INTRODUCTION This white paper explains how to calculate the expected return on
an investment made in site search. It shows how to build a robust
calculation from existing site performance metrics to help make
an informed data-driven decision about the value that search can
add to your online organisation.
The case for search is dicussed in greater detail in the White Paper
“The importance of search: how site search is crucial to the online
customer experience” available at www.colbenson.com
In that paper we discuss how your site search function is your
customer’s main dynamic interaction with your data and how this
impacts your bottom-line conversion rates.
Established industries have consistent and widely-understood
metrics which are clearly linked to industry key success factors.
New industries don’t.
In many cases, there is not even a settled view on the key success
factors. Building business cases based on analytics is tricky
enough, but even more so when there is no clearly understood
and accepted link between the analytics and the bottom line.
Search is a good example of this.
Outside of the sector, it is not obvious that search is a complex
area critical to online conversion. The customer experience is
often so smooth that the hard work behind the scenes is invisible.
Only when it fails to work do we notice it.
The rest of this paper will look at how we can measure search
success and how we can calculate the impact a search investment
will have on your organisation.
The importance of search
The problem
02
MANAGED SEARCH
In this paper we refer to
Managed Search. This is an
active approach to search
which involves the
organisation taking action
to improve search engine
performance based on
robust search analytics.

Not all businesses are alike, and the applicability of a managed
search function will vary by industry, sector and business. Some
sites are very search driven, others much less so. Some sites serve
information in return for advertising revenue, others want to sell
products or build customer engagement. The value of these
varies enormously.
This section looks at how important search is in terms of
delivering value to your organisation.
ONLINE SERVICE
There is also the role of
online service provision, such
as online banking. This is not
included in these calcula-
tions as it is not the role of
the search engine to provide
this service: it would proba-
bly work equally well
without site search. Therefo-
re, the business case for
search within service provi-
sion sites must still rest on
one or more of the three
search objectives: sales,
leads or information.
The objective is either:
1. A sale: the customer finds a product or service and clicks to
purchase it.
2. A lead: the customer finds a product or service and clicks to
move to the next stage, for example they request further
information, a demo or a quote.
3. Information: the customer finds the information they need.
Value of the search objectives
The second step is to understand the total value that the
searchable section of the site creates for your business.
Taking the above three objectives in turn:
For sales, the total value is simply the total value of all direct
online sales.
For leads, the total value is the total value of the leads.
The search objectives
The
Business
Case 1:
THE VALUE
OF SEARCH
TO YOUR
BUSINESS
03

For information websites it can be more complicated.
If the objective of the site is to provide information in order to
generate revenue through advertising sales and premium
subscriptions, then the total revenue can be used in the same way
that total sales works for a direct sales site. This number should
be an average for a selected period of time (usually a financial
year) because, unlike sales revenues, it is not directly linked to a
successful conversion. If you allow fluctuations of advertising
revenue to seep into your measure, then your site search
numbers will not accurately reflect the value of the search
engine.
A second option which works for sites which do not directly
generate revenue is to make the assumption that the money
invested in the site is equal to the value of the site. This means
that if your organisation spends a total of €10,000 per month on
site hosting, SEO, search, design, promotion, maintenance etc.
then you can assume that someone somewhere has already made
a robust business case that this investment is worthwhile and is
measuring its effectiveness (if this is not so, then this assumption
must be tentative and temporary!).
THE VALUE OF LEADS
This is usually a well-known
metric is the sales business,
each lead has a value assig-
ned to it depending on its
source.
It is calculated by adding up
all the sales made from
leads generated by the
website, less any other steps
in the conversion chain.
This latter part is important,
if your leads are difficult to
convert and this takes
further resources (usually
time and commission for
sales teams), then that lead
is less valuable and you
cannot assign all the value
created by the sale back to
the website. To do so is
misleading and could mean
we concentrate resources in
lead generation and not on
lead conversion which could
be where most value is
added.
04

To know if your search engine is delivering the results, you must
be sure that you can define what you mean by results.
This section looks at how you can define and measure
“conversion” for any website.
Usually we think of conversion as a successful sale. However,
search conversion is about making the correct connection
between the customer and the product (or information). The sale
– the things we’re measuring – is beyond the reach of the search
engine and is influenced by many other factors.
So we need to be able to isolate the search performance and
measure that the customer is being connected to the correct
product or information, but because we cannot know if a
customer is looking at the right thing, we must create conversion
proxies.
The
Business
Case 2:
HOW TO DEFINE
SUCCESSFUL
CONVERSIONS
What is search conversion?
The clearest way to know if the customer is looking at the product
they intended to find is to track what they do next. Do they
proceed to the relevant call to action (purchase, quote, demo
etc.)?
This is fraught with problems:
• Many site visitors use the site for research. They may find
exactly what they want (the site search did its job) but not buy
now because they are still researching.
• Research has shown that many visitors may take up to 28 days
before making their first purchase.
Creating conversion proxies
SEARCH CONVERSION
Search conversion can be
defined on either the user or
search level:
User: The percentage of
customers using search who
convert (buy a product),
where the customer was
connected to that product
by the search engine, over
the total number of
customers using search.
Search: The total number of
conversions (usually sales)
through search, over the
total number of searches.
05

• There are many other factors which influence a purchase
decision: price, customer reviews, delivery and returns policies,
payment methods, security concerns ... nothing to do with
search.
• Seasonal variation in demand. More sales will happen around
holidays and special gift days such as Mothers Day or World
Book Day.
• If the site does not include a call to action - such as
informational sites, how do we know that it was the right
information?
In order to apply the conversion proxy to the business case for
search, we must be able to isolate performance due to search
from other factors.
We do this by tracking the customer journey:
1. Number of site visitors
2. Number of searches
3. The CTR: are customers clicking through onto something
4. The conversion: are customers responding to the "call to
action"
The customer journey
At each stage, variations in the above metrics can be explained by
many variables, not just search engine performance.
Although the impact of search can be felt throughout the above
process, to simplify we can focus only on conversion (4).
If (1), (2) and (3) are within the bounds of what we would expect,
but (4) is off, then it suggests that either the customer is suddenly
going to the wrong product, or there is a new obstacle to
conversion, for example the price is no longer competitive.
To try to understand which it is, we need to look at what else
happens after (3) if not conversion.
The customer journey
AVERAGE SALE
In addition to conversion
rates, it is also useful to
track the value of the
average sale.
Sales driven by search tend
to be higher than those from
other methods. Good search
should not only drive higher
conversion rates, but also
lead to more valuable sales
as customers engage more
with the site, return more
often, and are attracted to
related products and other
search-driven features.
Customer
engages
(visits site
or mobile app)
Customer
searches
Customer finds
(CTR)
Customer buys
(Conversion)
1 2 3 4
06

There are two alternatives to success:
• Failure: the user abandons the site.
• Repeat: the user tries again, searches for something different
(an indication that they clicked the wrong thing).
If the user is put off by price, we'd expect to see failure rates
sneak up, if the user is not finding what they want, we'd expect to
see a rise in repeat rates.
This digging is worth doing even when the number seem to fit our
expectations, just to be sure that multiple variables are not
cancelling each other out.
For sites which sell information, either by advertising, premium
subscription or a non-revenue service (in which case it sells its
service to whoever provides the funding), a call to action is not
usually realistic.
We can try by including buttons which say "Did this answer your
question?" or by monitoring how many users share via social
media or bookmark the page, but most often this is just not
sufficient, covering only a small percentage of visits.
That doesn't mean it shouldn't be used and promoted, just that it
should be used alongside other metrics.
The relevant conversion proxy can be as simple as page hits, or
page hits minus bounces.
A more sophisticated approach, is to look at engagement: the
time spent on the page, and make the assumption that more time
equals more engagement.
Exactly what you define as engagement will vary. Is 30 seconds
enough? Is 5 hours really better than 5 minutes or does it mean
someone just left their browser open?
Other measures can be factored in, such as social media sharing
or “likes”. The important thing is to ensure the metric is linked to
the important user behaviour: the conversion. It cannot be linked
to advertising revenues, for example, or total site hits (this is not
conversion).
However you choose to construct your metric, you should get a
number which describes a successful conversion in a way you can
measure.
Conversion in non-sales sites
07

Once we have defined conversion and limited the scope of the
business case to the value generated by the search objectives, we
need to narrow the business case down to just those areas where
a new investment can make a difference.
The next step is to factor in how many of your successful
conversions come from search.
If you haven't got a measure for the percentage of conversions
that come from search, you can estimate that conversions via
search are usually twice those of non-search, and therefore
two-thirds of your "success" will be from search (although this will
vary by industry sector, numbers of products, ease to describe the
product in user-language, other ways to find products etc.).
For example, if a website which provides information has 90% of
its conversions from search, and we know that the organisation
invests €10,000 per month on the information section (and has
chosen this to represent the site's value), we can now say that
€9,000 is generated via search.
In an e-commerce example, let's assume 60% of conversions
come via search and total sales are €100,000 per month. We can
now say that search adds €60,000 per month.
This is an important distinction, investing in your site search can
only improve on sales that come from search!
The
Business
Case 3:
NARROWING
THE FOCUS
What is search conversion?
08

We now need to differentiate between searches that are already
working and those that are failing.
Three things can happen when a user searches:
1. Success: conversion!
2. Failure: the user gives up and abandons the search or, all too
often, the entire site.
3. Repeat: The search is repeated with different terms or the
results are refined.
For the purposes of this part of the analysis, we can ignore the
last of these because they will be counted when they eventually
stop repeating and either convert or abandon.
Usually a user who is forced to repeat their searches, time and
again, is more likely to give up, so the negative impact of this
repetition is captured in the higher failure rates.
Therefore, we can focus only on the ratio of conversions against
abandonments, or success against failure. It is not recommended
you do this for routine search performance metrics, only at this
business case stage to keep things simple.
Abandonments in non-managed search can be around 80%, or
closer to 30% in managed search. If you don't have an exact
figure, estimate 30% as a fair average.
Where is search failing
09

If we assume that the search investment will deliver a fairly
conservative improvement of 25%, we can arrive at a number
which represents the value you should expect to see.
• The €9,000 value from search came from only 70% of the
searches (the other 30% fail).
• The new search investment will reduce failures by 25%. 25% of
30% is 7.5%
• Therefore, we are pushing an extra 7.5% of searches through
the middle of the funnel along with the original 70%. Adding
these two together gives us 77.5%
• If 70% was worth €9,000 we can calculate how much 77.5% is
worth if we assume the same conversion rate: €9,965
• This means that this investment should give us €965 added
value per month
For the ecommerce example, we apply the same logic. €60,000
comes from 70% of the searches, so 77.5% would give us €66,429
- or €6,429 in monthly added value
The
Business
Case 4:
SO HOW MUCH
IS SITE SEARCH
WORTH?
10

In summary, putting the right numbers into the above logic leads
to a good calculation of what to expect from investing in a quality
site search solution in the short-term. This does not take into
account intangible benefits such as how much you can improve
engagement, increase frequency of visits and generally improve
margins based on increased customer satisfaction. All of these
things are likely, but can only be judged over the long-term.
A good search solution can also work as the foundation for mobile
shopping, apps, and social media search tools. In investing in a
good search platform now, you create a flexible foundation for
future development. This benefit is also not included because it
cannot be quantified in hard cash in the short-term.
SUMMARY
11

Business case for site search

More Related Content

What's hot (18)

Viewers also liked (9)

Similar to Business case for site search (20)

More from EmpathyBroker (13)

Recently uploaded (20)

Business case for site search