The LinkedIn Algorithm: How it Works
By Joshua BoydDec 13th
Published November 6th 2017
There are various definitions of brand health. Each includes a collection of metrics and approaches, but they all lead to the same thing. Brand health is essentially measured by how effective your brand is in helping you to achieve your aims. That effectiveness can be measured in a number of ways, such as:
These are just a few examples, but they arguably all fall under brand health. Each metrics plays its own role, but they fit together to make your brand a powerful asset. Think about what metrics are most relevant to you and build a list that makes sense.
Each aspect of brand health is important in its own way. Looking at the topic as a whole gives you the bigger picture to make it clear where your strengths and weaknesses are. For example, your brand awareness might be great, but if you don’t know whether it’s positive or negative awareness, you’re missing half the picture.
If you consider all the factors together you can work out if your brand is unhealthy, fine, or thriving (or whatever scale you want to use). From there you can take specific actions.
If it’s unhealthy, your brand is actively damaging your organisation. Using the metrics you’re tracking you’ll be able to see why this happening and how to stop it. When it’s thriving, you can find out how to keep this going. And when it’s fine you can look for improvements.
Essentially it’s about getting a full understanding of your brand and how you can use it to your benefit.
First we can look at a few methods to gather data on your brand. With different factors involved in brand health, there’s plenty of different ways to do this. After that, we can look at how to take lots of metrics and use them to indicate brand health clearly.
Social listening tools, like the Brandwatch Analytics product we offer or the many others on the market, give an incredible amount of insight into your brand on a number of levels.
You can measure brand awareness by looking at volumes of mentions around your organisation and products. You can them compare that to your competitors to see how you how you’re doing comparatively.
From there you can also measure sentiment around those mentions. In other words, whether they’re positive or negative. This can give you an idea into your brand reputation.
Then you could look at you brand positioning by looking at how people talk about you. Do they associate you with the image you’re trying to project or are you viewed entirely differently?
You could also look at employee engagement and track how successful your employee are at spreading the word on places like Twitter or Reddit.
Another common way to do brand research. You can get a very good understanding of your brand awareness, particularly among specific groups (depending on how you conduct the survey).
You again get sentiment from these, but you can also get some very in-depth qualitative data too. Thinking around brand positioning and reputation you can choose exact questions and find out more details on how people see your brand.
Similar to focus groups and surveys, conducting regular interviews with your customers can give great insight into brand health.
For example, you can see whether their view of your brand has changed over time, why they chose you over your competitors, and what they like and dislike about you.
You can get this data with basic feedback forms, talking to your customer-facing employees, and conducting more in-depth interviews with customers. All of these will be integral to measuring brand health.
So when you’ve chosen your metrics and collated the data for each, what’s next? It’ll mostly look like a range of unrelated numbers with widely different contexts. The best bet is to visualize this information that quickly shows when things are changing.
In this case, a good old-fashioned spreadsheet is your friend. You could do this in Excel, Sheets, or any other program. We won’t go into the specifics of formulas, but none of this will be too complicated.
So, we want to be able to get an immediate idea of brand health, but also easily compare current figures to previous ones for context. To visualise this, we’ll need just a single table.
1. Organize your data
With so many sources you’ll probably have a load of spreadsheet files with lots of different figures in them. The first thing you need to do is pull out the exact figures you want and bring them into one file. Then set up a table including each one with a time series of your choosing.
Although it’s a bit more complicated, you could bring all of your spreadsheets into putting each on its own sheet. Then you could have your main table pull in data from each one automatically. This saves the effort of copying and pasting all the time. It’s worth doing, but always double-check everything is correctly set up.
Here’s an example:
Now you have all your data ready, but it’s not instantly clear how everything compares to each other. We need to add some extra visual info so we can draw conclusions easily. The easiest way to do this with a table is to use conditional formatting. This is a set option in Excel and Sheets, so it’s pretty easy to do.
You select one column’s figures at a time and apply formatting to each one individually. The best option is a colour ramp. You can choose whatever colours you want, but in this case we’re going to pick red to show bad results, and green to show good. This formatting uses the data in the table to choose the right colours. These will change as more data is added in over time.
Once you’re done, you’ll get something like this:
And, well, that’s it. This is obviously very top level, but you can instantly see how things are. In the above example, December is looking pretty good with all greens, while January was looking pretty bad. You can also see when some metrics are looking unhealthy even if others are looking good. This can be really handy for evaluating past performances.
There is a big caveat to this though. You need to be careful with some metrics for the time periods you’re covering. Sentiment and share of voice are usually fine, but others not. Essentially, you should be expecting your brand awareness to grow over time, so comparing yourself to figures from two years ago won’t make sense. It’ll also mask bad results as the average is being skewed.
In this case there’s two solutions. One, only include a certain time period, like the past year. Or two, just look at percentage change. This means you can see any recent changes quickly. Unfortunately percent change means it’s hard to compare to historical data and removes context. It’s often worth including this along with the raw numbers, plus only considering a certain time frame.
Of course you could set up the table differently if you like. Simply tailor it to your specific needs and metrics. From there you can get an overall view of your brand health with a simple table and take action accordingly.