Selling via online chat

June 3rd, 2015 in Olark Development

If you didn’t already know, an arm of Coding Futures develops WordPress Plugins and sells them to end users and like a lot of similar companies, we have a live chat system (in our case, a system called Olark) on our website. Many customers seem to like it; it’s a great way of getting hold of either sales or technical staff without waiting for them to answer emails and it gives us yet another way to get across to our customers what personal service we offer.

We decided to combine Tim’s online sales optimisation and user tracking experience with my old-school sales attitude to turn it from a user-activated communications system, into an effective sales tool.

Olark already provided basic information about each person browsing the site but we soon took advantage of its extensible nature to include geo-location information and attempted to find the user in both our CRM system and our online support forum.

This meant that if the visitor was identified as an existing user (or even a lead we’d previously been in contact with) both sales guys and techies could enter a conversation armed with information of what products and licenses they have, what previous support tickets they’d raised, even which email promotions they’ve responded to. Whether you’re looking at customer service or sales conversions, the more information you have means the better job you do.

To set a username for the user, simply use:

api.chat.updateVisitorNickname({snippet: 'Customer Name'});

From a sales perspective, the real power of our extensions to the Olark systems comes from its tracking. It tracks every page a user visits that session (as it runs on every page) which allowed us to set up some very clever rules to identify certain types of “hot leads”.

For example if a user visits 4 pages including the pricing page they are flagged up, as are people who made it into the shopping cart but didn’t buy. Likewise, our support guys have people flagged up if they hit a certain numbers of support pages or start to raise a post on the forums but don’t hit send.

The code to flag this is relatively trivial:

api.chat.sendNotificationToOperator({body : "User has visted 4 pages including the pricing page"});

One of the nice new features of Olark is the ability to group operators; this means we can separate the sales and support teams. Each team has control over a given set of pages on the site and if a user begins a conversation then they are directed to that specific team. who can then look at the data we have for that customer and decide if it’s worth us initiating the conversation with the user rather than the other way around.

To assign the chat to a group, use:

olark.configure("system.group", "group_id");

Sure, a small percentage of users find this a little intrusive and don’t respond, but the vast majority are happy to engage with us as we are actually helping them find what they are looking for.

A good analogy is a shop assistant in a clothing store. Nobody likes to be followed around repeatedly being asked if they can help, but when you’re stood in the shop looking for help, you’re so grateful when the helpful assistant pops and not only wants to help, but also knows just the kind of thing you’re looking for.

As every company has different customer data systems and different types of users we can’t really offer what we’ve done as an off the shelf package, but if you’d like this kind of system implementing at your company, we’d love to hear from you.

Authors

Glenn Pegden
Glenn's background is a unique mix of sales, technology and customer service. In 2010 started acting as the sales force of NewMedias. When NewMedias were acquired by Coding Futures in 2011 he took up to role of Sales Director.

Todd Francis
Shortly after qualifying from Leeds Uni with a degree in New Media Todd became a professional coder. Soon after he caught the eyes of Tim and Kevin who were looking for a developer for their new company and he became Coding Future's first employee. Two days later he was in Rome talking to very large clients and things haven't slowed much since