In partnership with
How Jennifer Aniston’s LolaVie brand grew sales 40% with CTV ads
The DTC beauty category is crowded. To break through, Jennifer Aniston’s brand LolaVie, worked with Roku Ads Manager to easily set up, test, and optimize CTV ad creatives. The campaign helped drive a big lift in sales and customer growth, helping LolaVie break through in the crowded beauty category.
Open your credit card statement. Scroll to the subscriptions. Count the AI tools.
If you just winced, congratulations. You are the reason this article exists.
Most entrepreneurs running a small business in 2026 have turned their tech stack into a junk drawer. There is an image generator nobody opens anymore, a transcription tool that got replaced two quarters ago, a writing assistant that sits untouched because the team forgot the login, and a workflow automation platform nobody has actually connected to anything. It all charges on the first of the month. It all keeps charging whether you use it or not.
This is the first Monday of a new production week, so let me offer you a better opening move than reviewing yet another shiny launch announcement. Run the audit. Strip the bloat. Keep what actually pulls its weight. Then start the week lean enough to sprint.
Here is how to do it in under ninety minutes.
Pull every recurring charge into one place
You cannot fix what you cannot see. Start there.
Open your bank and credit card statements for the last three months. Pull every recurring software charge into a single spreadsheet. One column for the tool name, one for the monthly or annual cost, one for who owns it, one for the last time somebody on your team actually used it.
If you want a shortcut, Rize.io tracks which applications you and your team actually open during work hours. You will discover, usually with some embarrassment, that a handful of tools you are paying for have not been launched in six weeks. That is audit signal number one. Nobody is touching it. Kill it.
You are looking for three categories. Active, meaning someone on the team uses it weekly and can point to a specific outcome it produces. Dormant, meaning nobody has logged in during the last thirty days. Redundant, meaning it does exactly what another tool in your stack already does, just slightly differently.
Everything dormant gets canceled this week. Everything redundant gets consolidated. Everything active gets scored against what follows.
Score every tool against a revenue outcome
Most small business AI stacks were built by the worst kind of decision maker. The enthusiastic one. Somebody read a tweet, watched a demo, signed up for the free trial, forgot to cancel, and now you are paying for it.
That is not strategy. That is impulse shopping with your operating account.
For every tool in the active column, ask one question. What specific revenue, retention, or margin outcome does this tool contribute to, and can I measure it?
If the honest answer is some version of “it kind of helps me write faster,” you are not measuring. You are rationalizing. Faster than what, produced for what outcome, generating what return? If the number is zero or you cannot find it, that is a tool on probation.
A good litmus test. Imagine the tool disappeared tomorrow. What would break? If the answer is nothing measurable, the tool is a hobby, not a line item. Hobbies do not belong on the company card.
The tools that earn their seat have clear answers. The meeting recorder from Fathom saves me three hours a week and gives me searchable client notes I use in every follow up. That is specific. That is measurable. That is paid for, ten times over, every month.
Consolidate your model access before it consolidates your budget
Here is a category where most operators are bleeding cash without realizing it. Foundation model subscriptions.
Many businesses are paying separate monthly fees for ChatGPT, Claude, Grok, and HeyGen, plus a few more they forgot about. Each of those runs twenty dollars a month, sometimes more for the pro tiers. Stack four or five on one team and you are looking at several hundred dollars a month for access you could buy once.
This is where Galaxy.ai earns its keep. One login, access to most of the major frontier models. One invoice. No more phantom subscriptions. No more team members expensing three different AI tools that all do essentially the same job with slightly different branding.
I am not saying this as a universal rule. Some teams need enterprise features only a direct vendor provides, and that is fine. Most do not. Most are paying for brand logos on invoices when they could be paying for capability.
Run the math. Add up what you currently spend on individual model subscriptions across the team. Compare to the cost of a consolidated aggregator. If the difference is not meaningful, stay put. If it saves you two hundred a month without losing anything that matters, move.
Audit your automation layer next
After models, the second biggest leak is the automation layer. Zapier plans. Make.com workspaces. Scattered low code tools. Internal scripts nobody maintains.
Start by listing every automation currently running in your business. Not the ones somebody built six months ago and forgot about. The ones actually firing today. For each one, note what triggers it, what it does, and what would happen if it stopped working tomorrow.
You will find three types. The automations that save real time and produce a measurable outcome. The automations that fire into the void because the downstream system was turned off and nobody noticed. And the automations nobody remembers building but somebody apparently did.
Kill the second category immediately. Rebuild the third category only if the outcome matters. Keep and document the first category like your life depends on it, because if the person who built them ever leaves, it will.
My standing recommendation here is Make.com. Visual, cheaper than most competitors at scale, handles branching logic without turning into spaghetti. If you are on something more expensive and not using the advanced features, you are paying for a premium you do not consume.
Kill the tools that pretend to be AI
This one is uncomfortable, but someone has to say it.
In 2026, everything has an “AI” feature. The invoicing tool has AI. The scheduling tool has AI. The email client has AI. The grammar checker has AI. Most of these are wrappers around the same three or four foundation models, layered under a marketing refresh.
Ask yourself. Is the AI feature here actually useful, or did the vendor just bolt a chatbot onto an existing product to justify a price hike? If you are paying a premium for AI features you never use, you are subsidizing a roadmap that does not benefit you.
The test. Turn off the AI add on for one week. Keep notes on what you miss. If at the end of the week the answer is “nothing specific,” downgrade the plan. Vendors love quiet upgrades. They rarely remind you when you could quietly downgrade and save two thirds of the monthly cost.
Consolidate your CRM and outreach stack
If you have a CRM, an email tool, an outreach tool, a lead scoring tool, and a scheduling tool, all from different vendors, and they only partially talk to each other, you have a stack that looks impressive in a vendor presentation and drains cash in practice.
Pick one hub. Most small businesses do not need five specialized tools. They need one platform that covers the core and integrates well with what they already use. Go High Level is a reasonable starting point if you want CRM plus outreach plus automation in one place at a price that does not require a finance meeting. Clay is a better pick if your primary motion is outbound and you care about enriched data.
The principle is the same either way. Fewer tools, tighter integration, less leakage. Every handoff between systems is a place where data gets lost, leads get dropped, and somebody ends up blaming the intern.
Rebuild your content production stack around outcomes
The content side of most stacks is where the clutter gets truly impressive. Image generators, video tools, scheduling platforms, editing suites, brand asset managers, template libraries, analytics dashboards.
Here is the filter. What content actually drives a business outcome, and what content exists because you felt guilty about the empty feed?
If your social posts do not convert, produce leads, or build meaningful distribution, you do not need five tools to produce them. You need a hard conversation about whether that channel is worth the energy at all. Once you decide yes, pick one scheduler. Buffer does the job for a reasonable fee and does not try to sell you forty adjacent features.
For writing, pick one primary model and stop switching. For images, pick one generator and learn its style instead of bouncing between three. For video, if you need talking head content without a camera setup, HeyGen delivers. Pick the tool that matches your output. Stop paying for optionality you never use.
Build a one page operating stack
Here is what you want at the end of the audit. A single page document that lists every tool in your stack, what it does, what business outcome it serves, who owns it, and what it costs annually. Total at the bottom.
This document does three things. It forces honesty, because you cannot hide bloat on one page. It creates a target for next quarter, because once you see the total, you will naturally start asking whether that number should be higher or lower. And it protects continuity, because if anyone on your team leaves tomorrow, this is the document that lets you keep running.
Most businesses skip this step because it feels like overhead. It is not overhead. It is the difference between owning your operation and being a passenger on a subscription treadmill.
Kill the hidden enterprise seats
Here is a bleed that almost nobody tracks. Per seat pricing on tools that only two people actually use.
You pay for five seats on a platform that three people have logged into once. You pay for ten seats on a communications tool because “the team needs access,” and then half of those team members communicate in a completely different tool. Audit every per seat charge. Reduce to actual heavy users. Give everyone else view only or free tier access. You will be surprised how much of your stack was sized for a team you imagined, not the team you have.
Set a rule for every new tool
Now that the stack is clean, protect it. Write a one line rule for how any new tool enters your business. Here is mine.
Any new recurring charge requires a thirty day trial with a defined success metric, a named owner, and a calendar reminder to reassess before the trial converts to paid.
That is it. Three conditions. No new charge without them. This single rule will prevent nine tenths of the bloat you just spent the morning cleaning up. It is also the discipline most operators refuse to adopt, which is why most stacks look like yours used to before you ran this audit.
The Monday discipline
Here is the uncomfortable truth. Most small business owners are great at buying tools and terrible at retiring them. Tools accumulate. Subscriptions persist. The stack grows until somebody has a bad month and finally looks at the bill and panics.
Do not be that operator. Make the audit a Monday ritual every quarter. Ninety minutes, four times a year, and your stack will never get out of control again. Set the calendar invite now. Title it Stack Audit. Make it recurring. Treat it like a board meeting because that is effectively what it is. You are the board, the operator, and the shareholder. You owe yourself an honest look.
One more thing before you close this tab. After you finish the audit, tell one other operator what you cut and what you saved. Not because you are bragging. Because stating the number out loud makes the discipline stick. The operators who save money quietly tend to quietly start spending it again next quarter. The ones who say the number publicly tend to protect it.
Lean stack, clear outcomes, measured cost. That is the foundation everything else sits on. Build it this morning and the rest of the week gets cheaper, faster, and a lot more honest.
Ready to stop paying for tools that do not pay you back? Reply to this email with the word BLUEPRINT and I will send you the AI Workflow Blueprint, the exact forty seven dollar system I use to map a lean AI stack to a profitable business. If you want the full walkthrough with live workshops and implementation support, reply ACCELERATOR for the AI Business Accelerator.
Tomorrow we get tactical with automations that pay for your whole stack in a week.
Jordan Hale The AI Newsroom
Crash Expert: “This Looks Like 1929” → 71,048 Diversifying Here
Mark Spitznagel, who made $1B in a single day during the 2015 flash crash, warned markets are mimicking 1929. Seems extreme but we did just see the worst quarter for the S&P since 2022.
So it’s not so surprising that Vanguard and Goldman Sachs forecasted 5% and 3% annual S&P returns respectively for 2024-2034.
Almost no one knows this, but postwar and contemporary art appreciated 10.2% annually with near-zero correlation to equities from 1995–2025 overall.*
And sure… billionaires like Bezos can make headlines at auction, but what about the rest of us?
Masterworks makes it possible to invest in legendary artworks by Banksy, Basquiat, Picasso, and more – without spending millions.
28 exits. Net annualized returns like 14.6%, 17.6%, and 17.8% on works held over 1 year+. $1.3 billion invested. 500+ offerings.*
Shares in new offerings can sell quickly but…
*According to Masterworks data. Past performance is not indicative of future returns. Investing involves risk. Important Reg A disclosures: masterworks.com/cd.