Your Competitive Advantage in E-commerce

AI in E-commerce: The Opportunity

Most professionals running or working in e-commerce businesses hold at least one belief about AI that is quietly costing them time, money, or competitive ground. Not because they're uninformed, they've read the headlines, maybe tried a chatbot, perhaps experimented with a product description tool. The problem is that the loudest narratives about AI in e-commerce tend to be either wildly overhyped or frustratingly vague. The result is a strange paralyzis: smart people either ignore AI entirely, waiting for it to 'mature,' or they bolt on a tool here and there without a coherent strategy. This lesson breaks down the three most common misconceptions, replaces them with a clearer picture, and shows you exactly what's actually working for e-commerce teams right now, no technical background required.

Myth 1: AI in E-commerce Is Mainly About Chatbots

Ask ten e-commerce managers what AI means for their business and at least seven will mention chatbots first. It makes sense, chatbots are visible, they're marketed aggressively, and they're the most consumer-facing application most people have encountered. But equating AI in e-commerce with chatbots is like equating electricity with lightbulbs. The lightbulb is real and useful. It is also about 2% of what electricity actually does. The same ratio roughly applies here. Chatbots handle a narrow slice of the AI opportunity in e-commerce, and for many business types, particularly small and mid-size stores, they're not even the highest-value starting point.

The real AI opportunity in e-commerce spans at least eight distinct workflow categories: product content creation, personalized merchandising, dynamic pricing, demand forecasting, ad copy and creative generation, customer review analyzis, email and SMS campaign writing, and search optimization. A marketing manager at a mid-size apparel brand, for example, might use ChatGPT Plus to write 50 product descriptions in the time it used to take to write five, then use Canva AI to generate lifestyle imagery for those products, then use Google Gemini to analyze which product attributes appear most frequently in five-star reviews. None of that involves a chatbot. All of it directly impacts revenue.

According to a 2023 McKinsey report on generative AI, retail and e-commerce rank among the top three industries for AI-driven value potential, with the majority of that value coming from marketing and sales productivity, not customer service automation. The report estimated that generative AI could add between $240 billion and $390 billion annually to the retail sector globally. Chatbots contribute to that number, but they are far from the dominant driver. The dominant driver is content and decision-making speed, getting the right product in front of the right customer with the right message, faster and more accurately than a purely manual process allows.

Click to enlarge

Myth 2: You Need a Big Budget and a Tech Team to Use AI

The second myth is the one that keeps small business owners and lean e-commerce teams on the sidelines. They've seen the enterprise case studies, massive retailers deploying custom AI recommendation engines, building proprietary data pipelines, hiring machine learning engineers. The implicit message seems to be: this is for companies with seven-figure technology budgets. That was partially true in 2019. It is not true in 2024. The shift happened because of large language models, tools like ChatGPT, Claude, and Gemini, which put genuinely powerful AI capabilities behind a simple chat interface that anyone can use on day one.

A solo e-commerce operator running a Shopify store can access the same underlying AI writing and analyzis capability as a Fortune 500 marketing team, for $20 per month via ChatGPT Plus or Claude Pro. They won't have the custom integrations or the proprietary datasets, but for the tasks that matter most, writing compelling product pages, generating ad variations, drafting email sequences, summarizing customer feedback, the gap between a solo operator and a large brand has never been smaller. Tools like Canva AI (built into Canva Pro at around $15/month) and Grammarly AI (included in Grammarly Business) further extend this without requiring any technical knowledge.

The critical insight is that most high-value AI tasks in e-commerce are language tasks, writing, editing, summarizing, analyzing text. Language tasks don't require custom models, engineering teams, or API access. They require knowing how to give a good instruction to a tool that already exists. That skill is called prompt craft, and in business terms it's simply the ability to brief an AI the way you'd brief a talented junior colleague: give context, specify the format you want, and describe the audience. A marketing manager who can write a clear creative brief can write an effective AI prompt. No coding required, ever.

Myth 3: AI Will Replace E-commerce Jobs

The displacement fear is real and understandable. Headlines about AI eliminating jobs have been consistent and alarming. But the evidence from e-commerce specifically tells a more nuanced story. The roles that AI genuinely threatens are narrow, repetitive, and often already partially automated: bulk data entry, basic templated email responses, simple image resizing. The roles that AI actually strengthens are the ones that require judgment, customer understanding, brand voice, and strategic thinking, which describes most of what e-commerce managers, merchandisers, marketers, and buyers actually do. A 2023 Stanford HAI report found that AI tools in marketing contexts consistently acted as amplifiers of human output, not replacements for it.

The more accurate model is this: AI handles the production grunt work so that professionals can spend more time on the decisions that actually move the business. A copywriter who used to spend 80% of their time drafting and 20% refining now inverts that ratio. AI drafts, the human refines, approves, and redirects. The output volume goes up, the quality goes up (because a human is reviewing everything), and the copywriter's strategic contribution goes up because they're no longer buried in first drafts. The same pattern holds for e-commerce buyers analyzing trend data, HR teams drafting job posts, and customer success managers reviewing support tickets. AI does more producing; humans do more deciding.

Click to enlarge

Myth vs. Reality: A Clear Comparison

What Actually Works: The Real AI Opportunity in E-commerce

The e-commerce professionals seeing the clearest results from AI right now are not the ones who've deployed the most sophisticated technology. They're the ones who've identified the three or four tasks in their workflow that are high-frequency, time-consuming, and language-based, and systematically handed those tasks to AI tools. Product description writing is the most common starting point, and for good reason: a typical e-commerce catalog update might require 200 new descriptions, each needing to reflect the brand voice, include relevant search terms, and convert browsers into buyers. Done manually, that's weeks of work. Done with ChatGPT Plus using well-crafted prompts and a clear brand brief, it's days, and the output is often stronger because the AI doesn't get fatigued or inconsistent on description 147.

Email marketing is the second high-impact area. E-commerce email sequences, welcome series, abandoned cart recovery, post-purchase follow-up, win-back campaigns, are proven revenue drivers, but they require consistent copywriting output that most lean teams struggle to maintain. AI tools like ChatGPT Plus or Claude Pro can generate complete email sequences in minutes, including subject line variations for A/B testing, personalization tokens, and segmented versions for different customer types. A small team that previously sent one campaign per week because that's all they could write can send three or four, dramatically expanding their touchpoint frequency without adding headcount.

The third area, and the one that surprises most people, is customer intelligence. E-commerce businesses sit on enormous amounts of unstructured text data: product reviews, support tickets, returns feedback, social comments. Most of this data is never systematically analyzed because doing so manually would take weeks. AI tools can process this in minutes. Paste 100 one-star reviews into Claude Pro and ask it to identify the top five recurring complaints, grouped by theme, with direct quotes from customers. You'll have a product development brief, a customer service training document, and a marketing messaging correction, all from data you already owned but couldn't use. That's not automation replacing judgment. That's AI making your judgment faster and better-informed.

Frequently Asked Questions

• Q: Do I need to share my customer data with AI tools to use them effectively? A: No. The highest-value tasks, writing product descriptions, drafting emails, generating ad copy, require no customer data at all. You simply describe your audience in the prompt. For review analyzis, you can paste anonymized text without any identifying information.
• Q: Which AI tool is best for e-commerce: ChatGPT, Claude, or Gemini? A: All three are capable. ChatGPT Plus ($20/month) has the largest user base and the most e-commerce-specific prompt examples available online. Claude Pro ($20/month) tends to produce longer, more nuanced text and is particularly strong for brand voice consistency. Google Gemini is useful if your team runs on Google Workspace. Most professionals eventually use two of the three for different tasks.
• Q: Will AI-generated product descriptions hurt my SEO? A: Not inherently. Google's guidance is that it evaluates content quality, not content origin. AI-generated descriptions that are accurate, specific, and genuinely useful to shoppers perform well. The risk is generic, templated output, which is a prompt quality problem, not an AI problem. Specific prompts produce specific, SEO-friendly descriptions.
• Q: How much time does it realiztically take to get useful output from AI? A: For a first attempt at a new task type, expect 20–30 minutes including prompt refinement. Once you have a prompt that works well for a given task, a product description formula, an email template structure, reusing and adapting it takes 5–10 minutes. Most professionals report saving 5–10 hours per week within their first month of consistent use.
• Q: What if the AI gets facts wrong about my products? A: This is the most important quality control point in e-commerce AI use. Always provide the AI with the factual specifications in your prompt, dimensions, materials, certifications, pricing, shipping times. Treat AI output the way you'd treat output from a talented new hire: review it before it goes live. Never publish AI-generated content about product specifications without a human check.
• Q: Can I use AI tools if I'm on a Shopify, WooCommerce, or BigCommerce platform? A: Yes, and increasingly these platforms have AI features built in. Shopify Magic, for example, generates product descriptions directly inside the Shopify admin panel. But even without native integrations, you can write content in ChatGPT or Claude and paste it directly into any platform. No technical connection between the tools is required.

Key Takeaways from Part 1

1. AI in e-commerce covers at least eight distinct workflow categories, chatbots are one of them, and for most small to mid-size teams, not the highest-ROI starting point.
2. Enterprise-grade AI capability is now accessible to any e-commerce professional for $15–$20 per month through tools like ChatGPT Plus, Claude Pro, and Canva AI, no technical background required.
3. The displacement fear is largely misdirected: AI amplifies the output of e-commerce professionals who use it and poses the greatest risk to those who don't.
4. The three highest-impact AI applications for most e-commerce teams right now are product content creation, email marketing, and customer review analyzis.
5. Effective AI use in e-commerce is primarily a prompting skill, the ability to give clear, specific, context-rich instructions. This is a business communication skill, not a technical one.

Three Myths That Are Costing E-commerce Teams Real Money

Most professionals in e-commerce believe that AI tools are either too complex to use without a tech team, only useful for giant retailers with massive budgets, or that they replace human judgment entirely. These beliefs are understandable, they come from years of tech industry hype and watching Amazon announce warehouse robots. But they're wrong in ways that matter. Acting on these myths means leaving real efficiency gains on the table, or worse, avoiding tools that your competitors are quietly using right now to cut costs, speed up decisions, and sell more. Each myth deserves a direct takedown.

Myth 1: 'AI in E-commerce Is Only for Technical Teams'

The most common reason non-technical professionals avoid AI tools is the assumption that using them requires coding, data science, or at minimum a dedicated IT department. This assumption made sense five years ago. It doesn't hold up today. Tools like Shopify Magic, Klaviyo AI, and Canva AI are built directly into platforms that e-commerce managers already use daily. There is no setup, no script to write, no engineer to call. A marketing manager can generate a product description, a promotional email, and an ad image variation in under fifteen minutes, entirely through point-and-click interfaces. The barrier to entry is lower than learning a new Excel formula.

Take a small outdoor gear retailer as a concrete example. Their marketing team of two people was spending roughly twelve hours per week writing product descriptions for new seasonal inventory. After turning on Shopify Magic's AI description generator, that same work now takes under three hours. The team didn't hire a developer or change their tech stack. They clicked a button that was already in their admin panel. The AI draft still gets reviewed and lightly edited by a human, but the blank-page problem is gone. That's the actual experience of AI adoption for most e-commerce professionals: not a transformation, but a quiet removal of tedious, repeatable work.

The mental model worth adopting here is this: think of AI tools as capable interns who are available twenty-four hours a day, never get tired, and can produce a first draft of almost anything in seconds. They still need direction from you. They still need a human to check their work. But they remove the friction that slows teams down. You bring the strategy, the brand knowledge, and the final judgment. The AI handles the volume. That's a partnership, not a handover, and it requires no technical skills whatsoever to manage.

Myth 2: 'AI Gives You a Significant Edge Only If You Have Huge Data'

The second myth is that AI's value in e-commerce scales directly with the size of your customer database. The logic goes: Amazon has millions of transactions to train on, so their recommendation engine is powerful; your store has a few thousand orders, so AI won't do much for you. This logic confuses two very different categories of AI. Training a custom machine learning model from scratch does require large datasets. But using a pre-trained AI tool, which is what every mainstream e-commerce platform now offers, doesn't require your data at all. The model was already trained on billions of examples before you logged in.

When you use ChatGPT Plus to write a promotional email for your Black Friday sale, the AI isn't learning from your customer data. It's applying patterns learned from millions of marketing examples to follow your specific instructions. When Klaviyo's AI suggests the best send time for your next campaign, it's drawing on aggregate behavioral data from the entire Klaviyo network, hundreds of thousands of brands, not just your list. A store with five hundred email subscribers benefits from the same underlying intelligence as a store with five million. The playing field is genuinely more level than the myth suggests.

For small and mid-sized e-commerce businesses, this is arguably the most important reframe in this entire lesson. Tools that previously required enterprise budgets and data science teams are now available as monthly subscriptions starting at twenty dollars. A boutique skincare brand with a lean team can now produce personalized email sequences, dynamic product copy, and customer service chat responses at a quality level that would have required a full agency retainer two years ago. The advantage isn't data volume, it's knowing which tools to use and how to give them clear instructions.

Myth 3: 'AI Will Make Human Judgment Obsolete in E-commerce'

The third myth runs in the opposite direction from the first two. Instead of fearing that AI is too complicated or too limited, some professionals fear it will simply replace their role. This anxiety is understandable, but it misreads what current AI tools actually do well. AI is exceptionally good at producing first drafts, spotting patterns in structured data, and handling high-volume repetitive tasks like tagging products or answering common customer questions. It is genuinely poor at understanding brand nuance, making judgment calls in ambiguous situations, managing supplier relationships, or knowing when a customer complaint signals a deeper product problem. Those gaps are where experienced professionals earn their value.

A useful way to think about this: AI handles the throughput, humans handle the judgment. An AI tool can generate fifty product descriptions overnight. A merchandising manager still needs to decide which products to feature, how to position the brand against competitors, and whether a particular description strikes the right tone for a culturally sensitive occasion. The AI produces options at speed; the human makes the call. Companies that understand this dynamic, and build workflows that reflect it, outperform both those who ignore AI entirely and those who deploy it without oversight. The professionals who thrive aren't the ones who resist AI or the ones who blindly trust it. They're the ones who know exactly where to apply it.

Myth vs. Reality: A Direct Comparison

What Actually Works: Building a realiztic AI Workflow

The e-commerce teams getting the most out of AI right now share a specific approach. They don't try to automate everything at once. They start by identifying their highest-volume, lowest-complexity tasks, the work that takes time but doesn't require deep judgment, and they apply AI there first. Product description writing, customer service response templates, promotional email drafts, social media captions, and FAQ pages are the most common starting points. These tasks have a clear input (product details, brand guidelines, customer question) and a clear output (finished copy), which makes them ideal for AI assistance. The learning curve is minimal and the time savings are immediate.

The second characteristic of effective AI adoption in e-commerce is that teams build what practitioners call a 'prompt library', a saved collection of instructions that reliably produce good outputs for recurring tasks. Rather than starting from scratch every time a new product launches, a merchandising team maintains a master prompt template for product descriptions that already contains the brand voice guidelines, the required format, and the SEO requirements. Anyone on the team can use it. The output is consistent. This turns AI from a personal productivity trick into a genuine team system. Tools like Notion AI, ChatGPT's custom instructions feature, or even a simple shared Google Doc can house this library.

The third characteristic is a defined review step. Every AI output in a high-performing e-commerce workflow passes through a human check before it goes live. This isn't about distrust, it's about quality control and brand protection. AI tools occasionally hallucinate facts, misread tone, or produce something technically correct but off-brand. A thirty-second review catches these issues before they reach customers. The review step also keeps the human in the loop, which means the team builds intuition about when AI output is strong and when it needs more guidance. Over time, their prompts improve, their review time decreases, and the overall system gets faster without sacrificing quality.

Click to enlarge

Put It Into Practice

Frequently Asked Questions

• Q: Will customers know our product descriptions were written with AI? A: Not if you review and edit them before publishing. AI-assisted copy that has been checked and lightly refined by a human is indistinguishable from fully human-written content. The review step is non-negotiable, but it doesn't need to take long. Most experienced users spend two to four minutes reviewing a standard product description.
• Q: What if the AI writes something factually wrong about one of my products? A: This happens occasionally and is the main reason a human review step is essential. AI tools generate text based on patterns, not verified facts, they don't know your specific product specifications unless you tell them. Always include accurate product details in your prompt, and always check any specific claims (materials, dimensions, certifications) before publishing.
• Q: Is it safe to put customer data or proprietary business information into ChatGPT or Claude? A: Exercise caution here. For general writing tasks like product descriptions or email drafts, you don't need to include sensitive customer data at all. If you're using an enterprise version of these tools (ChatGPT Enterprise, Claude for Enterprise), your data is not used for training. For free or standard subscriptions, avoid entering personally identifiable customer information or confidential pricing strategies.
• Q: How much do these AI tools cost for a small e-commerce business? A: Most entry-level AI tools are either free (with limits) or cost between ten and thirty dollars per month per user. ChatGPT Plus is twenty dollars per month. Claude Pro is twenty dollars per month. Shopify Magic is included with any Shopify subscription. Klaviyo AI features are included in existing Klaviyo plans. For most small businesses, the cost is recovered in the first week of time savings.
• Q: Do I need to tell customers that AI helped create my product descriptions or marketing emails? A: In most markets, there is currently no legal requirement to disclose AI assistance in marketing copy, provided the content is accurate and not deceptive. This is an evolving area, and some platforms (like Amazon) have specific policies about AI-generated images. Always check the policies of the specific platforms where you sell and advertise.
• Q: What's the biggest mistake e-commerce teams make when starting with AI? A: Trying to do too much at once. Teams that attempt to automate their entire content operation in the first month end up with inconsistent output and no clear process. The teams that succeed start with one specific, high-volume task, build a reliable prompt for it, prove the time savings, and then expand from there. Small wins build the confidence and the team habits that make broader adoption stick.

Key Takeaways from This Section

1. AI e-commerce tools are built into platforms non-technical professionals already use, no coding or IT support required to get started.
2. Pre-trained AI tools deliver value regardless of your store's size. A small business with five hundred customers uses the same underlying AI intelligence as an enterprise retailer.
3. AI handles volume and repetition; human professionals handle brand judgment, strategy, and quality control. These roles are complementary, not competitive.
4. The quality of AI output depends directly on the quality of your instructions. A well-constructed prompt, reviewed by a human, produces publish-ready content.
5. The most effective AI workflows start narrow, one high-volume task, one reliable prompt, measurable time savings, before expanding to other use cases.
6. A shared prompt library turns AI from a personal tool into a team system, ensuring consistent output and making onboarding new team members faster.

What Most E-commerce Professionals Get Wrong About AI

Most professionals in e-commerce hold three beliefs about AI that are either flat-out wrong or dangerously incomplete. First: AI is only for large retailers with big tech budgets. Second: AI will automate away the human judgment that makes great merchandising and marketing work. Third: you need to overhaul your entire tech stack before AI can help you at all. Each of these beliefs causes real damage, they either freeze teams in place or lead to expensive, misdirected investments. The evidence tells a different story, and once you see it clearly, the path forward becomes obvious.

Myth 1: AI Is Only for Amazon and the Big Players

This belief made sense five years ago. Enterprise AI platforms cost hundreds of thousands of dollars annually and required dedicated data science teams. That era is over. Tools like ChatGPT Plus ($20/month), Canva AI (free tier available), and Klaviyo's built-in AI features (included in standard plans) have put meaningful AI capability within reach of a solo Shopify seller running a $200K/year business. The barrier shifted from budget to knowledge, which is a very different problem.

The mental model shift here is critical: stop thinking of AI as infrastructure (something you build) and start thinking of it as a skilled contractor (someone you hire by the task). You don't need to own a construction company to renovate your kitchen. You hire the right specializt for the job. AI tools work the same way, you bring them in for specific tasks like writing product descriptions, drafting ad copy, or summarizing customer reviews, and you pay almost nothing per use.

Click to enlarge

Myth 2: AI Replaces Human Judgment in Merchandising and Marketing

This fear is understandable. When you see AI generate a full product description in four seconds, it's natural to wonder what the copywriter is for. But the professionals thriving with AI aren't replacing their judgment, they're offloading the mechanical parts of their work so their judgment gets applied to higher-stakes decisions. AI drafts; humans decide. AI suggests; humans approve. That division of labor is where the real productivity gain lives.

Consider what AI genuinely cannot do well: it doesn't know your brand's quirky voice that took three years to develop. It doesn't know that your core customers are 45-year-old women who hate corporate jargon. It doesn't know that your best-selling product has a backstory that resonates emotionally with a specific community. You know all of that. AI is a fast, tireless drafting assistant, not a strategist. The strategist is still you.

A Stanford HAI report from 2024 on AI in creative work found that the highest performance gains came from humans using AI as a first-draft generator, then applying domain expertise to refine outputs, not from teams that tried to run AI outputs directly without review. In e-commerce terms: let AI write the first version of your product description, then spend two minutes applying your brand knowledge. The result beats both a pure-AI output and a from-scratch human draft for speed and quality combined.

Myth 3: You Need a New Tech Stack Before AI Can Help You

This is the most paralyzing myth of the three. Professionals hear 'AI integration' and picture months of IT projects, API connections, and system migrations. For some advanced use cases, that's true. But the majority of high-value AI tasks in e-commerce require nothing more than a browser tab. You can paste your customer reviews into ChatGPT and ask for a sentiment summary. You can upload a product photo to Canva AI and generate lifestyle variants. You can paste last month's email metrics into Claude and ask what the numbers suggest. No integrations required.

The trap is perfectionism, waiting for a seamless, fully integrated AI workflow before doing anything. Meanwhile, the practical wins (faster copy, smarter customer insights, better ad creative) are available right now through tools you can open in the next five minutes. Start messy. Get results. Build sophistication over time as you understand what's actually worth connecting.

What Actually Works: The Practical AI Playbook

The e-commerce professionals getting real results from AI share a common approach: they pick one specific, repetitive task that costs them time every week, apply an AI tool to it, measure the output quality, and then move to the next task. They don't try to 'implement AI across the business.' They solve one concrete problem at a time. Product description writing is the most common starting point because the time savings are immediate and the quality bar is easy to evaluate.

Customer review analyzis is the second high-value entry point. Most e-commerce managers have hundreds or thousands of reviews they've never systematically analyzed. Pasting 50 reviews into Claude or ChatGPT and asking 'What are the top five complaints and the top five compliments?' takes three minutes and surfaces insights that would take a human analyzt half a day. Those insights directly improve product pages, FAQs, and ad messaging, because you're using the exact language your customers use.

The third area that consistently delivers results is email subject line and campaign copy testing. Tools like Klaviyo's AI subject line assistant or simply running five variations through ChatGPT and asking which is most likely to drive opens, then A/B testing the top two, produces measurable lift without requiring any technical expertise. McKinsey research consistently shows that personalization and relevance drive e-commerce revenue, and AI makes both significantly more accessible to teams without dedicated data analyzts.

Frequently Asked Questions

• Q: Will AI-generated product descriptions hurt my SEO? A: Not if you edit them. Search engines penalize thin, duplicate content, not AI-assisted content. A well-edited AI draft that includes your target keywords and genuine product details performs well. The risk is publishing raw, unedited AI output at scale without reviewing it.
• Q: How do I make sure AI doesn't make up product details? A: Always provide the AI with your actual specs, materials, and features in the prompt. Treat AI as an editor and writer, not a researcher. If you give it accurate inputs, it works with accurate information.
• Q: Is the free version of ChatGPT good enough for e-commerce tasks? A: For most copy and analyzis tasks, yes. ChatGPT free handles product descriptions, email drafts, and review summaries well. ChatGPT Plus ($20/month) adds faster responses, image analyzis, and access to newer models, which helps with visual product work.
• Q: How long does it take to see real results? A: Most professionals see time savings within the first week. Conversion rate improvements from better copy typically show up within 30-60 days, depending on your traffic volume.
• Q: What if my team is worried AI will replace their jobs? A: Reframe the conversation. AI handles the mechanical, repetitive parts of creative work. Your team's expertise in brand, customer relationships, and strategy becomes more valuable, not less, when they're not spending four hours writing product descriptions.
• Q: Can I use AI to analyze my competitors? A: Yes. Paste a competitor's product description or ad copy into ChatGPT and ask: 'What positioning strategy is this brand using, and what customer pain points are they addressing?' It's a fast, free competitive intelligence method.

Key Takeaways

• AI tools are accessible and affordable for e-commerce businesses of any size, the barrier is knowledge, not budget.
• AI accelerates execution; your brand expertise, customer knowledge, and strategic judgment remain essential and irreplaceable.
• You don't need new software or technical setup to start, most high-value AI tasks work with copy-paste in a free browser tool.
• The highest-impact starting points are product description writing, customer review analyzis, and email copy generation.
• Start with one specific, repetitive task. Measure the result. Then expand. Trying to 'implement AI everywhere' at once produces nothing.
• Editing AI output is not optional, it's where your expertise turns a decent draft into an on-brand, accurate, conversion-driving asset.