The success of any business is wrapped up in the team’s ability to answer the big questions. Stuff like…
How much of the market have we left untapped?
What does our audience actually care about?
Which products or new attributes should we consider for development?
Traditionally, these types of questions require qualitative data.
Collecting that data is slow, the sample size is often small, and the resource investment is big. Meanwhile, the market is evolving so quickly, results might not be relevant in a matter of months.
Businesses need a better way of answering critical questions, fast.
In this article, we’ll introduce a new way to produce market insights in a fraction of the time - without surveying - using digital data most sites already have.
We’ll cover:
Digital Market Intelligence is an accelerated method for gathering consumer and market intelligence at scale, more quickly - and often more accurately - than traditional market research.
It collects, aggregates, and analyzes vast data sets from publicly available online sources, including search queries, comments, blogs, social media, news, forums, and reviews. Then it uses that data to answer complex business questions without cumbersome surveying or consumer insights initiatives.
Note: With all of this being said, DMI is most effective in well-established verticals or well-known topic areas. Extremely niche verticals don’t always have enough data to support a DMI approach.
In most cases, we look at data in silos. Keyword data is useful for SEO and content creation. Social media engagement is useful for social content and strategy… so on.
Unlocking the power of DMI starts with a shift in perspective. If our goal is to understand the market, any user behavior data is connected because it's part of that market.
We need to take a step back and look at different data types holistically. In the DMI model, we break different types of data into three distinct categories, based on the primary ways users show interest online.
When we search for something online, we’re actively looking for something we want — whether it’s a product, a specific brand, information on a topic, etc. It’s active demand.
When SEOs use keyword data to understand how many people search for different terms, they’re really measuring demand. When we use the same data at scale, we can quantify demand across entire markets, verticals, and more.
The content we see on social media - especially in feeds (content that’s literally “fed” to us) - is served for scrolling. Users are browsing passively. So when they like, share, or otherwise interact with a piece of content, it’s a signal that the content cut through the noise and caught their attention.
At scale across platforms, social media data is a living archive of engagement. It tells us what inspires different audiences and grows affinity for brands.
It’s also especially useful for assessing trends, since topics take off on platforms like TikTok before they ever make their way to search engines like Google. (AKA trends generally appear in engagement data before they appear in demand data!)
Forums, comments, and reviews are a massive pool of user feedback, filled with uncensored opinions. We can tap into that qualitative data to get a clear understanding of what different audiences care about, how they react, and their pain points and motivations.
Just as importantly, it can help brands understand the information that’s most helpful to users, so they can make their content more useful and relevant.
The crux of using everyday data for DMI is the ability to analyze it at scale. Our method pulls publicly available data from sources in each category using a short list of methods.
Many platforms provide their own reporting sites that aggregate and report data, like Google Search Console or Pinterest Trends.
Pros: Easy to search, filter, and view data.
Cons: Data is often limited in some capacity.
In a lot of cases, platforms provide APIs that allow analysts to pull large amounts of relevant data.
Pros: APIs export data in bulk very quickly.
Cons: The data takes more cleanup, categorization, and analysis before it’s telling or useful.
Tech like Apify enables the collection of public-facing web content for analysis.
Pros: Sources data that is otherwise unavailable at scale through APIs or other tools.
Cons: Crawl tools cost money, and the data is usually more helpful for content analysis versus sourcing behavior metrics.
Third-party tools like Semrush aggregate large amounts of behavior data, while also providing analysis, categorization, and calculation of key metrics.
Pros: Data is aggregated at scale and reported in a clean interface with targeted and often customizable analysis.
Cons: You have to pay to play.
DMI supplements our knowledge where first-party data falls short. At the same time, first-party CRM and analytics data can help connect the dots from market data to business outcomes.
If your data isn’t self-hosted and lives on an external infrastructure like GA4 or Salesforce, it’s technically second-party data. But you can still use it in the same way.
Pros: Detailed information about how users behave on your site, including conversion metrics.
Cons: Limited to behavior on your own domain.
Third-party organizations aggregate large data sets in the form of reports, some of which are free. (It’s especially common in industries like finance, real estate, and healthcare.)
Pros: Many reports contain survey data already, eliminating the need to conduct qualitative research of your own.
Cons: The price range can vary drastically, and the data you get is the data you get.
We pull data where interest and behavior happen, exactly as they happen, using the methods we just talked about. The sources vary from one data category to the next, based on the purpose of each platform.
The metrics most marketing teams already have in their measurement plan are the inputs that allow us to calculate insights using the DMI model.
Honestly, it’s not a monumental shift in perspective.
The key is thinking about what each metric tells us at the highest level. When we back out of the channel silos, dots start to connect between data points and the types of questions they can answer — often by identifying patterns across data at scale.
Quantifies how often people are actively seeking products, information, or brands at a given time.
It can answer: How much demand is there for [product/vertical/market]?
Illustrate changes in demand for a topic by looking at shifts in keyword volume over a window of time.
It can answer: How does seasonality affect demand for this product?
Identifies where users are in their customer journey plus common language and behavior at different funnel stages.
It can answer: How many people are actively researching products and services?
Compares brand demand apples-to-apples, while also defining how much demand each captures within the landscape.
It can answer: How much demand is there for us versus our competitors?
Provide valuable demographic data based on interests and motivators.
It can answer: What’s the true value of our brand community in terms of reach?
Illustrate how well brands are growing a loyal, organic following.
It can answer: What’s the potential reach of our own community, or working with different creators based on the sizes of their communities?
Quantifies how much content is created around a topic or trend over time.
It can answer: Which trends are going viral? And which are worth jumping on?
Show how well content engages users and generates interest or inspiration.
It can answer: Which topics or tactics are engaging people the most?
Represent real problems that real people are trying to solve while telling us how prevalent these issues are.
It can answer: What are the biggest pain points for our audience?
Show which types of information (and which authors!) answer those questions best.
It can answer: Is there a gap between our content and the answers people find to be helpful?
Provide real, uncensored consumer sentiment about products, trends, and topics.
It can answer: Which attributes or use cases do most people cite who have negative sentiments about our brand?
Tune into the language audiences use to solve problems and express opinions.
It can answer: Which topics prompt the strongest reactions from people?
The beauty of DMI is its flexibility. There are so many different questions we can answer and scales we can work at!
The insights it yields are useful across different channels, levels of the team, and even the business at large.
We’re just introducing DMI, so let’s start with some approachable - but still powerful - use cases for each of the three categories of data. In future content, we’ll introduce more complex use cases and how we use machine learning models to get there.
Oh, and these all have to do with food because we’re hungry. Hope you’ve got snacks! 🍿
One of the simplest and fastest ways to find competitive insights between two or more brands is creating a domain scorecard.
It’s a simple table that includes the following information:
You can add engagement data points such as followers, branded hashtag usage, etc., but the basic version above can still tell us a lot.
To illustrate, let’s quickly compare two brands and break down the insights. Since Tory’s been eyeing ceramic pans, we’ll use two of the more popular brands in this vertical: Caraway and Our Place.
We’re pulling metrics from Semrush, but Ahrefs, Moz, and other tools can help you get to the same data points (with variance expected from platform to platform).
You’ll notice this is similar to the work a lot of SEOs already do.
It’s a matter of framing. By breaking down the search silo, we can connect the data to insights outside of SEO, like the following:
1. Caraway has likely developed more products and content overall, which is helping them win more active demand from brand-agnostic searches.
A supplementary content gap analysis could help yield topics or products that Our Place may not be capitalizing on right now. Or for Caraway, it could yield areas of differentiation to lean into.
2. Both brands drive a considerable amount of their organic traffic from branded terms, which indicates users are finding them via word-of-mouth, high return rate, or marketing on other channels.
Further investigation into engagement data could help each brand identify what’s working for the other in terms of offsite marketing strategies.
3. Our Place isn’t able to turn brand demand into site visits as effectively as Caraway (some users are likely going to Our Place products on other sites).
Our Place could combine this with first-party sales data to better understand how driving more of this traffic to their own site could impact profit margins.
While these insights aren’t going to help Tory choose which pans she wants, they’re certainly useful from a business standpoint for these brands or potential competitors. The great part is, putting the table together and drawing insights took about half an hour.
As we write this, the “Triple Dipper” from Chili’s is having a moment. (Yes, you read that right.)
Let’s say one of us works at a major food blog, and a big part of our content program is capitalizing on food trends. Should we create our own elevated variation and share the results/recipe online? We need to understand whether people still care or if this trend is on its way down.
Since TikTok is a major source of new trends, let’s see what TikTok Creative Center has to say.
Here, we can filter down to the top 100 newly popular hashtags for the food and beverage industry on TikTok. Looking at the last seven days is a great way to find the newest trends.
We found #TripleDipper down at #98, which isn’t too surprising considering it’s up against heavy seasonal hitters like #meatball.
In the analytics for this specific hashtag, we get lots of data we can use to discern whether to “skip” or “dip” on this trend.
First off, we can gauge the virality by calculating the percentage of all-time posts created in the last seven days for this hashtag.
Doing the math ([687 posts in the past 7 days / 4k posts all time] x 100), it’s ~17%. That’s a rather high number compared to something like #meatball (<2%). The trend has virality.
But is it still trending up? The Interest Over Time chart can answer that, but it’s hard to tell by looking at such a small window. Time to zoom out a little bit and look at the trend line over the past 120 days.
Uh oh, looks like #TripleDipper is starting to slip.
Furthermore, our target audience is millennials. So before creating any content, we need to make sure it’s going to reach the right people.
Turns out by looking at the demographics, it might not be the right topic. Only 19% of the interested audience is older than 25. Pretty low compared to a popular hashtag like #SoupRecipes, where 54% of the interested audience is 25 or older.
Overall, it looks like it’s probably not worth the resources to lean into this trend.
The internet is flooded with content. Most of it isn’t actually helpful at all, because for years, many sites have focused on keyword-optimized content, rather than user-focused content.
On search engines like Google, that’s led to a big gap between the content that’s recommended in search results and the content that users actually find helpful. So people started abandoning Google for sites like TikTok and Reddit. Lo and behold, results from Reddit are dominating SERPs.
Google understands that UGC sources like Reddit are where differentiated, helpful content comes to the surface. (Of course, there’s also humor that doesn’t always translate, bad players, spam, etc. But that’s a story for a different day!)
As a brand, tapping into that pool of knowledge can tell you what people actually want to know and which answers serve them best. Both are huge in the context of creating effective, relevant marketing for your target audience.
Sure, we can go to Reddit, read through top posts in relevant subreddits, and see what’s generating responses. But if we want to look at broad swaths of content across many subreddits, that could take weeks or even months!
There are different ways to find insights from Reddit at scale. For now, we’ll share an example using a tool called GummySearch.
Start by creating an audience made up of the members of relevant subreddits. Going back to our food blog, we quickly created an audience that spans 57.1 million members. Not exactly a small sample size.
Then we look at the Reddit activity for this audience to find patterns at scale. “Advice requests” is particularly useful for getting a sense of what people want to know.
Clicking in to browse posts gives the most popular posts based on upvotes or comments, which is an idea goldmine in itself.
Hello quick video for Instagram Reels or TikTok…
This would be an awesome campaign to generate community engagement and sharing on social.
Round these up to create a blog post people actually care about, with tips they actually want to use.
This should come in handy for Thanksgiving content too, so let’s bookmark it.
We can also look for patterns that pop up across a large number of posts. For example, we found these patterns by analyzing a sample of 500 posts using AI features. (It’s possible to do the same using your own machine learning models - which we do - but that’s a whole article in itself!)
Not sure which subreddits your target audience interacts with most? Another tool, SparkToro, can give you a short list to plug into your audience.
Enter a keyword or website that you know your target audience uses. SparkToro will spit back an audience profile, including demographics, social accounts popular with them, subreddits they engage with, and more.
For our food blog, we could enter our own site. Or, we could check out the audience for a site like Bon Appétit.
Think through the biggest questions you have in your role, across your team, or across the business at large. Chances are, there’s an answer in the data you already have.
As you take a step back and look at the numbers more holistically through the DMI model, the key is thinking creatively to make the connections. It all starts with a simple shift in perspective from marketing data to Digital Market Intelligence.
We’re here to help illuminate the insights.
Home Search DataDigital Market IntelligenceOpinionated SEO Opinions Social Impact Careers Products Testimonials FeaturedIndustriesResourcesContact
