LinkedIn's Kendra Speed on measuring brand density

Kendra Speed, Director of Consumer Market Research at LinkedIn, joined us on our podcast recently to chat about brand density, why it matters and how it can be measured. Kendra works closely with product marketing, brand and product leaders to develop and deliver consumer and market insight that helps guide strategic decision-making. She's dedicated to understanding consumer and premium business lines at LinkedIn. The most significant thing she shared with us during this episode is how they’ve started to look at brand density and leverage it at LinkedIn. She says, “As we think about our traditional ways of tracking the strength of our brands, I want to talk a little bit about some of the innovations we put into place recently.”

She shares more about her history tracking brands, dating back 20 years ago at Procter & Gamble, specifically working to understand the company’s hair care brands and their performance within the broader hair care category. During this time she asked questions like, 

“How is each brand doing surrounding what they said they want to be doing? Where are the whitespace opportunities? How do they differentiate themselves from each other? And then what are some of the aspirations that the brand should have?” Over the next few years, she’s found that the world has become even more challenging when it comes to tracking brands. 

One way she’s found to get more insight is by measuring and tracking brand density. She says, “When you think about brand density, it really speaks to the strength of your brand. Think about what are the associations that your brand has and then go a step further to what is the interrelationship between all of those associations? Ideally, we want our brands to have lots of great associations with things that matter in the category.” She talks about the history of measuring brand density, as it is not a new concept. 

“It really relies on taking things that we have traditionally done, understanding associations, and then taking a step further and thinking about what those relationships are. If you think about some of the research work that's been done, it really comes down to showing brand density when you're deriving it from actual open-ended questions, it correlates very highly with market share. And then also showing that these open-end associations, direct from consumers, carry so much more weight than what we see from a more static battery of attributes, just asking people to go through and rate them. At some point in time, we're probably running into a degree of fatigue.”

Kendra said she’s chosen brand density as a key metric because it gives a more holistic picture of performance - showing how insights intersect. LinkedIn has experimented with different approaches to measuring brand density. For example, her team might use a study to help get all of the different category associations and understand the interrelationships. Then they cast a broader net by leveraging technology or a tool,  such as chat GPT, using a multistep process to help iterate on the correct AI prompt. Then they take all the data together to do an assessment of, for example, how often specific things are mentioned together. Finally they look back at their traditional brand trackers and ask “is this making sense given what we already know about the category?” 

They are constantly refining different approaches and the process, making it faster and easier to get insights. Kendra admits that uncovering brand density metrics is a complex process. As a word of advice, she says, “Iterate, iterate, iterate. This is not one of those things that you can just go and say, hey, I'm going to hop on chat GPT and get back one listing and we're good to go, we're gonna run with that. There's really art and science to it, as well as a degree of experimentation. Go in with the mindset that there will be some trial and error.” She also says it is important to find a good partner who specializes in uncovering these kinds of insights. 

Throughout the process, Kendra talked about the high level of human expertise involved. She covers the importance of overseeing the process, looking at “key points to make sure that what we're getting out is actually making sense.” Also, it’s not time to throw the baby out with the bathwater. She says “We're not yet at a place of departure, where we're shelving all tracking studies and we're going to go all out using chatGPT to get a sense of our brand strength. It really is the marriage or the merger of the two.”

She gives many more examples of how measuring brand density can be a powerful tool, and we can adapt the way we approach business questions - using what we already have in the toolkit and new technology to give us more complete data. Kendra shares some specifics on how they’ve tackled the complexities of this approach at LinkedIn, giving some tips for others looking to do the same - including details on training AI and developing prompts that provide the best feedback. “For me, it's a really exciting time for market research. I think whenever we have things that come in the marketplace that feel disruptive, it can always be an opportunity to totally shut down or it can be this prompt to say, what is the opportunity here? How do we adapt, how do we recreate the value that we bring in the organizations that we are in?” 

During our conversation, the topic of shrinking budgets on the client side came up, and how technology can help them become more efficient with the dollars that they do have available. Technology can take on rudimentary tasks, evolve the way we work, so we are spending time on those highest impact activities and those places where we add the most value. Kendra shares some examples of how she’s integrated new technology into workflows over the years. 

She says there’s no shortage of data at LinkedIn, but she’s always asking “how might you look at this data differently that drives greater insight, illumination for internal stakeholders to help them in their business decision making? So for me, that's really the most exciting part.” 

Listen to our entire conversation here:

No Comments Yet

Let us know what you think

Subscribe by email