Here at Adexchange we love data. As contact centre managers, data gives us x-ray vision across our operation by allowing us to:
- Have a panoramic view of the touch-point efficacy in our omnichannel hubs.
- Identify any niggles in our lines of communication with our customers.
- Gain an intimate understanding of where our agents’ time is being swallowed up, and how we can give it back to them.
Already, that is a LOT to love about data. But we also understand there’s such a thing as too much data (don’t tell Beryl in analytics). In an increasingly digital world, state-of-the-art kit can spout reams of random numbers that result in:
- Spending hours poring over spreadsheets without really knowing why
- Paying someone to spend hours poring over spreadsheets without really knowing why
- Staring at graphs which translate to our job in zero ways
- Writing up reports full of numbers which we hope no one asks about, because yes Len, you’re right, no they aren’t relevant. Even numbers are better in quality than quantity. How do we, as contact centre managers, cut the noise and get to the good stuff?
IGNORE numbers until we work out what we need to know
Our objective has to come before the finer fodder of data. What do we want to achieve with these numbers? Take a clothing company that wants to revamp the FAQ section of its website. Analysing the page data of their bestselling skirt might be fun (we said might), but it’s not going to give the team an idea of the recurring questions our customers ask. Pulling data on the reasons for customer contact with agents would be much more helpful. Once it’s clear that 78 people ask about the returns policy per afternoon, that bad boy can get an FAQ spot right away. Equally, a gym wondering why their members aren’t signing-in on the fancy new app (and then complaining about getting charged as no-shows), wouldn’t analyse data around new members who haven’t even got the app yet. Instead, they’d home in on the customer journeys of seasoned members with online accounts. Data around the clicks would promptly reveal if there are design flaws in finding the app’s ‘sign-in’. Equally, looking at the open rates of emails telling people about the sign-in policy would let the team know very quickly whether it’s something that members are even aware they need to do.
Identify which numbers we can INFLUENCE
Let’s return to the bestselling skirt in our favourite fictional clothing company. As much as it would be great to know why customers love this skirt so much, the chances are that it’s not down to our emails or live chat, but to a great fashion designer. So, unless we were thinking of entirely redesigning this skirt’s page and placement on the website, there’s no point paining over its data, when it’s influenced by something outside of our operation. What the contact centre manager could influence though, and should analyse, is the average wait time of someone calling up to ask about when this magical skirt will be back in stock. Considering it’s a likely purchase if the call goes well, we’d want to avoid long wait times that would cause a customer to give up on it for another, less good, competitor’s skirt.
Deal with FEWER numbers
On this note, average wait times, handling times and reasons for calling are often valuable nuggets of data to consider. With the right objective, these three data values can tell a more comprehensive story than hundreds of others combined. They are measurable and within our realms of influence as contact centre managers. But whatever our chosen values, it’s not about looking at ALL the data, just all the data relevant to our objective.
Ensure our numbers are ACCURATE
Sadly, even if a number is relevant to our objectives, it might not be accurate. Net Promoter Scores (NPS) are a good example of this. Customers will often down-mark an agent in these surveys because the agent has told them something they don’t want to hear, in a classic case of shooting the messenger. It doesn’t mean that the agent hasn’t offered effective communication. If our objective is to optimise the efficiency of our agents, leaning too heavily on customer review scores might lead us down a rabbit hole UNLESS we’re aware of their larger context. Aka – did Habiba’s scores get slammed the same day that she had to tell 800 customers that their washing machines are being recalled and they’ll have to use Shelley’s extortionate laundrette for a week?
Make our data targets ACHIEVABLE
As well as making sure that the data we’re analysing is accurate and relevant, we want to make sure that the numbers we’re aiming for are viable. Our data aspirations should push our contact centres’ potential, but they shouldn’t overshoot it like a blindly proud mum on sports day. Chanting “let’s cut abandonment figures by 100%!” isn’t going to make it happen, any more than a hand-painted banner is going to get Eddie over those hurdles.
**
At Adexchange, we recognise that dealing with data overwhelm is a time-consuming bag. If you’d like us to beat the clock and mine the good stuff for you, get in touch– we promise not to spout too many numbers at you.
