Part 1: Finding Social Profiles
Social CRM is all the rage these days. Every
month dozens of articles try to convince us that connecting our CRM
systems to social networks will give us great opportunities for
sales, marketing and support. But one
important message seems to be overlooked among all these great
stories: How do I connect my CRM to
LinkedIn, Facebook and Twitter?
The current generation of Social CRM implementations is facing a
few problems when it comes to connecting CRM systems to social
networks. First of all, the most innovative solutions on
the market are all stand-alone products. But many
organizations are not willing to install yet another tool as a
separate island, and filling it with customer data. What they need
is a social dimension in the tools that they already use,
and a master data management (MDM) solution across multiple tools,
so that the social data can be shared. But in order to connect
social network profiles to contact data in our regular CRM systems
we have to overcome a couple of challenges. This
is the first of three articles describing these challenges.
The first challenge is that our systems are already full of
customer data. And to connect that data to social network profiles
we have to manually find customers on-line and
manually enter their user names and profiles into our
systems. Why? For thousands of customers that is a lot of
data entry! There should be a way to automate all that, don't you
think? On the other hand, no automated system will ever be able to
match the Steven Smith in my database with the 27,000
Steven Smiths on Facebook. But I know what our Steven
Smith looks like, because he was at our service desk yesterday. So
if I could just see some pictures…
Application programming interfaces (APIs) enable traditional CRM
systems to find people on social networks. But different APIs use
different methods, and each method has its advantages and
disadvantages.
Find People by Email Address
A search on a person's email address results in high quality
hits, because most email addresses uniquely identify individual
people. However, in many databases personal email addresses are not
(or sparsely) available. Furthermore, some people use different
addresses on different social networks, and most social networks do
not permit searches using email addresses. Therefore, this search
method has only few hits. But when there's a hit it is
probably a good one.
Find People by User Name
A search on a person's user name may result in more hits because
user names are public where email addresses are not. And the
quality of these hits is just as high. Some businesses have
customer-facing services with user names available, and we can
assume that many consumers prefer to use the same user name on
different systems. Unfortunately, in many cases the preferred user
name of a person will be unknown. This renders this method even
less useful, unless it is used as an additional step after
finding a user name with one of the other two methods.
Find People by Full Name
A search on a person's full name results in many more hits than
either of the previous two methods. However, you not only get the
person you're looking for, but also everyone else with a similar
name. Depending on the uniqueness of the name this might or might
not be a problem. By checking or filtering on additional data, like
country, city, and bio, one is often able to find the right person
in a relatively short time. A user's on-line profile pictures are
particularly useful with this method. No system in the world can
rival with our own experience at facial recognition.
The conclusion, with the current state of technologies, is that
providers of CRM systems and other contact management applications
might need a combination of all three methods to match
social network profiles with existing customer data. A
simple automated sweep of email addresses can work well to find a
portion of your contacts on social networks. But it may leave the
majority of your contacts in the dark. The second approach is then
to find the remaining people by name, location, and picture. As
this is a slower process it should perhaps only be performed for 10
or 20 percent of your most valuable customers. But, given the 80-20
rule, these contacts are probably responsible for 80 percent of
your business.
Plenty of APIs, and various aggregated search engines, are
already available. The only thing left for you to do is to convince
someone that your CRM needs an upgrade…
(go to
part 2: Hiding Network Issues,
part 3: Forming Unique Identities)