The Internet Marketing Agency | Design. Development. Marketing | Get a better website and marketing strategy today. Start seeing results tomorrow.
Los Angeles, CA
Joined January 2009
- Tweets4.5K
- Following99
- Followers5.2K
- Likes346
Pinned Tweet
The next new customer who calls your business may not be a person.
Personal AI assistants from Meta, OpenAI, xAI, and a startup called Instinct find businesses, fill out forms, and book for the people who use them. Since September 2026, two of them make phone calls too. Your customer gets back one sentence about you.
We just published How to Market Your Small Business When Personal AI Assistants Do the Shopping.
Inside:
🤖 The four assistants that matter right now - what each one does, where your customer uses it, and what it costs them
🔎 The three places you show up - the facts it reads, the finish it needs, and the message it hands back
🖱️ The one-page test - count the clicks to your most common answer; more than two, fix the page
✅ Six things that always worked and work better now - each with one thing to do next
📋 12 copy-paste prompts - six that show you how an assistant decides, and six errands that send one to book your own business, scored out of 24
None of it requires a redesign. It's the same work that has always won customers, and the assistant stops forgiving the gaps.
Read the guide: imforza.link/ai-assistants-g…
If your customer's assistant described your business in one sentence, what would it say?
ALT Blog promotion graphic: Marketing for AI Assistants. Get picked when your customer's AI assistant does the shopping.
Website sessions down, bookings flat. That report reads like bad news, and it isn't.
When an AI assistant does your customer's browsing, your site gets fewer visits and more work.
Judge it by completed bookings, not sessions.
imforza.link/aia-teaser-x
ALT Flat illustration of a balance scale. A pile of gray website cursors rides high on the left, with a few tumbling off the edge, while one yellow booking card with a confirmed calendar slot sits lower on the right because it weighs more. An AI assistant's chat bubble hands the booking down along a dotted arc.
❝ imQUOTES: "Your first impression is now a summary written by software."
When your customer's assistant books you, it sends them a message. Something like: "Booked you Thursday at 2 with Dr. Patel, $95 new-patient exam, they texted a confirmation."
You don't write that sentence. You do control two things:
📋 The facts it pulls from. Hours, prices, and services that match everywhere you appear, so the summary gets them right
📲 What lands next to it. A confirmation text within a minute of the booking, from you, with your name, the time, the address, and one useful line: "Park in the back lot, it's free."
That second message proves you're real and responsive, and it puts your voice in your customer's hand right beside the assistant's. A business that stays quiet after a booking exists only in the summary.
Want to see the summary an assistant would write about you? Paste this into the assistant you use:
"Pretend you just finished booking [your most common service] at [your business name] in [city] for me. Write the confirmation message you'd send me, using only facts about the business you can verify online right now."
A wrong price, missing hours, or a vague address in the answer is the fact to fix first.
Read the guide 👉 imforza.link/ai-assistants-g…
Share this with the person who sends your booking confirmations.
ALT imFORZA #imQUOTES graphic: your first impression is now a summary written by software.
Your next customer sends their AI first.
Google's AI answers now spotlight official pages.
Gemini 4 Argon is now open in AI Mode.
Muse Wonder remembers what people like.
AI makes the shortlist. It picks what it can check.
5 minutes with imNEWS: imforza.link/subscribe
📊 imSTATS: 74% of consumers only care about reviews written in the last three months.
That's from BrightLocal's 2026 Local Consumer Review Survey of 1,002 US consumers. The same survey found 97% read reviews for local businesses, and 47% won't use a business with fewer than 20 reviews. BrightLocal sells review management software, so it has a reason to want you thinking about reviews. Read the exact figures with that in mind.
The figure in that survey that ties it to AI: ChatGPT and similar tools became the third most used source for local business recommendations, at 45%, up from 6% the year before.
An assistant reads your reviews and sums them up in one sentence for your customer. Volume, rating, and recency all land in that sentence. A six-month gap reads like a dark storefront.
Keep the stream moving:
📲 Ask by text, with a direct link. Start with every customer from the past 30 days
🔁 Make the ask automatic. It goes out after every completed job, so nobody has to remember
💬 Reply to your last ten reviews in your own voice. Including the bad ones
Read the guide 👉 imforza.link/ai-assistants-g…
When did your business get its most recent review?
ALT imFORZA #imSTATS graphic: 74 percent only care about reviews from the last three months.
📰 #imNEWS: @Google is updating AI Mode and AI Overviews to make source links easier to find and explore. The changes include more inline links higher in responses, prominent official-site links, expanded news carousels, richer related-article links, and clickable creator profiles.
🔑 3 Key Takeaways
① Links move closer to the answer
Google says inline links will appear more often and higher in AI responses. That gives publishers a better shot at earning the click while users are actively evaluating the answer.
② Official sources get a dedicated spotlight
Google will test a section at the top of some AI responses that surfaces one or two official pages with context about why each is useful. Clear authority and task-focused pages may matter even more.
③ Discovery becomes more visual
Timely-article carousels are expanding globally, while related links gain thumbnails and creator names become clickable. Search visibility now depends on more than a blue-link ranking.
✅ Action Item
Audit the pages you most want cited by AI Search. Make each page clearly authoritative, answer a specific intent, use descriptive titles and supporting visuals, and give Google an obvious reason to surface it as the official next click.
Source: imforza.link/Msfj3rF
ALT Google Search graphic illustrating new ways to explore web sources through AI Mode and AI Overviews
📰 #imNEWS: @Google introduced Universal Cart, an intelligent shopping cart that follows shoppers across Search and Gemini, with YouTube and Gmail planned later. It keeps products from different retailers together, tracks deals and restocks, and lets shoppers check out through participating merchants without Google becoming the seller.
🔑 3 Key Takeaways
① Shopping becomes a cross-channel journey
Shoppers can add products while using different Google services instead of rebuilding a cart on every site. The cart keeps items, price history, deals, and restock information in one place.
② AI agents move closer to completing purchases
Google is expanding agentic shopping so AI can do more than recommend products. Universal Commerce Protocol enables participating retailers to support checkout through Google Pay while remaining the merchant of record.
③ Merchant readiness now matters beyond the website
Product data, pricing, inventory, and checkout compatibility increasingly determine whether a business can participate in AI-assisted commerce. Weak feeds and fragmented systems can make products harder for shopping agents to act on.
✅ Action Item
Audit your Merchant Center feed, inventory accuracy, pricing, and checkout flow. Make sure your product data is complete and current, then watch for Universal Cart and UCP eligibility so your store is ready when agent-led shopping expands.
Source: imforza.link/0mP2pmn
ALT Google Universal Cart and agentic shopping announcement graphic
📰 #imNEWS: @xai has launched Grok Voice 3.0, a speech-to-speech model built for fast, natural customer conversations. It ranks first on Artificial Analysis’s Speech-to-Speech Index, responds in under one second, and is already handling millions of daily interactions across Tesla and SpaceX.
🔑 3 Key Takeaways
① Top-ranked voice performance
Grok Voice 3.0 scored 86.9 on the Speech-to-Speech Index and 97% on Full Duplex Bench. That means stronger reasoning plus more natural pauses, turn-taking, and interruptions.
② Real-world automation at scale
Starlink uses Grok Voice for 20,000 calls per day across multiple languages. About 70% are resolved without a human, while one in four sales calls ends in a purchase.
③ Frontier performance without a price increase
The API costs $0.08 per minute, the same as Grok Voice 2.0. Existing users of grok-voice-latest have already been upgraded automatically.
✅ Action Item
Pick one repetitive, high-volume phone workflow and test a voice agent against real calls. Measure resolution rate, handoff rate, latency, customer satisfaction, and cost per completed task before expanding it.
Source: imforza.link/rXhlp6g
ALT Grok Voice 3.0 announcement graphic from xAI
💡 imTIPS: Count the clicks between your homepage and a complete answer to the question you get most.
When a customer sends an assistant to find "a dentist who can see me Thursday," it compares businesses the way a hurried person would, minus the patience. When Skift, the travel trade publication, tested Meta's Muse on a New York hotel search, it compared properties on Expedia and Hotels.com and picked one without ever opening a hotel's own website. The booking sites had every answer on one page.
Run the test in ten minutes:
1️⃣ Pick your most common request. A new-patient exam, a kitchen estimate, a table for six on Friday
2️⃣ Start on your homepage and count clicks until you've found all four: the price with fees, when you're available, what's included, and how to book
3️⃣ Write down where each answer lives. If the answer is "on the phone," software can't find it
4️⃣ Two clicks or fewer passes. More than two, and the page needs the fix, not the assistant
Can't publish an exact price? Publish a range and what changes it: "Most kitchen remodels we do run $45,000 to $90,000; we'll give you a fixed quote after a one-hour visit." An assistant can use that sentence. "Contact us for a quote" is a dead end.
The full guide 👉 imforza.link/ai-assistants-g…
Follow us for one actionable marketing tip every Wednesday.
ALT imFORZA #imTIPS graphic: put your top answer within two clicks.
📰 #imNEWS: @Meta is giving Facebook users a direct way to shape what appears in Feed and Reels. New AI-powered controls let people add preferred topics and explain, in their own words, what they want to see more or less of, with recommendations updating in real time.
🔑 3 Key Takeaways
① Users Can Talk to the Algorithm
A new “Your algorithm” area lets people add and manage topics directly from a post’s three-dot menu. Facebook is moving beyond passive behavioral signals toward explicit user instructions.
② Feedback Is Becoming Conversational
After selecting “Interested” or “Not interested,” users may be asked why. Their freeform responses can refine recommendations during the same session.
③ LLMs Are Changing Content Ranking
Facebook is combining natural-language feedback with existing engagement signals such as likes, comments, and watch behavior. That gives its ranking systems more context about user intent, not just actions.
✅ Action Item
Review your Facebook content strategy for clear topical signals. Publish focused posts that make your expertise obvious, then monitor whether the new controls change reach, engagement, and the audience segments discovering your content.
Source: imforza.link/kOm50ci
ALT Facebook interface showing new AI-powered controls for shaping Feed and Reels recommendations
📰 #imNEWS: @Google has launched Gemini 4 Argon, its latest frontier model for complex work across coding, finance, legal, cybersecurity, and science. Available globally through the Gemini API, AI Studio, Antigravity, the Gemini app, and AI Mode in Search, Argon ranks No. 1 on the Vals Index and launches at $2 per million input tokens and $10 per million output tokens.
🔑 3 Key Takeaways
① Built for deep, multi-step work
Argon is designed for complex professional workflows, not just quick answers. Google says it combines stronger reasoning, multimodal analysis, writing, and instruction-following across high-stakes domains.
② The productivity gains are already measurable
Inside Google, Argon increased agentic code submissions by more than 35% while cutting rollbacks by nearly one-third. It also reduced long-session prep time in Google's Agent Development Kit by 40% and sped up cold imports by 21%.
③ Access is broad and pricing is aggressive
Developers and enterprises can use Argon through the API, while paid subscribers can access it in Google's consumer products. The introductory API pricing makes it practical to test against production-grade workflows now.
✅ Action Item
Pick one costly workflow that requires several steps, deep context, or large-document analysis. Run a controlled Argon pilot against your current model and compare output quality, completion time, human corrections, and total token cost before changing your stack.
Source: imforza.link/EeGS96P
ALT Google Gemini 4 Argon launch graphic
🛠️ imTOOLS: Hermes is a free, open-source AI assistant from Nous Research that you can send to shop your own business every month, then hear back from by text.
Why this one: your customers are handing errands to personal assistants like Meta's Muse, which search, compare, and book for them. The fastest way to see what those assistants find is to send one of your own to shop you. We run one of our own internal assistants on Hermes, and it's the one we'd point you to for this job.
How it helps:
🔎 It shops you like a customer would. Ask it to find a [your service] in [your town] without naming you, then to confirm your hours, prices, and reviews. It searches the web and reads the pages itself
⏰ It repeats the job on its own. Tell it "do this on the first of every month" in plain words, and it runs without you opening anything
💬 It reports back where you already are. Results land in WhatsApp, Telegram, Slack, email, or text, among 20-plus places it can message you
🧠 It remembers the last run. Hermes keeps notes between conversations, so ask it what got better or worse since last month
Why the AI chat app you already pay for may not cover it: Hermes is yours. It's free software you install, it can run on the AI plan you already have, and its memory and schedule stay on your machine instead of inside one company's app.
The real limits:
- It's a setup, not an app you open once. The desktop installer is quick, but connecting an AI model and then WhatsApp or Telegram takes a few guided steps, and some of them happen in a terminal window.
- Scheduled jobs only run while Hermes is running. If your computer is asleep, the job waits. Always-on means paid cloud hosting through Nous Portal.
- It sees what its own AI model sees, not what Muse sees. Treat the results as a stand-in for your customer's assistant, not a copy of it.
- It can act, not just read. Tell it to stop before it submits a booking or calls anyone, and connect only the accounts the job needs.
Pricing: Hermes is free under the MIT license. You pay for the AI model it runs on. Nous Portal offers free models plus Plus, Super, and Ultra plans at $20, $100, and $200 a month, which include $22, $110, and $220 in monthly credits. You can also connect an eligible ChatGPT plan you already pay for, with tools and cloud hosting billed separately.
👉 imforza.link/hermes
ALT imTOOLS graphic: The Best Completely Customizable AI Agent, showing the Hermes Agent home page on a laptop.
📰 #imNEWS: @SpaceXSI says X is unifying Grok, Cursor, and X Premium under one subscription. The new lineup ranges from Free to Ultra at $200/month, combining AI chat, media generation, coding agents, API access, and X creator features. Existing subscribers can keep their plans or switch as access rolls out.
🔑 3 Key Takeaways
① One bill for a broader AI stack
The subscription packages Grok’s AI tools, Cursor’s coding agents, and X Premium’s creator features. That could reduce the friction of managing separate plans for teams already using all three.
② Pricing now spans five tiers
Plans include Free, Plus at $8, Premium at $30, Super at $100, and Ultra at $200. The wider range gives individual users and power users clearer upgrade paths.
③ Bundling raises the switching cost
Linking Grok, Cursor, and X accounts can make the ecosystem more convenient, but it also concentrates more of a company’s AI workflow with one provider.
✅ Action Item
Audit your current AI and social subscriptions before switching. Compare total cost, included usage, and which tools your team actually uses so the bundle saves money instead of adding another overlapping plan.
Source: imforza.link/jh40N0K
ALT Promotional graphic for The X Subscription combining Grok, Cursor, and X Premium.
📰 #imNEWS: @Meta introduced Muse Wonder and Muse Wonder Full, new frontier models built for personal and professional agents. They pair Active Memory, which recalls relevant context across conversations, with Active Focus, which keeps working through blockers and long-running tasks. Muse Wonder is available now through the Meta Model API and Muse Code.
🔑 3 Key Takeaways
① Memory becomes operational
Muse Wonder is designed to retain what matters about a user and retrieve it when a later task needs it. That moves agent memory beyond chat history toward useful continuity across real work.
② Agents are being trained for sustained execution
Active Focus measures whether a model can try alternatives, adapt to new information, ask for clarification, and keep going on difficult goals. Meta evaluates this with FocusBench and SWE-Marathon, including long-running software projects.
③ Meta is targeting broad, real-world capability
Meta says the models lead across agent, memory, and coding benchmarks while generalizing better to unfamiliar tasks. The company is also emphasizing safeguards for agents that may access messages, calendars, and private documents.
✅ Action Item
Pick one multi-step workflow your team repeats every week and test Muse Wonder against it through the Meta Model API or Muse Code. Score it on context retention, recovery from blockers, completion quality, and how often a human must step in.
Source: imforza.link/S9nW7Hl
ALT Blue lines and arrows spiral around a bright central point in Meta's Muse Wonder artwork.
💬 imFAQs: Will AI assistants make company websites obsolete?
No. Your website picks up two new jobs on top of selling to the people who still visit.
When your customer tells Muse or another personal assistant to "find a plumber who can come Thursday," the assistant goes looking for facts it can confirm and a way to finish the job. Your website is the one place you control both:
📋 The fact sheet. Services with prices or ranges, hours, the towns you serve, all matching your Google Business Profile and directories word for word. If the assistant can confirm you from your own site, it doesn't have to trust someone else's summary of you
✅ The finish line. A real time slot, a real total, and a confirmation the moment it's booked. When Walmart let shoppers check out inside ChatGPT, those purchases converted at about a third of the rate of shoppers who clicked through to Walmart's own site, and Walmart moved checkout back home
👀 The pitch for people. The assistant sends your customer a one-line summary, and some of them tap through to look. The page still has to sell in ten seconds
One thing will look worse in your reports: human visits from customers who send software ahead of them. Watch completed bookings instead of sessions.
And skip the separate "AI version" of your site. The assistant reads the same page your customer does.
Read the full answer 👉 imforza.link/ai-assistants-f…
What's one question your customers still ask by phone because your website doesn't answer it?
ALT imFAQs graphic asking whether AI assistants will make company websites obsolete, with three answers.
It looks finished. Is it right?
Google will now tell anyone if a file came from AI.
Google's new Gemini takes a goal, not steps.
GPT-6 turns answers into tools you can use.
Polish is free now. Knowing what to trust is the part you keep.
5 minutes with imNEWS: imforza.link/subscribe
Every channel needs a name next to it.
Search, referrals, email, paid, social. Fill in the owners and every name is you.
Unowned channels don't get worked.
imforza.link/mp-video-2
A goal you can't count is a goal you can't miss.
"Grow the business" never fails, so it never teaches you anything. The 3-3-3 and 70-20-10 rules won't count for you either.
Three numbers will. Here's how to work them out: imforza.link/mp-carousel-x
ALT Card titled "A goal you can't count is a goal you can't miss." Two goal cards side by side. The first says "Grow the business," with an empty progress bar and the note "No way to tell." The second says "25 new clients," with a progress bar partly filled and the note "4 of 25. Now you know."
ALT Card titled "The rules won't count for you." Four cards for the 3-3-3 rule (limits scope), the 40-40-20 rule (splits effort), the 70-20-10 rule (splits budget), and the 5 C's (sizes up the room). Each card has a blank box labeled "Your number" with a question mark in it.
ALT Card titled "Three blanks replace 'grow.'" A one-page plan form with three empty fields: new customers needed, leads needed, and qualified visits needed. A cursor blinks in the first field.
ALT Card titled "Fill the blanks. Then plan." A form labeled "The guide's made-up law firm" shows 25 new clients needed, 63 booked consultations, and 263 qualified visits a month. An arrow points to two cards: the 3-3-3 rule, "Pick your three by the visits each can deliver," and the 70-20-10 rule, "Split a budget you actually worked out," each marked "Now useful."
📰 #imNEWS: @Forbes and @Forbes_Councils asked 19 agency leaders how businesses can generate more demand without blowing their budgets. imFORZA Founder and CEO Vinny La Barbera recommends reactivating the leads already sitting in your CRM instead of paying to replace them.
🔑 3 Key Takeaways
① Start With The Demand You Already Own
Old leads, existing customers and unconverted website traffic can produce pipeline without adding another acquisition channel.
② Follow-Up Speed And Persistence Matter
Many leads are lost because businesses respond too slowly or quit after a few emails. Better systems can recover opportunities you already paid to create.
③ Precision Beats More Spend
The strongest ideas in the article focus on authority, conversion, personalization and smarter use of existing data rather than simply increasing media budgets.
✅ Action Item
Pull every lead that inquired during the past two years but never bought. Segment them by original need, then send a short, personal reactivation message that starts a conversation instead of another generic campaign.
Source: imforza.link/AaeujG9
ALT Forbes article: 19 Ways To Generate More Demand Without Blowing Your Budget
📰 #imNEWS: @GoogleCloud introduced Gemini agent, a universal work agent that answers questions, handles knowledge work, creates media, writes and runs code, and operates across Workspace and enterprise systems. It adds persistent memory, reusable skills, multi-agent orchestration, model routing, governance, and cost controls so companies can delegate outcomes instead of managing isolated prompts.
🔑 3 Key Takeaways
① One agent, many jobs
Gemini combines research, content, coding, analysis, and task execution in one interface and API. It can work autonomously, respond to events, schedule tasks, and create temporary sub-agents for complex workflows.
② Context becomes infrastructure
The system connects to tools like Salesforce, ServiceNow, Slack, Microsoft 365, BigQuery, and MCP servers. It maintains session, semantic, procedural, and episodic memory so teams do not have to re-brief it every time.
③ Enterprise controls are part of the product
Google is pairing agent capabilities with identity, permissions, sandboxing, policy management, spend caps, and multi-model routing. The practical shift is from experimenting with chatbots to governing agents that perform real business work.
✅ Action Item
Pick one repeatable workflow that crosses multiple apps, then document its desired outcome, required systems, permissions, and approval checkpoints. Test whether an agent can complete the workflow end to end while keeping a human at the highest-risk decision points.
Source: imforza.link/5YHGruC
ALT Google Cloud illustration for Gemini at Work 2026