Marketing is from Mars, Sales is from Saturn (we’re all from Earth)

Posted on by tombhanna

You’ve all heard the cliché about Marketing and Sales being from separate tribes, but the fractures that can occur within Marketing teams are the leading causes of poor performance, reactivity, and ultimately churn.

The biggest factor for why “Marketing isn’t working” is that organic and paid search, organic and paid social, and organic and paid content live in two completely different worlds within our teams Layering on product lines, customer segmentation, or a B2B vs D2C approach just exacerbate the issues further.

It’s easy to blame other teams for issues, Sales, but I’d like to propose that tooling, specialization, and segmentation allow space for marketing teams to create silos within themselves and for complacency to thrive.

How do we get Marketing talking to Marketing again?

I firmly believe that Marketing leadership needs to lead the charge in paving communication pathways within the organization. To do so, they need to make sure that everyone is aligned towards the same goal.

The quickest way I’ve seen to isolate team members is by setting a short term goal(s) with no consideration for the down funnel journey. An example would be pushing a team to generate 1,000 MQLs in a month under the assumption that conversion rates down funnel will hold. This type of Marketing is fundamentally siloed because it pays no mind towards the sales hand-off, customer relationship management, onboarding process, and worse yet to the lead’s experience.

Why do goals like this exist? Because they look good on a slide, it’s something that’s easy to measure against, and it possibly makes you or the team look good at the next all-hands to say you hit 100% attainment.

Molly Graham put it best in her article, “‘Give Away Your Legos’ and Other Commandments for Scaling Startups.” My take is that while it may seem to be in your self-interest in the short term to show progress towards vanity metrics, they’re ultimately holding you and the company back in the long run. Learning to delegate, share ownership, and collaborate is how we can all grow together as one team.

Too many cooks in the kitchen

Secondly, it’s important for leadership to clearly define roles and responsibilities, procedures, and service level agreements within teams. The clearer it is who owns what and how to include others in projects or requests the smoother your team will work together.

I’ve been in too many awkward meetings where a great idea is presented, but no one steps up to own its execution for fear of stepping on another’s toes. Or, even worse still for morale, when the wrong person does step up. This can lead to frustration by the ‘true’ owner for not being included and cause further breakdowns within teams.

How many times have you come up with an idea only to find out that it was already done before (or in a similar fashion)? And, yet, you and others weren’t aware on your team. Rather than shooting these ideas down, I can’t help but think that there is a clear breakdown in how work is being documented and shared. What could be learned by saying instead, “Yes, we’ve tried that before. Let me send over the results. If you’d like to you can set up some time on my calendar and we can review them together to see if there is anything new to be learned or tried.”?

The thread across Marketing teams

I’ve typically pictured a Marketing team as something like a train being loaded up and taking goods to the market or as a line moving from left to right with clear delineation between each stage. While this paints a nice picture, it fails to capture the reality that agile and capable Marketing teams face today when it comes to being and staying relevant.

I’ll be the first to admit that there are fundamental differences in how one manages SEO and SEM or Organic and Paid Social, but from far away they’re two sides of the same coin. Some of my best work has come from sharing insights not only across the two but across the organization. One such instance involved taking a piece of content that was being heavily engaged with organically and creating a remarketing campaign from it. Another included shopping my top-performing keywords across the organization only to learn several new opportunities for long tail versions.

If you want to break down these siloes and perform truly exceptional marketing start by setting the correct goals, define roles, and being open to learnings, wherever they may come from.

Red flags when auditing organic & paid channels

Posted on by tombhanna
Red Flag

When interviewing or starting a new job I’m frequently asked what would you do with a limited budget to hit our upcoming quarterly goals around leads? I think that before anyone can begin to answer that question effectively they first need to understand how that company’s marketing is perceived by their customers today. It is by researching this that it becomes clear that many companies are not marketing themselves effectively.

In this post, I’m going to break down my approaches, provide tools, showcase strategies for identifying these red flags, and in turn propose alternative ways of marketing.

Auditing organic channels

Who are they?

I start by first getting to know who is asking the question. The first place I’d look is on a company’s About Us, Careers, News, or Press pages to better understand who they think they are. I especially like to see how they talk about themselves in recent press releases as sometimes websites can be a bit out of date.

From there I like to branch out onto their social media accounts to see if they are consistent in this messaging and/or how it might differ across their banner and about sections. I should have a good idea of their story at this point.

What is ‘their’ story?

That said, I can’t tell you how many brands I’ve come across that they’re “The #1 leader/solution/provider in X Industry.” While not as egregious I also get concerned when companies are too focused on their own products, “Our *insert confusing string of big words* solution helps you to *insert fluff*.” It’s lazy and begs the question; “What’s in it for me (WIIFM)?” It’s also easy to replicate from the competition and does nothing to differentiate their brand. So, how would I recommend fixing this?

Who is the customer?

The first approach I’d recommend is taking a page from Ann Handley in making the customer the hero of the story. Recommend they get on the phone or visit their current customers to understand why they chose this solution.

It may be because their solution is the “fastest” on the surface, but underneath being fast may have helped the customer to get to market or iterate and in turn beat out their competition. Besides gathering key insights into how to answer WIIFM this is a great opportunity to generate content and insights to fuel future messaging and campaigns.

SEO

There are a number of free tools to pick from when it comes to auditing SEO. My purpose here is to understand what the top keywords they’re matching for are and if they’re in alignment with my expectations and the story the company is putting out.

Once I have an idea what the top keywords are I then look at the “related searches” or “people who ask” results for ideas on where they could beat the competition or gain a foothold with organic content.

That said, my favorite auditing tool by far is Page Speed Insights by Google. By running a domain through this engine you can see if the company places any stock whatsoever in SEO and can basically tell if they have anyone on their team that knows web development. More often than not pages will have high paint times due to poor CSS and Javascript. Knowing that a company scores a 23/100 on this report makes it quite clear that they do not prioritize this channel. You can infer what you will from there about them and their team makeup. The mobile snapshot is important context as well in this modern world.

*Note: It’s clear that this company doesn’t care about SEO based on page load times, image optimization, and a poor mobile optimization score.

Social

I won’t claim to be an expert when it comes to creating community on social, but I do know that is its purpose. When I’m auditing I too often see companies pumping out content on these channels, or pasting the same content across multiple channels, with no regard to engagement with their audience. Even better still is when the 2-3 likes each post is getting is the same people every time and they’re the employees or owners. Social media channels have a vested interest in showing users engaging content so be warned that if only employees engage with content then only employees will see the content.

The solution here goes back to the hero story in that it’s okay to talk about yourself and your solution some of the time, but the majority (80/20) should be spent providing value to your community. I would also recommend prioritizing the channels your customers are on.

If a company doesn’t yet have a community my recommendation would be to tap into similar ones that exist already such as a LinkedIn Group to get ideas on the types of things your customers like to engage with. It may become clear that your Marketing or Content teams are not capable of communicating effectively with this audience. If that’s the case, you should look to recruit internally or hire someone who reflects the voice of the customer. I’ve seen this approach work well in highly technical fields.

Auditing Paid Channels

Goal Setting

I could write several pages on how to audit paid channels, but when it comes to looking for red flags the first was presented in the prompt for this post. Any company setting goals around leads, MQLs, or even pipeline should be tying their goals back to Revenue.

Those other metrics can provide good leading indicators in regards to a campaign’s performance, but if a campaign isn’t profitable you should find better investments. Tying marketing efforts back to revenue also unites the entire organization behind a goal and makes it a team effort.

I’d also like to add that attribution shouldn’t get in the way of good marketing. Just because you can’t measure something directly does not mean it isn’t working. If a company is concerned about such things my recommendation would be to turn off the campaign and see if you take a hit or to ask your customers if the campaign was valuable to them in their decision to purchase.

Paid Search

If you’re looking to learn how to do a thorough audit I recommend getting certified through Skillshop. These will depend, but common red flags I notice are if an account is using broad match keywords, if it is not integrated with their CRM, if Ad Quality scores are low, if Search Impression Share is low, if they’re using automated bidding strategies without knowing what their CAC or ROAS should be outside of the platform, not using extensions, if their keywords are targeting consumer or top of funnel leads, underutilized negative keyword lists, and if they’re not using segmentation.

If an account is this far gone I would recommend pulling in and focusing on 3-5 core campaigns with exact and phrase match long tail keyword strings using manual bidding to reset your baseline. In general it’s better to focus on too few than too many keywords and completely own those and build campaigns effectively around them (The same advice I provided around social).

Paid Social

*Note: The Drift logo/brand is repeated at least 5 times in this small post. The majority of the image is text. The title and headline are simple and don’t talk to the customer. They are vague and don’t clearly represent the value gained from reading it. The image is static. This is the type of buzzfeed like post that will get a ton of engagement because the brand is well know but the majority of leads will be worthless or a drain on resources.

Common mistakes I see are too much text on images, repeating the text from the title and headline on the image, only leveraging static images, asking for too much for little return, and gating everything. Given limited space and short attention spans on these channels these minor mistakes can tank a post. They also flood your database with low quality content leads that have no interest in purchasing that will waste future marketing spend and sales time.

To avoid these issues I recommend mocking up what the post will look like in its entirety before you begin design work. The majority of your time should be spent on building an enticing Title, and running experiments around it, as well as on the content itself.

When it comes to the content itself a static image that leads to a gated landing page is just not going to perform. Your first priority here should be to provide value to the prospect on this channel and engage them where they are, not interrupt their experience. In many instances that means taking this another step further and building a video or gif to explain your value proposition along with giving away the content for free within the post.

While you may not be able to tie attribution back to this recommendation I firmly believe that it is the future and the right way to market to your prospects.

Other Channels

If a company has these core paid channels down you can start auditing their marketing across other search engines, app stores, review sites, podcasts, publications, events, emails and nurtures, direct mail, chat, and more. Wherever their customers and prospects are engaging is where they should be.

Digital Marketing Certifications & Resources

Posted on by tombhanna

For those of you that are tired of binging Netflix, I’ve put together a comprehensive list of MarTech tools, programming languages, and advertising platforms with links to their corresponding certification and/or resource centers. Let’s start upgrading our skills together!

Feel free to make a copy or leave a comment in the Google Doc if you think of any others that should be added to the list 🙂

Getting Started with Testing – Part 2

Posted on by tombhanna

In part 1, found here, we worked on how to create a test, what metrics you should be looking at, and whether or not a test is worth running. Today, we’re going to take a look at some of the different testing methods, their strengths and weaknesses, and the value of seeing a test through to the end.

Split Testing, A/B (C, D, E..) Testing, and Multivariate Testing

A common test you might run would be to modify the form on a popular landing page under the assumption that less friction would increase performance.

Say your current form currently asks for First Name, Last Name, Email, Company, Industry, and Phone Number.

Whereas, the new form you’ve proposed only asks for First Name, Last Name, and Email.

This would be called a split, A/B, or multivariate test. These types of tests are composed of three parts:

  • Control Group(s): The group in the test that does not experience any change.
  • Experiment Group(s): The group in the test that does experience a change.
  • Variable(s): The elements or features that you changed.

Using the scenario above, to run a split test, you would split incoming website traffic equally between a control group (people who will land on the original page and form) and an experimental group (people who land on a new page with the new form) and then measure the difference in conversions the variable(s) (the removal of several form fields) had.

For A/B/C/D.. or multivariate testing you would still have a control group but also set up several different experimental groups with their own variable changes.

This has the added benefit of helping you to not only hone in on precise changes such as only updating one field at a time, but also allows you to test multiple hypotheses concurrently.

However, the drawback is that less traffic gets split between each new experimental group thus increasing the length of time you will need to run each test to see statistical significance.

That is why I normally recommend sticking to A/B or split testing unless you’re working on optimizing a page that receives a lot of traffic. A split test duration calculator can help you make that decision.

The Need for Control Groups

Another reason I recommend these tests is because they utilize a control group. There are many external, internal, and unseen factors that can either positively or negatively skew the results of any test.

Let’s assume that you read some articles that said that removing form fields was the right thing to do as well as saw several of your competitors creating shorter forms. So, instead of running a split test, you just go ahead and roll out the new form with a reduced number of fields ASAP.

For the sake of success, we’ll say that you have had this new form up for a month and that the conversion rate is a 3% increase in signups month over month. A common inference would be that the new form is a great success and that you should try removing even more fields to see even better conversion rates.

However, you truly have no way of truly knowing if that is the case.

What if the test month landed during the largest industry conference of the year? The sudden increase in qualified traffic may be due to your sales team’s performance at the event and subsequent word of mouth. Your website will have exploded with leads who are converting like never before. Looking back at the baseline from the month before you’ve crushed your goal. However, the reality is that this was going to happen even if the page stayed the same. In fact, performance may have been even better if you didn’t make any change at all!

Or, what if a competitor had a scandal recently and their customers decided to explore new offerings? What if a new article, advertisement, whitepaper, product, or feature launched and that was the cause of the increase in traffic that was more likely to convert.

On the flip side, say that the new form performed worse! Maybe you had a recent product launch that had some bugs in it and led to bad PR and word of mouth. Do you just write off the dip in performance as a fluke? How will you know if the original wouldn’t have also dropped the same if not more?

There are just too many questions to answer and not enough data to do so when you don’t have a control group.

The Waiting Game

This is probably the most difficult and most crucial thing to have when running any test. Patience.

As your test runs it’s not uncommon to notice the performance of either the control or the experimental group sharply rising or falling.

You may be tempted to end such an experiment thinking that you don’t need to wait for statistical significance.

Statistical significance is the likelihood that the difference in conversion rates between a given experimental group and its control group is not due to the null hypothesis or chance.

I would urge you not to do so. Ending an experiment too early, just like not using a control group, means that you’ll never know if the results were due to the change in variables or just random chance.

How to Win the Waiting Game

In another scenario, you may feel external pressures from managers, your budget, or monthly goals to make a call and just move on. I know I’m guilty of ending a test or two before its time.

While you can generally know how your performance will be going into a test, assuming that no change occurs, it’s much harder to manage other people’s opinions on the value of an experiment.

It’s not uncommon for a HiPPO (highest paid person’s opinion) or a manager to emerge and ask for a test to end in the interest of short-term results. This can come under the guise of praise for increased performance or reprimand for losing the company’s money. Don’t be fooled. As tempting as it may be to end the test right then and there it’s important to remember why you ran this test in the first place and to convey that importance to others. Explain clearly:

  • The data you observed that led to your hypothesis.
  • How if you end the test now you’ll never know if your hypothesis was correct.
  • That any further decisions based on false information will hurt the company in the long-term.
    • This could end up costing us more than we’ll make right now.
    • The information we learn will be more valuable than the short-term gains.
    • As the saying goes, don’t build your house on sand.

If you’d like to start forming some tests of your own, all of the work I’ve shown here can be found in my testing framework sheet. You’re welcome to make a copy and use it. I hope that it can help you get started testing as well.

Digital Marketing. “What’s in a name?”

Posted on by tombhanna

I’ve been fortunate to have been able to wear many hats in my marketing career. From my beginnings in social media, content and email marketing, to product, brand, and event marketing, and most recently specializing in digital, demand generation, and operational marketing.

However, as I’ve become more specialized I’ve noticed some confusion spring up around what exactly a digital, demand generation or lead marketer is and does.

As many new marketers start their careers and organizations look to build out their marketing teams I thought it important to reassure them that choosing the right title is not what is important. What is important is identifying the work to be done. To quote the great William Shakespeare;

“What’s in a name? That which we call a rose. By any other word would smell as sweet;”
– William Shakespeare’s play Romeo and Juliet

Here are just a handful of different titles I’ve seen for the same position at different companies:

  • Digital Marketing
  • Demand Generation
  • Lead Generation
  • Email Marketing
  • Growth Marketing
  • Conversion Rate Optimization (CRO)
  • Search Engine Optimization (SEO)
  • Search Engine Marketing (SEM)
  • Marketing Data Analyst
  • Marketing Operations

Clearly, there’s no deficit of marketing roles available. And, while adding such labels can help to distinguish and separate work at larger organizations, they may have the opposite effect and cloud our understanding of a person’s role in smaller ones. That’s why I’d like to take some time to define each of the above.

Digital Marketing

Digital marketing generally refers to marketing efforts performed through digital channels, mainly on the Internet. This can include but is not limited to search engines (both paid and organic), content, websites, email, social media, mobile, and marketing automation.

The reason why I love this role so much is that it’s all-encompassing of modern marketing strategies and tactics that revolve around reaching people where they are. In front of their computers and phones.

Digital marketing is not just about leveraging any one of the above channels on its own but about how you can utilize them all together to connect with a lead or customer at the right time and in the right place to provide a personalized and relevant experience with your brand or offering.

Demand Generation

Just as the name implies, demand generation is about building awareness for your company’s products and services. It’s their responsibility to fill the top of the funnel. This is typically accomplished through inbound marketing tactics such as search and display ads, content distribution, referrals, and other tactics.

sales_funnel_text

What makes demand generation unique though is their holistic approach towards optimizing every touch point in the sales funnel. This is done not only nurturing the leads through email automation, targeted content, and retargeting advertisements, but by collaborating closely with sales after a lead has been handed off. At a basic level, this should include passing off qualifying information to sales such as;

  1. Lead Scoring
  2. Lead Ranking
  3. Lead Routing

Lead Generation

You typically hear lead generation used interchangeably with demand generation.

However, the key difference between the two is that rather than focusing on the entire funnel, lead generation is solely focused on driving top of funnel leads through form fills, list building, and other advertising efforts. It is only with the inclusion of lead management that the two become indistinguishable.

Email Marketing

Email marketers utilize emails to build loyalty, trust, and brand awareness for their company’s products and services. To achieve this they build a list of subscribed recipients either through purchased lead lists, subscription forms, or current customers.

Their goal is to create engaging emails by leveraging design elements and great content for the purpose of increasing open rates, click through rates, and ultimately drive conversions.

To achieve this, email marketers need to segment their email lists so as to deliver targeted and personalized messaging at the right time across various audiences at various stages of the sales cycle. This typically requires the use of a Marketing Automation platform so that they can automate the process of identifying and sending relevant emails as triggers are hit as well as analyze open and click rates.

If you’re looking to move away from batch and blast sends to your entire subscriber list (which most certainly are hurting your open rates and wearing on your subscribers’ patience) I’d recommend checking out MailChimp.

For those of you interested in leveling up your email design and learning a bit of HTML I recommend checking out Litmus. Their HTML/CSS Builder is invaluable if you’re building a custom email template.

Growth Marketing

Similar to demand generation, growth marketing targets the entire funnel. However, it often goes a step further looking at how customer success teams or the product affects conversion and retention. This term essentially refers to anyone on marketing who runs A/B tests and optimization at any point in the funnel.

If you’re interested in becoming a growth marketer I suggest you check out my previous post, “Getting Started With Testing.”

Conversion Rate Optimization (CRO)

CRO is about optimizing your advertisements, landing pages, emails, or virtually any touchpoint that a lead can engage with so as to increase the % likelihood that they will convert into a customer.

This includes auditing your entire funnel to measure effectiveness and ROI, as well as leveraging the results to drive changes that will improve performance.

Search Engine Optimization (SEO) / Search Engine Marketing (SEM)

Search engine optimization (SEO) refers to organically optimizing your content to be indexed by search engines and more easily found by your audience.

Search engine marketing (SEM), on the other hand, is paying to increase your contents’ visibility on search engine result pages (SERP).

The responsibilities of an SEO marketer include improving the ranking of a website on a SERP through the following tactics:

  • Keyword research
  • Reviewing your site content or structure
  • Technical advice on website development: hosting, redirects, error pages, use of JavaScript, etc.
  • Content development
  • Management of online business development campaigns

The responsibilities of an SEM specialist involve using the same tactics through paid advertisements to increase website visits, conversions, and ultimately revenue.

Google offers a number of free resources to help you get started with SEO if you’re interested in learning more.

Marketing Data Analyst

When you want to understand the impact of your marketing campaigns or model out the potential for future initiatives you turn to your marketing data analyst.

They’ll use their quantitative and qualitative skills to identify how campaigns are performing or model industry trends so that you can make a more informed decision. Utilizing predictive modeling or analysis it’s their job to not only interpret the data but to help others understand what the numbers mean so that the marketing department can make better business decisions.

Marketing Operations

Similar to a lot of other roles, marketing operations management is focused on optimizing every stage of the funnel. The goal of marketing operations is to increase campaign efficiency and drive consistent results. It requires that they establish:

  • Plans and forecasts grounded by data
  • Key Performance Indicators (KPIs)
  • Repeatable best-practice processes

This is accomplished through the use of software tools such as Marketing Automation (e.g. Marketo), Marketing Resource Management (e.g. Bizible), and/or Customer Relationship Managment (e.g. Salesforce).

The Take-Away

The importance of hiring a marketer in any of these roles is in what they can do, not what you call them.

For those organizations getting started with marketing my hope is that you learned a little bit more about the different roles that marketing can play and as such are better equipped to identify the gaps and opportunities available to you for growth.

And for those of you starting your careers in marketing or moving in a new direction, I’d urge you to branch out and explore some of these similar roles and responsibilities as you grow.

Getting Started with Testing – Part 1

Posted on by tombhanna

In my last post, “The Million Dollar Question” I discussed how teams could better brainstorm and prioritize ideas for running more effective marketing campaigns. Today, I’m going delve deeper to show you how to set up a framework for experimentation.

1. Before you begin

The foundation for any test is its hypothesis and as such it’s important that you spend the time to form one before running your test. Trust me, there’s nothing more frustrating than making a change and not knowing why it either worked or didn’t. As a quick refresher, a hypothesis is;

“a prediction you create prior to running an experiment. It states clearly what is being changed, what you believe the outcome will be, and why you think that’s the case. Running the experiment will either prove or disprove your hypothesis.”

– According to Optimizely

Let’s assume for the purpose of this post that you’re thinking of rolling our a chat portal on your website.

You log into Google Analytics and see that net new visitors are spending more time on your homepage and that they are the bouncing and exiting from your site without engaging with your CTAs at much higher rates than returning visitors. That’s not good!

Using this data you infer that new visitors (we’ll call them leads moving forward) have questions that you just aren’t answering on the homepage. They’re searching for answers based on their long time on page, but since they can’t find those answers they are bouncing and exiting the site without visiting other pages.

A hypothesis you might form based on this observation is that if we give leads more ways to reach out to sales and ask questions by adding a chat portal then they may be more likely to engage with your main CTA on the homepage, free-trial signups.

  • What is being changed? – Adding a chat portal.
  • What do you believe the outcome will be? – Increased free-trial signups by new visitors. (I recommend that you quantify outcomes as much as possible and I’ll show you how to in section 2.)
  • Why do you believe that? – Because new visitors bounce and exit the page despite spending more time on it than returning visitors.

1.2 What are you really solving for?

Now that you have your hypothesis and an observation based on data, you need to identify how you will measure success.

Ask yourself these three questions to determine whether or not you want to proceed with a test.

  1. Is this test a high priority? To clarify, does it directly affect your KPIs? Another way to think of this is how close is it to the bottom of the funnel vs. the top?
  2. What is the business impact of this test? What is the return on investment?
  3. What is the technical effort required to see this test through?
Experiment Priority Business Impact Technical Feasibility Total
Add Chat to our Website P1 5 4 9
Add Customer Logos next to our Sign Up flow P2 4 2 6
Add social sharing buttons to landing page P3 2 3 5

To see a practical application of this methodology visit my Google Sheet and explore the first two tabs “Prioritizing Experiments” and “Experiment Brainstorm.”

1.3 Three tips for setting priority

  1. Pages with the highest amount of visitors and that are closer to the bottom of the funnel should have priority when testing. This is because these tests have the highest impact on revenue as well as enough volume to show results in a shorter amount of time. Helping you to learn and move forward quickly.
  2. Tests that are simple should be prioritized next. This is so you can achieve scale and velocity. Quick wins will keep you motivated and quick losses will help you learn so you can pivot and test again.
  3. Highly technical changes should come last as you don’t know how a test will perform until it is run. As a rule of thumb, only 1 in 10 tests wins and these types of tests force you to place all of our eggs in one basket.  As such, it’s safer to learn more before you take on that risk.

2. How will you measure the results?

Now that you have the framework and a hypothesis that you’d like to test it’s time to set up your experiment. In the past I’ve used Google Analytics, Optimizely, HEAP, and Google Optimize to run my experiments. Each of these platforms varies based on the features, flexibility, integrations, and customization they offer. Functionality is usually in direct correlation with their price tag.

If you’re a just getting started I’d recommend using Google Analytics and Google Optimize because they both have free offerings and can be integrated into your site through the header tag (or Google Tag Manager), reducing the technical effort of getting started with testing.

2.1 Establishing baseline metrics

The following calculations are in the testing framework sheet for reference.

Using the observations that you’ve made when forming your hypothesis you’ll now need to identify the KPI that you are solving for. I’m going to be making some assumptions with the data for the purposes of this test, but you should be grabbing these metrics from Google Analytics.

  • Looking at your past data what is the Lead to Conversion Rate on the homepage?
    • 4.66% Conversion Rate (Made up Google Analytics Destination Goal)
  • How much of a Lift do you expect this experiment to have on the Lead to Conversion Rate?
    • Targeting a 20% Lift based on past tests (Made up)
  • What is your Target Conversion Rate?
    • Baseline Conversion Rate * Targeted Lift + Baseline Conversion Rate = Target Conversion Rate
    • (4.66% * 20%) + 4.66% = 5.59% Target Conversion Rate
  • How many daily unique visitors does your homepage receive?
    • Monthly Unique Visitors / Average # of Days in a Month = Daily Unique Visitors
    • 14,326 / 30.6 = 468 Daily Unique Visitors
  • How many variations will you be running including the control?
    • I tend to run 2 variations at a time, usually referred to as an A/B or split test, to speed up and simplify the process.

Using these numbers, you can now calculate how long it will take to reach statistical significance for this test.

2.2 Calculating statistical significance

“Statistical significance helps quantify whether a result is likely due to chance or to some factor of interest. When a finding is significant, it simply means you can feel confident that’s it’s real, not that you just got lucky (or unlucky) in choosing the sample.”

– from Tom Redman, HBR

For this step, there are plenty of free calculators online. However, I prefer the VWO split test duration calculator. If you enter all of the bolded numbers from the last section into the calculator you’ll see that we need to run this test for 35 days to make a decision with any confidence that this test has either succeeded or failed.

However, the reality is that this process might be shorter or longer depending on the number of page visitors and the rate at which they convert. I would add a buffer here and plan to run this test for a month and a half.

2.3 Is this test worth running?

Before you spend any more time on this test, let’s calculate if it will have any impact on your KPIs.

  • What is the Planned Sample Size?
    • Daily Unique Visitors * Planned Days * Traffic Allocation = Planned Sample Size
    • 468 * 45 * .5 = 10,534 Planned Sample Size
  • How many Additional Leads will this test generate?
    • (Target Conversion Rate * Planned Sample Size) – (Baseline Conversion Rate * Planned Sample Size) = Assumed Additional Leads Generated
    • (5.59% * 10,534) – (4.66% * 10,534) = 98 Assumed Additional Leads Generated
  • How many of those Leads will become Closed Won? (Assuming 8%)
    • Additional Leads Generated * Lead to Won Ratio = Assumed Additional Closed Won Leads
    • 98 * 8% = 8 Assumed Additional Closed Won Leads
  • How much additional Monthly Recurring Revenue (MRR) will this generate? (Assuming the average lead generates $78 in MRR)
    • Assumed Additional Closed Won Leads * MRR = Assumed Additional MRR
    • 8 * $78 = $624 Assumed Additional MRR

These numbers tell me that not only does this test run the possibility of significantly increasing the number of leads generated each month, but it could also add a substantial amount of MRR.

Whereas before you might have simply run this test based on a hunch that it could work, these additional data points can help you to quantify the potential payoff to better be able to either make the case for or decision of whether or not you should start running this test.

3. Continued…

In Part 2 I go over different testing methods, their strengths and weaknesses, and the value of seeing a test through to the end.the common pitfalls of running experiments as well as what to do once your experiment ends.

If you’d like to start forming some tests of your own, all of the work I’ve shown here can be found in my testing framework sheet. You’re welcome to make a copy and use it. I hope that it can help you get started experimenting as well.

P.S. If this topic was of interest to you I’d recommend checking out the Resources over at Wider Funnel, Optimizely, and VWO.

The Million Dollar Question

Posted on by tombhanna

I’ve been going to a lot of interviews lately and one of the questions that really stood out was, “What would you do if I were to give you a marketing budget of one million dollars?”

Of course, my first reaction would be to jump over the moon a couple of times and make an embarrassingly poor attempt at cartwheeling around the office in joy while blasting Queen’s “We Are The Champions” through the Sonos. But once the excitement wore off I expect a lot of marketers would be nervous being responsible for such a large spend with little to no direction and all eyes on them.

Never fear, tried and true methodologies are here to help guide you through this problem and turn it into an opportunity that will make your career. So, let’s get started. What do you really do first?

Research, Research, and More Research

You’ve heard real estate agents spout location, location, location for years as their north star for investing. Well, as marketers I would say this is about as close as we’re really going to get as it depends on if the sky is cloudy, the captain ate dinner, or if you just lost your other good eye in a hook fight.

First things first, go out and look at what others are doing. Keep in mind that these companies don’t have to be in your same industry, and I’d actually recommend that they aren’t so that you can get a fresh perspective. The goal here is to get some ideas about what might work.

Another thing you should be doing at this stage is looking back on past proposed or failed ideas and thinking through why they never took off. You’d be surprised at how many experiments fail because the market, product, spend, or any other number of outlying factors got in the way, but might just be the perfect fit for you now.

At this point, I would recommend having a shared document with all of your potential ideas written down. For this first round, I put together a rather loose scoring system next to each idea based on technical feasibility (how likely it is that we can build this given limited engineering resources), and impact on the business (how likely this is to generate revenue). I use a 1 to 5 scale for each of these with the highest impact and easiest to implement is 5 each and lowest impact and hardest to implement scoring 1 each. By putting a total column after these you can prioritize the campaigns that score a 10 and work backward from there.

Idea Business Impact Technical Feasibility Total
Add Chat to our Website 5 1 6
Add Customer Logos next to our Sign Up flow 3 2 5
Change the Color of the CTA button 2 5 7

Now that you’ve got a list of ideas floating around it’s time to call a meeting of the minds. Get your best marketers, salespeople, product managers, and other creative types in a room and tell them about the opportunity that awaits them. They have this chance to influence how marketing is going to be spending the companies money.

Share the document you’ve been working on and then start them brainstorming. I actually recommend hiding any scoring you’ve done up to this point so as not to discourage any new ideas or get caught up arguing over scoring just yet. This meeting should be all about proposing new ideas and building upon the suggestion you’ve already built.

Start Testing

Now that you’ve got some ideas down on paper and some solid ones at that. Don’t just let them go to waste! Too often we get caught up wondering if a campaign right or wrong thing to do we forget that really important, what works and what doesn’t.

Set aside a small portion of your budget and start experimenting with these new ideas. Start with the potentially most profitable and easiest to implement to get some quick learnings and wins. This is the perfect time to share this information with the team to let them know how their awesome suggestions turned into actual revenue (and make it that much more likely that they’ll want to participate in your next brainstorm)!

In the meantime, start working on building out those hard to implement and more likely profitable programs using the additional time and new information you’ll be gathering along the way.

Now that you have some wins under your belt you’re going to have a much better idea of what really works with your audience and what doesn’t work. Which brings us to our next step.

Optimization and Why You Should Always Keep Testing

Now that you know which channels and campaigns work in your marketing mix it’s time to push more of your budget behind them and see just how much you can squeeze out of each. It’s here that you should be looking at the amount of revenue each opportunity being generated is bringing in so you can look to optimize these channels.

When I’m optimizing a campaign I generally focus on either increasing the amount of revenue per opportunity, increasing the lead to opportunity ratio, or reducing the cost per lead. Each of these data points tells a different story not only about the campaign but about the customer too.

But just because these channels are working today doesn’t mean that they’re always going to be this efficient or profitable for you. Your competition will soon enough get a whiff of what you’ve been cooking up and come over to steal a piece of the pie. That’s why it’s so important to continuously be iterating and launching new tests and experiments on both existing and new channels.

Aligning Marketing & Sales

Posted on by tombhanna
sf-skyline-night

Have you ever heard your marketing team say that their latest campaign or program would have worked if only sales had been able to hit their numbers? Or worse yet, sales complaining that marketing is passing them leads that they can’t work? 

“51 percent of marketers are not satisfied with the level of communication between the teams and 53 percent of sales professionals are not pleased with marketing’s support,”

– Study by Pandadoc

It’s understandable that such frustrations surface. Both of these teams have enormous pressured to drive revenue, usually with limited resources. While marketing is focused on creating a strong brand, acquiring new leads, and nurturing them for the handoff to sales, sales are focused on hitting their monthly or quarterly quotas.

Here’s what typically happens: marketing passes off what they deem to be a marketing qualified lead (an MQL) to sales as a potential opportunity and voila their job is done, right? Wrong. This is where the rift between these teams begins. If sales can’t convert this lead into an opportunity, they will blame marketing for handing off of a bad lead. Likewise, marketing then blames sales for losing a lead they spent time and money on.

So, how can we prevent this from happening?

Enter Marketing Automation

I recommend using a customer relationship management (CRM) system to store, collect, and track a leads opportunity and activity information. This will allow these teams to segment leads based on industry, company size, job role, and several other demographic, behavioral, or other qualifying pieces of information unique to your business. Ensuring that both teams have all the information available needed to properly qualify and prioritize leads as they move through your sales funnel.

Defining an MQL

The handoff process should start with a conversation between marketing and sales to make sure that everyone agrees on what an MQL looks like. These are leads who have:

  • Downloaded several specific pieces of content
  • Engaged with your nurture emails
  • Visited certain pages on your website
  • Match the demographic makeup of your current customers
  • Indicated that they’d like to hear from sales

As such, sales can focus their efforts on converting leads that are more likely to become paying customers, and marketing can continue to nurture those that are not yet ready.

By aligning sales and marketing at this stage of the funnel, marketing is able to help sales to close deals more quickly and improve the organizations’ performance.

sales_funnel_text

“Funnel management provides insight into which sales and marketing processes are effective and increase deal flow, as well as insight into how efficiently customer opportunities are moving through the stages of opportunity development.”

– Stated in the Journal of Digital Asset Management

Creating a Service Level Agreement (SLA)

The conversation between sales and marketing shouldn’t end here, though. It should also include what happens to a lead once it has been handed off to sales.

“Service level agreements (SLAs) need to be outlined for each phase of the revenue cycle. When an MQL is handed off to the sales team, what happens? How long does the sales team have to make contact? What happens if they don’t? If sales recycles the lead, how long does marketing have to respond, and what is their next step?”

– According to Marketo

Committing to a Culture of Experimentation

Since all this information is stored and easily accessible through the CRM, marketing and sales can answer these questions by reviewing the same lead information and reports. This helps to inform future content, additional requirements for nurturing leads, or new segments to pursue. It also ensures that no lead is left behind and that every lead is given a personalized experience.

By regularly examining how aligned sales and marketing are in their definitions of marketing qualified leads and sales qualified leads, your business will generate and pursue the types of leads that will ultimately generate revenue for you, saving your organization time and resources.