Cracking the code on eCommerce: Top 10 tips and tricks

eCommerce, B2B or B2C, can be exciting but also overwhelming. No matter how you view it, it is understood to be one of the most difficult and daunting marketing endeavors because with countless number of brands vying for buyer attention, simply having a great product isn’t enough.

The difference between a struggling startup and a thriving online store requires both a well-defined strategy and a well-oiled operational practice. One without the other in an eCommerce venture will mean the brand is limping on one foot and outcomes will reflect this weakness.

This article provides the top 10 essential tips that cover a wide range of eCommerce strategies and tactics such as how to manage a sales funnel or how to develop a practical promotions strategy. These tips are proven strategies to help you build a profitable and sustainable e-commerce business.

Let’s go!

1 > The limited utility of Branding

So, so many agencies or inexperienced eCommerce managers often push the importance of branding in the marketing mix, often when sales hit a wall. Maybe the brand was growing well and then plateaued. Or maybe the thinking is that if a brand becomes well-known, this can reduce the number of price promotions a company must run amidst a crowded landscape. 

The reality is that branding activities as a tactic is not useful for any eCommerce firm other than the top two or three players.

Challenger brands cannot spend enough to break through or worse, branding programs tend to favor leaders by reminding people of the product category. More bluntly, every dollar you spend on branding actually pushes people to your competitors losing potential sales for you. 

Instead, the right next move to expand sales lies in using a “Brand-to-Demand” approach to eCommerce. This creates an architecture for how a multi-channel program would be planned.  Using this Brand-to-Demand architecture, a campaign around a new product launch might include branding elements such as virtual product placement, (Ryff is a good example) and the demand elements might include a partner promotion. (See next item, #2, to understand how to architect a brand to demand marketing mix.)

Branding as a separate program is more a “hail mary” than productive for most eCommerce brands.  

2 > Funnel management  

The traditional sales funnel; upper funnel, mid and lower funnel, does not work exactly the same way anymore for everyone. Yet too many companies insist on living (and maybe dying) by this construct. Instead, given the diversity of eCommerce, there are actually three different varieties of sales funnels to manage. Understanding which funnel fits your business, B2B or B2C, is key to maximizing returns.

Funnel model #1: Upper to lower funnel:

This is the most typical funnel and it is best applied to businesses with a consideration/ close cycle that runs for weeks or months. This funnel is driven by content that can help prospects understand the value proposition. Typically, conversion is a considered buy and usually not an “impulsive” decision.  

What media channels drive this funnel (examples):

• Brand:
  • Traditional SEO
  • AI discovery SEO – very different than traditional SEO
  • Content – blogs, tools, white papers
  • Organic social
  • Sponsored content placements or product placements
• Demand:
  • Purchase level resource content, ie – calculators
  • Paid social
  • eMail
  • Chatbots (but ONLY well-developed ones – a bad chatbot is worse than NO chatbot)

Funnel Model #2: One funnel.

In this funnel model, all stages collapse into a “single” funnel that merges consideration and lower funnel audiences. This covers businesses with a very short consideration-to-close cycle – usually measured in days – not weeks or months. Brands that fall into this category typically include impulse buys or products that are largely commoditized.  

In this case, the marketing strategy is all about brand visibility during the short consideration cycle. Further, all marketing promotional efforts should center on product use cases, ie – problem/ solution. If, for example, a site sells premium auto products, marketing investments should highlight prospect use cases around common automobile issues like oil leaks or cooling system issues.  

What media channels drive this funnel (examples):

• Brand:
  • Sponsored newsletters from relevant publications
  • White paper
  • Organic social
• Demand:
  • PPC (as much as is affordable)Paid social
  • Digital media (direct buys – not programmatic)

Funnel Model #3: Topic funnel.

This funnel is the model for any conversion process, B2B or B2C, that is content centric. This funnel typically covers luxury brands, B2B or B2C for goods that have high user investment either financially or in a personal way such as customized hair care products. This funnel needs the most content that can carry a user through their topic journey to conversion. To succeed in this funnel requires a heavy investment in data (topic level) and content development programs.   

Most critically, this model is not dependent on user profiles or demographics nearly as much as having the right content to drive a user from consideration to a purchase.

What media channels drive this channel (examples):

• Brand:
  • Video
  • Sponsored newsletters
  • Onsite tools
  • Product placements
• Demand:
  • eMail
  • PPC
  • AI Discovery SEO
  • Trade journal advertising

Figuring out which funnel your business falls into help optimize all efforts in the most efficient way possible. That said, brand may shift from funnel to funnel if the product mix changes or expands.

3 > Understand the real role of creative to drive sales

When one thinks about eCommerce, one’s first thought is that creative is the main driver of sales. This is a common but incorrect conclusion. A successful eCommerce is the culmination of five core pillars:

1 > Product (s) and product strategy

2 > Site user experience including frictionless commerce  

3 > Business model including:

• Competitive environment and offerings

4 > Merchandising that addresses seasonality, market segments, partnerships, and new opportunities or line extensions

5 > Marketing including:

• Brand positioning in market
  • SEO discoverability
  • Communications

Notice that “communications” is just one part of one pillar of an eCommerce businesses.

This is a clue that while advertising is important, it might only account for 20% of a brand’s overall revenue model.

Yet, too too often, brands hyper focus on advertising because it is the most “visible” – literally. As result, the other pillars tend to be overlooked and languish in the shadows. The full recipe needed for eCommerce success, requires investment across all five pillars. This may include necessary tech solutions for analytics or new types of data. What is key to remember is that advertising’s contribution to overall success is notably less than assumed.

4 > Too much testing. Too little testing. How to get it right.  

The testing paradigm is firmly planted into the consciousness of every marketer; test more things, more often and faster.

This sounds right – more testing means more insights and hopefully better outcomes. Right?

The truth is more nuanced than that.  A lot of time is wasted testing things that either cannot be operationalized or duplicated.

In fact, testing too much is as dangerous as testing too little.

The right testing architecture means you match testing programs to the five pillars of a successful business (Item #3).  

1. Product (s) and product strategy
2. Site user experience including frictionless commerce
3. Business model including:
4. Merchandising that addresses seasonality, market segments, partnerships, and new opportunities in line extensions
5. Marketing

Testing can be structured to address one element in each pillar across all five pillars or ALL elements in a specific pillar. Once testing is done for an entire pillar, one can move into the next pillar. Either way, by bringing structure to testing provides a business with intelligence is a business that can thrive.

5 > AI is here and there but Intelligence is MIA

AI potential has gripped the imagination for every marketer reflected in the image below, the Gartner Hype Cycle.  

Everyone thinks everyone else is doing AI better in driving outcomes than they are. This is a distorted view.

AI utilization in eCommerce is a WIP (work-in-progress) and we are at the peak of “Inflated Expectations.” As most tech cycles, it will begin to crater as real world experiences bump up against the hype.

We see this play out in eCommerce too. All the major shopping platform have an AI component to provide insights yet too often the AI is painfully wrong. In some cases, it was obvious that AI glitched when it could not fetch sales data from last week explaining, “The data you asked for has not happened yet…”

Yet, too often and more troubling, the AI errors are less obvious and thus less detectible. Here is a real-world example with the leading shopping platform. The question was put to AI; “how many leads came through the landing page.” The AI answered Zero. This seemed odd since traffic patterns suggested the landing page had a lot of traffic so it seemed unlikely there were Zero conversions. To get the answer, the eCommerce site had do a special query to the database to get an accurate answer. You won’t be surprised to learn the answer was double digits. That’s a big deal because the brand might have thought there is no point to doing landing pages. Imagine ALL the missed opportunities because of AI misfires.  

The lesson to take away is to use AI correctly in eCommerce. AI is helpful in purely mechanical functions – like figuring out pricing models. However, if AI is used to answer questions about insights, verify its conclusions with data. Do NOT accept AI insights as accurate or even on point as AI accuracy performance is hit ‘n miss.

The real job is using AI is to learn when AI can be trusted and when it cannot be trusted. It means the difference between a successful quarter or a declining one.   

6 > The Case for Being a Data Skeptic

Data is the fuel of eCommerce but not all fuel is of equal value or octane that can go the distance.

There are three rules of the data skeptic:

• Understand that all data is not created equal. Similarly with people in the real world, some data can be trusted and some must be suspect.
• All data is not created ethically. Privacy is something we all need to take seriously but sadly too much data was not sourced ethically – with consent. This means before you bring data in – ask tough questions of the data service provider.
• All data has an error rate. The only variable is how high is the error rate. This suggests rigorous testing of data samples before any data is brought in wholesale into the company.  

This poses a real problem for marketers because generally speaking eCommerce firms don’t have data engineers on staff. This also explains why data providers are relentless in pushing the quality of their data even though, according Ad Age, “nearly 50% of data is wrong, (source: AdAge:  https://adage.com/article/measurement/nearly-half-data-used-ad-targeting-wrong-truthset-study/2521136/).

This statistic includes all the major players from Google to Meta despite all their promises to slice and dice their data endlessly.

This puts eCommerce brands in a real pickle because billions of dollars are spent based on data that is about 50% wrong. Astonishing.

No eCommerce can totally solve the data conundrum, but they can limit the damage bad data can do by creating data “sources of truth” for key functions.

Let’s take the example of attribution – especially tricky because EVERY platform creates its own attribution model that advantages their ad platform. Google uses last click whereas Meta uses all click and some shopping platforms use a first/ last click model. Some unsophisticated clients use all attribution data per each platform and then try to assimilate it, often leading to confused action items. By using each platform for attribution – this invites chaos and irreconcilable collisions because there is no source of truth.

There are two different strategies to address this issue.

1) Hire the right data company to create an attribution schema for a brand.

This is not something that is easy or cheap but it sets up a clear source of truth structure that is available to everyone. Expect this type of project to take about four months (for all data to get integrated and standardized) with a price of about $25K+ (annual). Yet this investment will be able to propel the company to make good decisions around performance. Funnel.io is a good option here with lots of integration choices to match the tech stack of a brand.

2) There a simpler approach which is to (gasp) let go of the need for attribution at a channel level but to measure campaigns in the aggregate – at a revenue level. This does not mean there won’t be optimization at a channel level, but are a few technical ways to unify channel performance data into a unified dashboard to reflect only revenue/ conversion data at a campaign level – not a channel level.

Source of truth always ends up being about sales generated. In the case of data complication and untrustworthiness, simplicity of measurements is the key – whether it is measuring campaigns or profitability – create a data architecture based on solid data that is highly trusted. Use all other data with extreme caution.

7 > Measurement and Attribution: Know the Difference About What They Can Teach You.

If data is the jet fuel of adtech, measurement and attribution together make up the engine. Each dimension is critical but too often these data insights are conflated thus overlooking nuanced contributions of each data category.

The Key Metrics In Analytics:

• Traffic trends
• Time on site
• Pages per visit
• Repeat versus new visitors

Ignore most demographic data because it is either irrelevant (as with B2B) or largely missing. It is common for Google Analytics to have demographic data on only about 10% of the entire audience.

The Key Metrics in Attribution:

• CLV at a campaign level
• Average transaction value at a campaign level
• Cost per Acquisition
  • Note – the CPA expenses only apply to new customers. Repeat customers do not have any acquisition expenses (that cost was already captured when they became customer). This is why repeat business is the lifeblood of almost every successful eCommerce business.

Of course, there are many other metrics one can look at, but one can get lost in the data trees and miss the forest. Stay focused on the important measures to understand what to do next and what makes money.

8 > Role of promotions

This is probably one of the trickiest aspects of eCommerce because no one wants to leave money on the table. Too many promo’s and audiences will be trained to wait for a sale. Too few promotions and competitors will eat your lunch.

To unravel this puzzle requires a more strategic look at the question based on the life stage of the eCommerce company.

As a company moves from startup to maturity, pricing protocols change to reflect the company’s life stage.

Here’s the key points.

The Startup:

In this stage, the company must experiment with price and promotions. The overall goal though is to capture as much revenue as possible to become stable enough to evolve. If a startup is too cautious about running promotions, then they risk the health of the company. At this stage, the price of promotions should be commensurate with the CPA (Cost per Acquisition) and Transaction Value of a sale. This is not the stage to fuss too much about channel optimization or any branding activities – this is the moment to do a land grab.

Challenger Brand:

At this stage, an eCommerce company has the best and the worst of eCommerce. At this stage, the company has a good sense of their top sellers, their CPA and average cart size. The key challenge now is to deploy tactics that maintain the growth momentum even though ad budgets might not increase.  

This is where clever “offers” play a central role. This is when programs like loyalty or membership programs are important. Instead of doing price promotions, one can offer extra loyalty points for certain classes of products. At this point, video content with multi-products focused on problem/ solution content will be powerful sales stimulators. eMail is a powerful channel to highlight new offers or seasonal packages discretely without having to promote it publicly.

In short, maintaining sales momentum is a function of product and service innovation – not just price off.

Market leaders:

This is where pricing promotions are usually based on new product launches and new offerings. For instance, let’s take the example a mature, highly respected consumer electronics brand. The product sell for $700 or more making it too expensive for a big segment of the addressable audience. To make matters worse, since the product is of such high quality, there is not a lot of sustainable repeat business revenue.

The answer for this brand is to launch a new customer acquisition option through a leasing program. This way, many more consumers can afford the product, which ensures recurring revenue, a ready pipeline for the resale side of the business AND a predictable repeat revenue model driven the by expiration of the lease.  

Promotions are clearly table stakes for any eCommerce venture. Used well, they can help evolve a company to be strong and profitable. Used badly, price promotions can cause a company to bleed cash slowly and painfully and over a longer, excruciating period of time. In eCommerce, promotions can be a company’s best friend or worst nightmare. Sometimes, price promotions are both.  Proceed with caution.

9 > It’s not your grandfather’s SEO anymore

SEO did not evolve for over 15 years. Then within the last 12 months everything changed. The SEO battleground has shifted completely so that fighting for higher organic ranking is being replaced with brands fighting for a place in AI results. This is a different battle indeed.

To succeed in the new age of AI, one is required to think like AI.

• Keywords are out. Search terms and topics are in.

• Blogs are out. Resource content (search terms based) is in.

• Site tags are mostly old news. Topic data is in.

From a practical point of view, here’s what it means for brands:

• Content needs to reflect multi-layered search topics – NOT keywords – that answer questions for consumers. Now authority content is how a brand achieves discoverability in this new AI terrain.
• PPC needs to be recast as providing resources – not just long tail ads.
• While there was always some opaque relationship between PPC and SEO despite Google’s attestations otherwise, now it’s wise to be ever more mindful of the direct connection between SEO and PPC. Plan content topic campaigns that span both channels.
• Backlinks become less useful for gaining domain authority UNLESS the backlinks are theme/ topic based and answer top; how to, how to, when to, why to type of consumer questions.

The SEO practice for the next 12 months will be quite different than the SEO practices of the last 15 years. It will require different, topic data that provides intelligence into what are search terms drive revenue, (ie –  Topic Intelligence addresses this new topic data paradigm).  

If you or your agency is doing SEO the old way, you could be less visible to your prospective customers than your competitors. More important, if your agency has not shifted to using topic data to produce resource content, you will be less visible to your prime audience. That is a hit no eCommerce brand can afford.  

AI search discoverability is the real discoverability challenge. Plan for it and plan on it.

10> Content is not king. Conversion topic data is.  

When it comes to content, there is a consensus that content is the “propellant” for eCommerce. Given the importance of content to drive revenue, it turns out there is a lot of guessing going on. A lot.

There’s a lot of guessing about what to write about.

There’s a lot of guessing about which topics are even worth writing about. There’s way too much guessing about the role of content in driving outcomes.

Sure, there is CMS data or analytics data about which pages work better than others but these data points do not provide the insights needed to understand which topics drive sales and optimize metrics.

Most data focus on profile data, demographic data, psychographic information, and interest classification data. The issue with all this data is that it can be wrong as we have seen but more troubling is that these data sources do not reflect the nuanced intent needed for conversion.  We know this to be true because how often has a retargeting ad been shown after a user already bought the product?

This means marketers are, again, relegated to guessing games. It is exhausting and stressful.

The answer is to understand which topics drive prospects forward and to understand the topic journey to conversion. Even more important is the need to understand the topics that are worthy of investment before any investments are made.

The new data solutions can create a custom language model for each brand’s business that defines the optimum content footprint a brand should develop to optimize sales. Data and analytics platforms like Topic Intelligence.ai provide both the topic data needed for conversion AND the analytics to understand the top topic journeys to conversion. This data drives AI SEO, email, social advertising and content development. Together, this game changing topic data and analytics changes the acquisition game because with the right data, acquisition becomes predictable and repeatable.

—————————————————————————————————————————-

eCommerce has always been a very difficult business model to crack because it is a delicate combination of merchandising, technology, advertising, and instincts.  AI is about to make this business both easier and harder. Easier in the sense there are more tools to facilitate more programs and harder because a brand can deploy more AI-driven programs more quickly that could overwhelm marketers’ ability to untangle all the data points that will be churned out.

For all the uncertainties in eCommerce, it is also one of the most exciting marketing businesses anywhere. When one cracks the code and revenue starts flowing – nothing beats that feeling. Nothing.