Welcome to our AMA with the Knotch team! Knotch is a very cool content intelligence platform. We have...
...here to answer questions on content strategy.
David: The starting point is to be clear about the objectives you are trying to achieve. Too often these are not defined, so the question can't be answered about is it working. When the objectives are defined, these can be translated into KPIs, and you should be collecting performance data to allow you to understand if the strategy is working.
We have seen repeatedly that marketers that do not have clear objectives, KPIs and measurement plans are defunded and cancelled within a year or two. So it's a race.
David: There's some great academic work on this subject by Byron Sharp that bottom of the funnel efforts is an optimization play, not a growth play.
He reports that companies that grow have a consistent and significant effort to grow the top of the funnel to make it broad and deep.
Without ToFu, there is no BoFu.
Ellen: We also see in Knotch data how middle-journey content is performing. It’s really easy to miss the middle because it’s not “source” attribution, and it’s not “last-touch” or conversion content. But SO MUCH of your customer journey happens in the discovery/consideration phase. For example, we’ve seen one of our blog posts which has very little to do with our value prop is actually bringing in a lot of traffic, and then referring on to more Knotch-specific topics and articles. Without that data, it’d be impossible to know that article’s value.
This is the article: https://knotch.com/blog/to-gate-or-not-to-gate/
BUT it has very little to do with our actual product
Ellen: I do see this a lot and think it’s a huge missed opportunity. Content and demand gen need to work hand in hand! Each should inform the other. Think of it as a demand gen person: what are you doing if not distributing great content? And from a content perspective, the data a demand gen partner can provide back can be very instructive to what you do next and what is succeeding.
David: There are major differences in both content strategy and measurement because each of these customers have different objectives. For example, some are focused on their corporate reputation, and they will want to know that their content efforts are always enhancing reputation not detracting. Others may be focused on brand consideration, and what to know what content is particular good at this stage of the journey. Others still may have hard core lead generation goals.
What we have seen is that marketers are normally excellent at one dimension of the customer journey, and have blind spots about other parts of the journey where content can be really helpful. For example, post purchase where content can be used to aid product adoption, and to reduce the cost to serve the customer through human channels.
Ellen: Some metrics I’d look into are more qualitative, since you’re working with low numbers. If your conversion and acquisition is good, look next at time spent on page, and also what we call recirculation rates, aka are your thought leadership posts pushing people to stay on your site to learn more?
David: There are two ways to go on this front. On your brand objectives, you can run brand surveys to content consumers on your website to see if their brand impressions are improving. Knotch does this well. The other way to address is to attach yourself to what your sales team is trying to achieve and then report against traffic to product pages or to lead form, or said another way find the bridge that connects content engagement to sales activity, and take credit for what you are driving.
David: It may be easier to check your content taxonomy/tags to see what tags are driving the positivity. You can ask them directly as well. On user retention, you could look at your email nurture programs for clients, and track users who are both retaining or lapsing, and see how their email click throughs vary. This may also give you an indication of who is at risk of lapsing so you work to save them before the lapse occurs.
David: Trusting your gut is taking a real risk. It's worth doing the advanced analytics work to know what behaviors are a pre-cursor of a conversion. For example in the food category, if you were to follow content that was great at generating social shares, you would not have as much impact as following recipe content that drive lots of "print now" actions.
Ellen: Tricky one. My take is, consider your target audience. It sounds like he’s honing in on this in the last phrase, “doesn’t behave the same & is usually not as valuable.” If you are bringing in an audience, but it’s not the right one, that’s not worth your time. You risk optimizing for a dolphin when you’re fishing for tuna, ya know?
Ellen: Our Chief Customer Officer @Andrew Bolton has joined the convo.
Andrew: It depends on your overall strategy and the business goal you are trying to drive with content. If your remit is brand building, then looking at a combination of page views and engagement will give you directional information. If you can, you can also use survey questions at the asset level (part of what Knotch offers) to understand if the content is providing value to the audience. If your remit is driving lower funnel actions (downloads, demo request etc.) then looking at what content can be attributed to those actions. This is also something Knotch focuses on, looking at content attribution across first, last and MIDDLE touch. At the end of the day you want to identify what content is influencing action.
David: Building on Andrew's thoughts, you can create two tables of your URLs - one ranked by page views, the other ranked high value actions, and then calculate the conversion rate. For high conversion rate pages with low views, you need to grow traffic to those pages. For high traffic pages but low conversion rate, you need to find a way to improve conversion. Approach this decile by decile.
Ellen: What I’ve done in the past (the time I refer to as “Before Knotch”) I would look at traffic, and if possible, layer sentiment on. There’s also a lot of collation of data you could do; aka was it popular on LinkedIn? If you work with sellers, did they get more feedback or better responses? I once make an executive briefing that booked 6 meetings off of 12 emails… … but that data would never have shown up in a web portal stat. I also would warn to think critically of a high-traffic piece if you don’t have that sentiment layer, especially if your brand has recently had a stroke of bad news. If you’re getting traffic that poorly affects the brand, back away quickly.
Now that I have access to Knotch, it’s all presented in a platform and what I look for is positive sentiment, scroll depth WITH good time on page, and — if the page has a high-value CTA (book a demo, register, etc), is that a good ratio with the traffic it’s getting?
David: On the subject of writing more, you can topic expansion tools to understand the adjacent topics that your audiences are also interested in.
Also, you may want to write less. Because we find that 5% of content generates 90% of engagement, so the opportunity may be to focus on creating new content types to intensify the impact for your 5%
David: That depends on your customer journey. It takes Lockhead Martin 25 years to sell a fighter jet.
Ellen: So, we’re working on defining our ‘leading indicators’ — what exactly are we going to look for as early signals that the engine is about to turn over. For example, in 30 days we might look at
It’ll be different from place to place, that’s why it’s important to understand your buyer’s typical cycle. Are you able to speed that up?
Andrew: That there is A LOT of content that is very important to the audience journey that doesn't get credit because it doesn’t show up in first/last touch attribution models.
For example, we found that for Fintech companies, it takes an average of 10 content assets over 14 DAYS to get an audience to convert. If that audience has never heard of the brand, make that 24 DAYS and 16 content assets. Most of that 10-16 assets would normally not get credit for that conversion.
Andrew: Unfortunately not, it varies tremendously from brand to brand and depends on the length of the journey, how they got to the page from a referral perspective etc. Its literally why we built Knotch because getting that data out of GA/Adobe is impossible.
Ellen: It’s a balance, in my experience. You can’t write only for the bots; people ultimately sniff it out. This is where the middle attribution would be able to unveil what is really working for the audience when they land on that piece. Ultimately, it needs to resonate so well with the consumer, not the bot, that it’ll be stuck in the back of their mind so the next time they need detergent they think of natural alternatives, and think of your brand as one that cares about that topic.
IMHO, my leadership has wanted content, and marketing in general, to try to map to what’s truly driving $$ (whether sales of detergent or SaaS). If your page is resonating with bots which in turn boosts readership and then sales, that last step is the step you’d want to prove.
David: This is Knotch's arena. We organize content performance around 4 pillars - enhance brand, grow audience, improve ROI and lastly understand the content journey. Rather than include metrics, you should focus on KPIs. All KPIs are metrics, but not all metrics are KPIs.
Ellen: Here’s an example of what you can see in our dashboard.