least a
couple of times, but once they have, if they like what they see, they'll
jump in. I'm guessing there are other factors as well, but I don't have any
evidence that explains causation.

So, I pose the question to the legion of expertise on this list. Have you
noticed a delay in results (positive or negative) due to design changes? If
so, how do you explain it?

-r-

Comments

18 Apr 2008 - 2:53pm
Victor Lombardi
2003

On Fri, Apr 18, 2008 at 3:25 PM, Robert Hoekman Jr <robert at rhjr.net> wrote:
> So, I pose the question to the legion of expertise on this list. Have you
> noticed a delay in results (positive or negative) due to design changes? If
> so, how do you explain it?

This could be a corollary to the "cognitive lock-in" phenomenon that
Gerald Lohse described. Lohse found that familiarity with a particular
website makes visitors less likely to switch to a competitive site
because of the effort and time needed to become familiar with another
site. But if the change isn't dramatic, your visitors may adjust
instead of abandoning and subsequently become accustomed to the
change, and eventually buy something.

I put some of Lohse's findings in context in this article on rate of
change in homepage design:
http://boxesandarrows.com/view/the_evolving_homepage_the_growth_of_three_booksellers

Another explanation might be that by making a small change you drew
attention to something valuable that visitors wouldn't have seen had
you not made a change. David Danielson's work on Transitional
Volatility explains how subtle changes can draw the reader's
attention...
http://noisebetweenstations.com/personal/weblogs/?p=1010

Victor

18 Apr 2008 - 3:04pm
Robert Hoekman, Jr.
2005

>
> This could be a corollary to the "cognitive lock-in" phenomenon that
> Gerald Lohse described.

I'm familiar with this idea, but in this case, we're talking mostly about
new, first-time site visitors. On WordPress.com, it's people who have yet to
sign up, and on my friend's site, it's people who have just heard about him
for the first time (he's a new author).

David Danielson's work on Transitional
> Volatility explains how subtle changes can draw the reader's
> attention...
> http://noisebetweenstations.com/personal/weblogs/?p=1010
>

Oh, I'm all about small changes. On the WordPress site, for example, we
stripped out two relatively unnoticeable sign-up links and made one big one,
and the conversion rate skyrocketed. That's all we did. I know the impact of
small changes very, very well, and I constantly evangelize this type of
thinking.

What I'm interested in is why the rise in effectiveness takes so long to
kick in? I mean, logically, you should be able to compare one day's metrics
with another and see the effect of a design change, but this is not the
case. There's a delay, and this delay is absolutely fascinating to me.

-r-

19 Apr 2008 - 9:28am
Bruno Figueiredo
2007

I guess that you could track the slow-rise phenomenon to the speed of
word of mouth. When small tweaks happen, they're usually not
advertised. So what's happening, I suppose, is that people see
something as easier and they spread the positive change amongst their
friends. While people who bumped on the site, if they can easily see
the path to completion with the change in place, then it leads to
conversion.

I've always been a sucker for this kind of studies focusing on small
changes and the impacts they made. I remember that a while ago eBay
rolled out their new layout by placing minor changes every couple of
days until they reached the full design overhaul. Don't know how it
worked out in the end, but I wouldn't mind reading about it.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Posted from the new ixda.org
http://www.ixda.org/discuss?post=28215

20 Apr 2008 - 5:32pm
Victor Lombardi
2003

On Fri, Apr 18, 2008 at 4:04 PM, Robert Hoekman Jr <robert at rhjr.net> wrote:
> What I'm interested in is why the rise in effectiveness takes so long to
> kick in? I mean, logically, you should be able to compare one day's metrics
> with another and see the effect of a design change, but this is not the
> case. There's a delay, and this delay is absolutely fascinating to me.

Can you share the metrics you're seeing?

21 Apr 2008 - 3:27pm
Robert Hoekman, Jr.
2005

>
> Can you share the metrics you're seeing?
>

Not specifically, no, but I can tell you that I'm comparing page views,
site-wide and then by segment. There are multiple types of site users
(visitors, visitors who search, members, content creators, etc).

Most of the traffic on this site enters from a search engine, and most of
that traffic bounces immediately. It's this group in particular the design
changes were geared towards—things to help them understand the purpose of
the site within a few seconds, improve sideways discovery, etc.

For it to be a cache issue, most of the current users would have to
have already
visited the pages they're viewing now (thereby caching them), and that's not
the case. This was also not the case for WordPress.com.

So, the delay seems to occur even when the largest audience segment is
first-time site visitors. When a new design is successful, it seems, the
effect is not immediate. It still takes a week or so for the metrics to
start reflecting the success of the new design.

-r-

Syndicate content Get the feed