Harsha Potapragada

‍Correlation vs. Causation

Correlation occurs when no cause and effect can be established between two variables that have a relationship. For example, the level of education of parents is positively correlated with the salary levels of their children. In other words, higher levels of education of parents has been observed in higher salary levels of their children. However, this does not mean that a direct causation can be established. If that were the case, to increase your salary level, you would simply have to get your parents in schools and universities. Another such example of correlations exists between heights and weights. Your height is not causing your weight but taller people tend to be heavier than shorter people. 

Causation means that there exists a cause and effect relationship between two variables. In the education example, a direct relationship may exist between education level of a child and the average salary he earns. Someone who just completed an undergrad and someone who just finished an MBA might get different salaries even at the same experience level regardless of their parent’s education levels. 

Correlation  ≠ Causation

It is important to be able to distinguish between causations and correlations. The best way to differentiate the two is to consider all other factors that are involved in the outcome. For example, there exists a strong correlation between the data for ice cream consumption and murders. This correlation is a complete coincidence. But if you were to apply causation, it becomes worse because then it implies that ice cream consumption leads to murder. 

Applying causation in less subtly absurd correlations can be even more harmful, especially if budgeting decisions are based on cause and effect relationships between touch-points. Ideally, most data analysts avoid establishing causations. First, because its hard and correlations are easier to establish. Second, direct causations are very rare. 

Correlations in B2B Marketing Analytics

Establishing correlations and causations is fundamental to any and all data analysis. Marketing analytics is no exception to this. Correlation insights help marketers make sense of their data points. In turn, this contributes to optimizing marketing efforts and determining the impact of marketing on KPIs and revenue.

In other words, correlation analytics identifies valuable patterns within the story, your marketing data is trying to tell you. Here’s how:

1. Understand the impact of your SEO/PPC 

2. Test campaign decisions during implementation

3. Determine the revenue impact of customer touchpoints 

There can be several pitfalls to correlations data, particularly in cases where coincidences can be mistaken for statistically significant relationships. Some can be very obvious, others are not so much. For example, there exists a strong correlation between the number of pool drownings and films that Nicholas cage has appeared in through the years. Another perfect correlation is between total revenue generated by arcades and CS doctorates awarded in the US. But as is plain, these events have nothing to do with each other. 

Let’s take a marketing example. Say a company decides to mail catalogs of their retail products to their target audience in Karnataka. Soon after, they Ef a stark rise in orders placed from Odisha. Intuitively, the right move would be to send more catalogs to Odisha to support the growing demand for your product. However, as a result of the strong relationship between the two touch-points, correlation analytics would suggest shipping catalogs to Karnataka instead.

Best Practices for Correlation and Causation in Marketing Analytics. 

Avoid confirmation bias

Confirmation bias in correlations data occurs when your data inaccurately confirms a bias. Say, a preferred channel is performing better than another and a correlation that confirms your belief, you are likely to assign causation that isn’t there. 

Anish is the marketing head of Company X. He recently had a celebrity promote X’s product. He worked hard on getting them on board and was sure that it will drive sales. Soon after, he noticed a spike in the number of website redirects from Facebook and immediately assigned the causation for this increased traffic to the celebrity’s campaign. Expecting similar results, he invests further resources and runs another ad with the celebrity. However, there is no change in performance. There is something amiss in the marketing head’s correlation analytics. Instead of checking for causation, he let your subjective assumptions take over. This is confirmation bias in play. 

To assign definitive causation, it is necessary to check for coincidences. In this example, tracking performance data for the campaign across channels is a good way to assign cause to the campaign. Simply put, if the celebrity is affecting more people to click on this ad, then there should be a percentage increase in clicks in all channels that carried the ad with the celebrity. So Anish should’ve tried to corroborate the results, keeping all other things (like the intent of the target audience) constant across all platforms (Google, Facebook, Instagram, etc). On running such an analysis, he notices that only Facebook had a spike in traffic after the first ad, which wasn’t replicated across other platforms or even on Facebook itself when the second ad was shared. On further research, he learns that the platform had made changes to its algorithm around the same time, which seems to have impacted all ads on Facebook, including X’s. 

Using quantitative data from all channels can help avoid making decisions or causations around subjective assumptions. 

You can use a marketing analytics tool like Factors can help you check how a touchpoint is helping or hurting pre-determined conversion goals. The funnel feature allows you to customise your queries to check for specific correlations. Funnels can be created for website redirects, and in this example, the celebrity ad could be compared across channels in a few clicks and Anish could check whether to attribute the change to the celebrity ad or if there’s something else at play.

A/B testing

One of the best ways to establish effective correlation is A/B testing. Let’s say you’re revamping your website homepage and want to test the impact on traffic and conversions. A/B testing involves testing a variable (for example, the position of a “schedule demo” button). This change is tested across two-time frames — pre-change and post-change. 

Let’s change the previous example and assume that the spike in Facebook redirects did not happen immediately after running the ads but happened a few weeks later. In the absence of a proper pre and post analysis, it is human nature for Anish to attribute it to the ad campaign. But if he did a pre-and post-analysis of the impact of ad campaign on redirects, he might find that the cause for the change is something else. 

You can use tools like Factors.AI to record changes like new ads when they occur and use data from the various channels like Facebook as well as your website or conversions to A/B test campaigns. The funnel feature allows you to use campaign naming conventions to get data pre-change and post-change. 

Analyse the impact of correlations across channels.

Looking out for correlations and establishing possible causations can help understand how a specific touchpoint is affecting pre-determined conversion goals. If you want to check impact on goals like say, web event sign-ups, white paper downloads or even deals won, you can use correlation and causation analytics to figure out what touchpoints are saying, helping you schedule demos, what touchpoints on your website is driving down form fills, etc. 

Factors allow you to compare metrics on a week on week basis to catch changes in any of the metrics. The explain feature allows you to check for what URLs or web pages your users have visited before submitting a form. Apart from identifying URLs that have influenced the users to convert, you can also see which webpages aren’t performing well. Weekly sessions data can help see short term changes, apart from A/B testing. Correlations can also be checked at a segment level, like demographics, industries, business model types, etc. 

Choose the right graphs for correlation analysis/reporting

Data collection is only the first step to understanding correlations. The second step is to read the data and share the insights. After getting the insights, you act upon the data as well as build data-driven strategies. To understand how a touchpoint is interacting with each other and the impact of a change on your conversion metrics and revenue, you can use graphs.

There are several kinds of graphs that can be used for correlation analysis.

Time-series graphs:

These reports compare metrics over time periods. They are most appropriate for trends or changes in metrics post a change in a touchpoint or campaign strategy etc.

Distribution Graphs:

These graphs can easily show when there is a correlation. They show changes in distribution against a mean.

Funnel comparison graphs:

These graphs can be used to see a side by side comparison of funnel queries. Say you want to see how ad 1 and ad 2 have impacted the conversions, you can see a side by side strategy comparison of the two. You can also compare the same funnel before and after a specific time period. 

There are also other graphs like relationship graphs that help see the relationship (positive, negative or nil) between two or more metrics. 

In closing...

In the age of data-driven marketing, it is important to know how to treat your data. Every customer journey and every touchpoint weaves a larger story where the channels are connected and touchpoints impact each other to influence each potential customer to convert. Correlations can help bring forth these insights that are invisible to the naked eye and can help you craft a winning marketing strategy for your organisation.

B2B Marketing Attribution (or B2B Revenue Attribution) empowers demand gen teams to map out their customer journeys and connect the dots between marketing and revenue. At a high level, attribution weaves the story that your marketing data is trying to tell about the influence of each touchpoint on core business objectives. As multitouch attribution technology improves, B2B attribution is becoming an increasingly powerful tool for CMOs to wield. Here are 5 way in which CMOs and marketing leaders can take advantage of B2B marketing attribution.

1. A Bird’s Eye View Of Marketing Efforts 

B2B marketing attribution empowers marketers to capture nearly every touchpoint across the customer journey. This is valuable information as most B2B buyers are already halfway through the sales cycle before they explicitly engage with a sales rep. 

Your customers have likely interacted with plenty of marketing channels and content before being picked up by the sales team. Moreover, many of these customers become high intent buyers even before sales or marketing identifies them as such. In such a case, it becomes important to know:

• Which touchpoints help them make their decisions 
• What content or marketing activity influences them to further pursue a product or engage with your company 
• What content helps users narrow down your product over your competitors
• At which touchpoint do customer generate buyer intent,
• At what touchpoint do customers lose this intent

This helps CMOs understand user journeys as well as the efficiency of various marketing efforts in influencing customer decisions. It gives insight into the precise point in the funnel during which to target customers and optimize conversion rates, which campaigns to allocate budget to, which touchpoints are weak links in the buyer journey, and more.

2. Achievable Targets

Marketing attribution, being the data-driven technique that it is, helps CMOs undertake goals in terms of achievability and feasibility. More importantly, attribution uses metrics that can be used to track the progress as well as the success of various campaigns across various channels. This also helps in planning larger goals as well as yearly sub-goals with forecasting, tracking and analysis of campaigns and their impact on revenue. Such goal-setting is not vague as it is thoroughly backed by data. 

3. Improve Productivity and Alignment Across Demand Gen

As your business grows and your marketing campaigns and sales processes start to scale, it can be challenging to track which campaign brought in which leads. Sales and marketing activities tend to become more siloed and communication gaps between the two teams can widen. This can lead to a lot of inefficiencies in the handing over of leads from marketing to sales. Marketing may have insights on which touchpoints impacted most positively to a certain lead that can help sales reps during their engagement. Conversely, sales reps may have insights through their engagements on what information or campaign content helped customers make their buying decision. CMOs can use marketing attribution to align the processes of these teams and improve the productivity of each campaign and each SDR by unifying customer journey reports and touchpoints onto one platform. 

4. Accountability and Reporting

With attribution, marketing leaders can easily generate reports of the most important metrics for their business and board. Moreover, it’s convenient for CMOs to track the performances of their various teams and understand the contributions of each team on conversions, pipeline, and revenue. For example, if a certain blog posts incurs recurring URLs for all leads that have converted, then it is a good idea to give more resources to the content team and perhaps even hire more writers. Attribution gives you hard data on metrics like website traffic and what pages they visited and how much time they spent, whether they filled a form or if they left without any activity, whether they clicked on a discount code or a free whitepaper or if they were not able to notice it — this can give a CMO a good idea on the interface and content of the website. In essence, attribution helps you hold each team accountable by getting a data-backed view of their performances.

5. Driving Growth

Marketing attribution recognizes trends and makes sense of the confusing quagmire of touchpoints in any marketing and sales funnel. Data is unequivocally important in driving sustained, scalable growth. If there is seasonality to when you get more qualified leads or there are specific blog posts, ad campaigns or social media platforms bringing in higher traffic and driving growth, attribution makes it easy to identify these high performing channels and take advantage of them. Most attribution tools have built-in integrations for various ad platforms, social media sites, CRMs and website tracking tools that ensure that regardless of how big you grow, you always have a handle over your customer tracking and don’t lose out on important insights that may get lost in high volumes of data.

In conclusion,

B2B Marketing attribution is a powerful tool for any CMO in 2022 to get the best insights from both internal and external data sources that an organization has. Forecasting, tracking trends, revenue impact, ensuring accountability, saving time and human resources on reporting to focus more resources on analysis and implementation, ensuring accuracy in reporting — are all foundational to building and executing powerful marketing campaigns. With marketing attribution, CMOs can make data-driven, informed decisions and enable their teams to deliver more with less spending and better, useful insights.  

Now more than ever, B2B Sales and Marketing teams share the same objective: drive conversions and revenue. Here are a few reasons why alignment between the two teams is crucial — plus a couple of tips on how you can ensure the same.

But first, let’s discuss sales and marketing misalignment

For the most part, Sales and Marketing interact with the same leads and accounts. Once marketing has identified a high-intent lead, they pass them on to Sales, who are then responsible for converting them to paying customers. Too often, however, relevant lead data is siloed between marketing and sales. Crucial information may be missing or inaccessible for either team. This misalignment can lead to misinterpreted data, poor conversion rates efficiency, unorganised customer support, and ultimately, a loss of revenue and pipeline.
This issue is further fueled when both departments use different tools and platforms, inconsistent data storage practices, and deficient analysis. Another, qualitative symptom of this misalignment is poor communication between teams. This can manifest as dissatisfaction amongst sales representatives with the quality of leads being passed down to them and a similar dissatisfaction amongst marketers for an inadequate number of deals being closed by Sales.

The importance of Sales and Marketing Alignment

Alignment of strategies

‍Often, the strategic outcomes of both sales and marketing are dependent on the toils of each other’s departments. Transparent communication across strategy, challenges, insights and more will ensure that both sales and marketing efforts are complementing each other in driving revenue.

Improve productive prospecting

Often, when sales and marketing are misaligned, the leads coming down the funnel may not seem very valuable to the SDRs. This can lead to:

1. SDRs ignore a majority of the leads being sent to them by marketing
2. SDRs recycling old leads

Both of these symptoms signal inefficient prospecting. Sales and marketing lead to both teams setting up clear parameters for which contacts to send to sales and sales also understands why a certain prospect showed promise from the marketer’s perspective. This leads to increased productivity for both salespersons and marketers as well as improved conversion rates. 

Seamless workflows

Sales and marketing alignment requires alignment across technology and data as well. Data, tools and platforms should maintain consistency across the board. This ensures that information sharing and interpretations are seamless and accurate.

Shorter sales cycles

‍B2B sales cycles tend to be long due to more touchpoints and conversations with reps before the final purchase decision. The process tends to be easier further down the funnel. However, most people avoid initiatives like sales calls and emails. A more collaborative marketing-sales dynamic can help shorten the cycle and improve conversions through content strategy, nurturing activities, etc — that have inputs and perspectives of the salesperson as well as the marketer. 

Tips to improve Sales and Marketing alignment

1. Define common terms

Definitions as simple as qualified leads, MQLs and SQLs can be different for sales and marketing within the same organisation. "This may become a major cause of miscommunication and dissatisfaction with lead quality. Ensuring that everyone is aligned on the definitions and parameters of terms that are integral to both departments can avoid async activity and productivity loss." - says Milosz Krasinski, Managing Director at miloszkrasinski.com

2. Identify target audience 

Aligning the goals of ‘lead generation’ and ‘lead conversion’ begins when both teams sit down and identify the ideal target audience. Dissatisfaction arises when lead identification by marketing and sales are not aligned. For B2Bs, it involves knowing the firmographic features like firm size, industry specifications, titles, revenue etc. This also involves creating core messages together so that both teams are aligned on positioning as a lead goes through the buyer journey. 

3. Define goals and strategies together 

It is imperative for both sales and marketing to be clear on outcome metrics like pipeline and revenue. This ensures that sales have input on defining sales readiness, making communication between teams clear and productive.

4. In addition to sales funnels, perform revenue attribution

The traditional sales funnel is linear in nature as it only comprises the following structure:

Lead->Prospects->Clients. Attribution modelling is a holistic way to look at all the non-linear touch-points during conversions.

5. Create a process for leads engagement

Another consequence of organisational misalignment is the formation of distinct funnels — one for lead generation and another for conversions. Combining these two funnels will encourage comprehensive, high-efficacy engagement across the buyer journey going through the customer journey.

6. Alignment across tools and tech

The best way to ease communication and close down data silos between sales and marketing is to use tools that promote alignment. Attribution and analytics tools that collate data from all touchpoints of the user journey across ads, web, and CRM (ie. both marketing and sales touchpoints) allow seamless data analysis, reporting and insight derivation for both teams. This can promote further collaboration and synergy between both organisations.

Marketing Analytics, Web Analytics, and Customer Journey Funnels

Now more than ever, marketing analytics, web analytics, and customer journey mapping is at the core of every marketing strategy. That being said, tracking, collecting, cleaning and formatting data is a laborious chore. Most organisations, especially SME firms, have neither the time nor the resources to devote to these steps. What's more? Only after you have all the data in place can you analyse, report and optimize marketing efforts.

This is where organisations use self-serve marketing analytics solutions to collect and analyse data. There's no shortage of tools trying to solve for quality, self-serve analytics. Picking the right one, however, can be tricky. One such web analytics solution, Oribi, was recently acquired by LinkedIn for over $80 million. As a result of the acquisition, former Oribi-users are on the hunt for alternate solutions — one of them being Heap.

Heap

Founded in 2013, Heap is a San Francisco based product analytics platform that provides insights and data visualization to track customer engagement with a company’s site or product. It maps user behaviour and enables users to quickly access and organise data to recognise sources of friction within the user journey.

Oribi

Oribi is an Israeli based web and journey analytics platform founded in 2015. Oribi helps track site interactions and key conversions. It also allows marketers to get action-oriented and data-backed trends and insights. Additionally, Oribi helps users understand visitor journeys with intuitive, user-friendly reporting mechanisms.

Heap Vs Oribi: Analytics and Integrations

Although marketers can (and do) use both Heap and Oribi to access user journey data, Heap is marginally more intuitive when it comes to tracking user journeys on web-based products. Oribi, on the other hand, is better suited for pure web analytics.

Another point: Heap does not support direct integrations with ads platforms like Google ads or Facebook ads. To be fair, Oribi’s integrations with Google and Facebook is also set to be discontinued as a result of the Linkedin acquisition. When it comes to CRM integrations, Heap allows for both Hubspot and Salesforce. Meanwhile, Oribi only users to push data back into HubSpot.

Heap works by placing a snippet of code at the top of the site and tracks user journeys only on your website or your product. Its primary use cases are product adoption, product-led growth and funnel tracking for the digital experience over the website or application. Heap also enables site search tracking and campaign management. Oribi does not.

Oribi’s funnel helps marketers understand what journeys buyers are taking and where they are losing more users so that marketers know what they have to work on to improve. Similarly, it gives insights as to which type of content works best and drives more buyers to convert.

Shameless plug but Factors.ai delivers the best of both worlds. Strong campaign analytics, web analytics, revenue attribution, funnels, button tracking and more — across ads, web, and CRM. Schedule a personalized demo to learn more :)

Heap Vs Oribi: User Interface

Oribi has often been praised for its simple to use UI. Heap, on the other hand, has been found a bit wanting in terms of ease of use. Oribi has consistently ranked higher across factors like UI, ease of set-up, ease of admin, real-time reporting, etc. However, Heap may have the edge in terms other features like retroactive reporting, integrations and custom event tracking. Although Heap is a non-code platform, users with zero experience have often found the tool a bit complex to set up and the learning curve steeper than in the case of Oribi. 

Factors also ranks high (in fact, higher than even Oribi) across ease of use, onboarding, customisable filters and breakdowns for reports. Learn more here! 

Heap Vs Oribi: Multi-step digital journeys & multi-channel digital journeys

Both Heap and Oribi help organise and track customer journey funnels. But the funnels are of different kinds.

Heap has been proven to be best for tracking the funnel in a multi-step digital journey, this means that if the user has to take several steps in their digital journey over the application/product or website to get to the end goal or to convert, Heap gives insights as to what steps the user took, in what sequence, when did they complete the goal, where they faced frictions, what step took more time, etc. Their effort analysis features allow you to see what parts of the site give more trouble to the user and why. 

On the other hand, Oribi is preferable for marketers to track the funnel in a multi-channel buyer journey. In other words, if you want to see where your potential buyers are coming from, and what actions they’ve taken before they’ve come to the website, a tool that focuses on tracking multi-channel journeys is more useful. Particularly in the case of B2B user journeys, where there are multiple decision-makers, each of which interacts with your product/service on various marketing channels over a longer sales cycle, multi-channel attribution tracking and efficiency measurement of overall campaigns becomes more important for the marketing team.

Pros and Cons of Oribi and Heap

Heap Advantages

1. Real-time reports: Heap’s auto-tracking and data governance tools ensure that every single event and every single user is tracked and these data points fit into a data structure from the moment that they are collected. This ensures that the reports are always real-time and the data structure is able to adapt even when events change — without any code or engineering support.
2. Allows for retroactive analysis: Since users can retroactively define events and conversions, the data structures and dataset organisations evolve to fit deliverables when they change.

Heap Disadvantages

1. High cost of data storage: Because every single user and every single event is automatically tracked on a real-time basis, this leads to a large quantity of data that has to be stored. 
2. Website analytics focussed: Although Heap supports several integrations, it is more focused on user’s interactions and journeys on the website/product. It also misses ad platform integrations due to the same reason. B2B marketers cannot map out entire customer journeys which in turn, can make it harder to derive insights into overall sales patterns. 
3. Difficult to use: Its UI is a little complex as compared to Oribi. The learning curve is steeper. 

Oribi Advantages

1. User interface: The interface and dashboard are intuitive and easy to use for anyone within the organisation. 
2. Automated data orchestration: Oribi’s ability to automatically send data back into platforms like Hubspot and Google Analytics helps with data orchestration and breaking down of siloes across different storage locations.

Oribi Disadvantages

1. Oribi’s CRM integration allows it to send data automatically to Hubspot but it cannot take data from the platform to integrate CRM data for attribution on the user’s larger customer journey on its own platform. 
2. Oribi’s reporting capabilities have been found lacking as it does not allow for custom filters, breakdowns and formats for data visualisation. The reporting section only allows for pdf reports which can limit how much you can include/exclude.

In closing,

The biggest difference between the two is that Heap is primarily a product analytics tool and Oribi, a web analytics tool. However, because most B2B SaaS products are web-based, the functions of product and web analytics bleed into each other. So Heap is also used for web analytics and vice-versa. At the end of the day, there are several analytics tools that help marketers automate grunt work like data collection, organisation and formatting. They come with different features that help solve various use cases in the day-to-day working of the team. To choose which tool is best for you and your organisation, identify what you struggle with and what tools provide best for such use cases. 

We suggest you check out Factors to get the most out of your data!