Facebook Topic Data has been a hot button issue in the social listening space of late. Yet many brand don’t fully understand what it is or why they should consider asking for it with their listening platform. If that wasn’t bad enough, I spoke to a representative of a known social listening platform last week who said about Facebook Topic Data, “We don’t offer it because we looked into it and just didn’t see any use for it.”
Yes, I had to pick my jaw up off the floor. But in full context, I understood what he meant. (More in a moment.)
The point is there is not a lot of clarity in the marketplace on what Facebook Topic Data is and why your brand should be looking for it. So here’s my take:
Facebook Topic Data Does Not Include Post-Level Data
Because Facebook chooses to protect the individual user’s posts from being used for data mining or targeting, post-level data is not available through Facebook Topic Data (FTD). Since the social listening industry is built upon the ability to search, surface and analyze individual posts, this runs counter to the way we’ve always seen social listening data presented. While FTD does surface the total number of posts, likes/shares/comments per post and the like, you will not see a verbatim feed of all the posts included in the topic.
Because the social listening representative I spoke to was looking at data with those blinders on, he did not see the bigger picture or value of Facebook Topic Data. It doesn’t speak well of his company that they do not, but in that context, I understand the opinion.
Facebook Topic Data Is Broad-Level Insight Data
Instead of granular analysis of who said what, FTD provides a high-level analysis of the topic in question. No, it doesn’t go very deep on analysis, but does provide enough information for a research-minded analyst to assert critical insights about the Facebook audience for your brand.
What FTD does surface includes:
- Total volume of posts and social actions (likes/comments/shares) on posts in a given topic
- A timeline of post volume and social actions
- Demographic makeup of action-takers including gender and age
- A broad sentiment analysis of the topic
- What languages the topic is discussed in
- Popular related and sub-topics within the main topic
- Popular links shared and referenced within the conversation set
- Popular domains shared and referenced within the conversation set
- Popular hashtags used within the conversation set
- Volume and types of media used in posts about the topic (images, videos, posts, links and shares)
- Author type broken down by individual user vs. brand page
Facebook Topic Data Is More Important Conversation Data
The final entry in the list above is the first insight that offers a critical piece of understanding for brands. On one topic I analyzed recently (craft beer), the number of posts from brand pages that surfaced in the conversation was a little over 20,000. The number of posts from individual Facebook users? Over 233,000. What that means is the actual volume of conversation we’ve been missing on Facebook could be over 10 times what we’ve been able to see to date with social listening tools!
Before FTD was available, the only conversations social listening tools could surface were public-facing ones on brand pages and public groups it could index and scrape itself. Anyone who has done a source analysis of social listening topics in any platform can tell you that the vast majority of the conversation comes from Twitter with blogs, news sites, forums, review sites and social networks all coming in a far distant second through whatever.
But if 90% or more of the conversation on Facebook isn’t included, what we’ve been looking at and making decisions from in the past is missing what I would argue could be the majority of online conversations of value. Jay Baer’s new book Hug Your Haters features original research he commissioned with Edison Research that correlates to the assertion. Their survey shows that 70 percent of customers complain on Facebook while just 17 percent complain on Twitter. If that is any judge then social listening platforms are missing more than the majority of conversations online.
Facebook Topic Data Presents Two Challenges For Listening Platforms
The challenge social listening platforms have with offering Facebook Topic Data is two-fold. First, DataSift charges them for the volume of data they use, so they’re mostly likely to pass on that per-volume cost to you. For analysis of 10,000 posts you’ll pay X. For 50,000, you’ll pay X-plus (or perhaps times 5). So they can’t price Facebook Topic Data the way they’ve priced their other, more traditional, listening services.
Even if the pricing wasn’t an issue, the data structure and components are different. Without post-level data, Facebook Topic Data doesn’t fit in the average listening platform’s structure. Putting broad, topic analysis into a tool that typically goes to the post-level in granularity is putting a square peg in a round hole. The formats don’t match, so one won’t look right in the other’s clothing.
Social listening platforms now have the challenge of turning their post-centric traditional data into an alternative view to match Facebook Topic Data’s format or continue to offer both as separate products. This will undoubtedly confuse and frustrate end users like you and me, not just because we’ll have to analyze Facebook Topic Data in a separate place, but it will also make it easier for the platforms to present it as an up-charge.
Facebook Topic Data Is More Vital Than What We’ve Had To Date
All of this information leads me to conclude that Facebook Topic Data, even though it does not offer post-level information and only goes a layer or two deep on insights within a given topic set, is more important in insights gathering than any social listening data we’ve previously had access to. I say that for three primary reasons:
- All data I’ve seen indicates that we’re not seeing as much as 90% of the conversations available on Facebook in social listening tools because of Facebook’s privacy walls
- Facebook is, by far, the most used social network for American (and many other nationality’s) users and thus a primary conversation point
- Twitter, which most social listening platforms lead with because their data has always been open and offered through the infamous “firehose”, has evolved to a near-meaningless onslaught of news-ticker updates and celebrity following (A generalization. There’s certainly merit there, but not as much high-quality value.)
My pithy quip from last week on Twitter was, frankly, quite honest. Offering social listening without Facebook Topic Data is like analyzing soft drinks and leaving out Coca-Cola. If you’re looking to social listening for any depth in understanding your customers or the insights online conversations can surface, Facebook Topic Data is a must-have, not nice-to have, data set. Here’s hoping the listening platforms can find ways to offer it at a price point and with a user experience that makes sense.
Have you used FTD yet? Tell us about your experience in the comments!