There is a specific feeling that arrives around 11pm when you are behind on something. A launch email lands, the tool sounds like the missing piece, the price is lower than it will be tomorrow, and buying it feels like progress. Two weeks later it is another login you never use.
This is not a discipline problem. It is a decision-making problem, and it has a fix: a short list of questions you run before looking at the price. The takeaway up front — the only tools worth buying are the ones that remove a specific task you are actually doing this week. Everything else is a plan to become a different person, purchased in software form.
Question 1: What exact task does this remove from my week?
Write the answer as a sentence with a verb and a time. "This removes the two hours I spend every Monday pulling together a content calendar." "This removes the manual export I do after every campaign."
If you cannot write that sentence, you are not buying a tool. You are buying a hope. That is the single most useful test in this article, and it disqualifies most impulse purchases in about fifteen seconds.
Two things it catches:
- Tools for work you are not doing yet. A tool that automates outreach is worthless if you have never done outreach manually — you will not know if the automated version is any good, and you will not know what to fix when it underperforms.
- Tools for work you do once a quarter. Automating a two-hour task you do four times a year saves eight hours annually and costs you a subscription plus the learning curve. Do the arithmetic before, not after.
Question 2: Who made it, and what else have they shipped?
Every product has a person or a company behind it, and in the digital-product world that name is usually visible right on the sales page. Search it. You are looking for a track record, not a reputation — has this vendor shipped things before, do they still exist, is there support after launch, and what do people say six months later rather than during launch week.
The pattern worth noticing: many tools in this category are launched by known independent creators who list their name on the product. That is a good sign compared to an anonymous product, because a name is something you can follow up on. Curated storefronts make this easier by listing the vendor alongside every product — Eric Salo Online Business Offers, for instance, is a handpicked storefront of AI, traffic and marketing-education tools that names the vendor on each listing rather than presenting an anonymous catalogue, with longer written reviews on its companion site.
That does not outsource the decision to anyone. It just means you start from a shortlist someone stands behind, and then you still do question one.
Question 3: Does this replace something, or add to the pile?
Every new tool adds a login, a monthly line item, a place data can live, and a thing to remember. That overhead is real and nobody accounts for it at purchase time.
So ask: what am I cancelling if I buy this? A tool that replaces two existing subscriptions is a genuinely good deal even at a higher price. A tool that sits alongside everything you already pay for needs to justify itself as pure addition — which is a much higher bar than it looks.
This is the same logic as choosing a side hustle: the constraint is never ideas, it is your attention, and every addition spends some.
Question 4: What does leaving cost?
Before you go in, look at how you get out. Three things:
- The refund window and how to claim it. Written on the vendor's own checkout page, not in the affiliate's pitch. If you cannot find it in two minutes, that is information.
- Whether your data comes out. Anything holding your list, your content or your customer records should let you export in a normal format. A tool you cannot leave is not a tool, it is a landlord.
- Whether it is a subscription or a one-time purchase, and — for lifetime deals — what happens if the company stops maintaining it. Lifetime is the vendor's lifetime, not yours.
Question 5: Does the pitch survive being read slowly?
Marketing tools are, unsurprisingly, marketed well. That is not disqualifying. What matters is whether the claims hold up when you read them at normal speed instead of launch speed.
Things that should slow you down:
- A promised outcome with no mechanism. "Unlimited traffic" is a result. Ask what it actually does to produce traffic, and whether that thing is something you would want done in your name.
- Income figures. Nobody can promise you earnings, because earnings depend on your offer and your market. Any product marketing that guarantees income is telling you about the marketer, not the tool.
- Countdown pressure. Real launch discounts exist. So do permanent timers that reset when you clear your cookies. The difference takes thirty seconds to check.
- A demo that only shows setup. Ten minutes of configuration and zero minutes of output is a common tell.
None of this means "do not buy." It means buy the mechanism, not the outcome.
The ninety-second version
Before any purchase:
- Write the sentence: "this removes ___ from my week." No sentence, no purchase.
- Search the vendor's name. Look for a track record and post-launch support.
- Name what you are cancelling.
- Find the refund window and the export path.
- Reread the pitch slowly and separate the mechanism from the promise.
Five questions, ninety seconds, and it will save you more money this year than any discount code.
FAQ
Are AI marketing tools worth it, generally? Some genuinely are — the ones that compress a task you already do into less time. The ones that promise to do the strategy tend to disappoint, because the hard part was never producing the output; it was knowing which output to produce. Buy the compression, be sceptical of the substitution.
Is buying through an affiliate link bad? No. Affiliate commissions are how a lot of independent reviewers and curators fund the work, and it costs you nothing extra. What matters is whether the commercial relationship is disclosed and whether the recommendation would survive without it. A curator who tells you the model up front is behaving normally; one who hides it is the problem.
How many tools should I be running? Fewer than you are. A useful check once a quarter: list every subscription, and next to each write the task it removes. Anything with a blank next to it gets cancelled. Most people find two or three blanks immediately.
What if I buy something and never use it? Claim the refund if you are inside the window — that is what it is for, and using it is not rude. Outside the window, cancel the recurring charge today rather than "next month", and add the purchase to the mental list of what you already know about your own buying triggers. That list is worth more than the refund.
I keep buying tools instead of doing the work. How do I stop? Add a delay rule: nothing gets bought on the day you first see it. Most launch urgency is designed to survive exactly zero days of thought, so a 24-hour rule filters out almost all of it while costing you nothing on the genuinely good purchases.
Next step
Take whatever tab you currently have open with a tool in it and answer question one. If the sentence does not write itself, close the tab — that is the whole exercise.
If you would rather start from a shortlist than the open marketplace, Eric Salo Online Business Offers is a curated storefront of AI, traffic and marketing-education tools with the vendor named on every listing, and Eric Salo Reviews publishes the longer written takes. Just run the five questions anyway — a shortlist narrows the field, it does not make the decision for you. And when you are back to the earlier problem of what to actually build, Mellow Ideas is here for that.
