Budget for a start-up online store: where to begin
The budget for a start-up online store is the sum with which you buy information, not sales. The first money spent shows which products sell, which message holds attention, where the people who reach checkout come from, and where the path breaks before payment. That is why getting started is an arrangement, not a gamble: first the measurement and the product pages, then one channel that goes first, then the split between it and content. Planning with an expectation of profit from the first month is the most common reason new stores stop advertising just before it has started to work. The three most expensive mistakes at the start often come from haste, not from the size of the sum.
My name is Hristiyan Panov — a performance marketing consultant with over 5 years’ experience and over €2 000 000 advertising budget managed for over 90 companies in Google and Meta. Below is what must be ready before launch, which channel goes first and how the money is split.
What needs to be ready before the first lev?
- Tracking, reconciled with real orders. Not “pixel installed”, but a verified match.
- Product pages that answer questions — price and delivery time, return policy, enough photos.
- Checkout without surprises. The delivery price should be visible before the last step.
- Stock and feed, if you are going to run Shopping.
- A calculated break-even threshold. Without it you have no criterion for what counts as success.
Advertising on a site that is not ready is not a launch, but paying to find out that the site is not ready.
Which channel goes first?
It depends on whether people are already searching for your product. If they are, Google captures intent that already exists and usually gives a faster first result. If the product is new or an impulse buy, Meta creates the demand, but needs more time and more creatives.
For a store with a clean feed, Shopping is often the cheapest way into Google. For a niche product with no search, Meta is the only option.
How is the budget split at the start?
Fewer channels, more on each. Spreading across Google, Meta, TikTok and email at the same time means none of them collects enough data to leave the learning phase. The model that works: one main channel with a budget that gives enough conversions per week, and a second one only when the first is stable.
Within a channel, concentration is just as important. One campaign with a clear purchase objective collects signal faster than several parallel tests. ABO gives control over which ad set gets how much while the volume is small; CBO makes sense when there is already more than one working audience. Also set aside part of the advertising budget for new creatives, because the channel wears out without them.
“Stable” is not a good week. It is checked by an equal CPA over several consecutive weeks with unchanged settings, by a conversion volume above the learning threshold, and by a match between the platform report and GA4 via UTM. The second channel is added without cutting the first one’s budget — otherwise both go back into learning at the same time.
How much should you set aside for content?
On Meta the creative does the job that keywords do in Google. A store that gives everything to budget and nothing to photos and video usually pays more per order — and you cannot test messages if you have only one. Plan the production as part of the budget, not as a one-off cost before launch.
The working minimum is measured not in the number of files but in the number of different angles: price, delivery time, composition or material, use, customer review. The same frame with a different font is not a test. Shoot in square and vertical, combine static with short video, and keep the first seconds clean — that is where it is decided whether anyone watches at all.
Fatigue is recognised by the numbers, not by feel: CTR falls, frequency rises, CPA climbs with an unchanged audience and an unchanged budget. Then the message is changed, not just the format. That is why production runs in a cycle — the new batch is prepared while the old one is still running, so there is no empty ad set.
When should you expect a result?
The first orders may come within days. A stable picture comes once enough conversions have accumulated, usually a few weeks. A decision on whether the channel works — no earlier than a month, and only with unchanged settings. Any change to the budget or the structure in this period restarts the learning and delays the answer.
“Enough” is measured in conversions per ad set per week, not in elapsed days. While the volume is below the learning-phase threshold, delivery is unstable and the cost per order jumps for no visible reason. That is why in the first weeks you touch things rarely: changing the optimisation event, the audience or sharply moving the budget puts the campaign back into learning.
Read the result at the level at which you will make the decision — campaign or ad set, not an individual ad with a handful of clicks. Reconcile the orders in the platform with the orders in the store for the same period and use the same attribution window throughout. Changing the window changes the number without anything having happened in the account.
What are the three most expensive mistakes at the start?
- Launching with unverified tracking. Everything after that rests on wrong numbers. The check is simple: reconcile the orders reported by the platform with the orders in the store for the same period and see whether a value is being passed, not just an event.
- Too many campaigns. Five campaigns with a little budget each learn more slowly than one with enough. Splitting by device, by gender and by many separate audiences looks like control, but it cuts the signal into pieces too small for optimisation.
- Stopping after a weak week. Starting and stopping is the most expensive possible cycle — every new launch starts from zero and you pay for the learning again.
The three usually come together: unverified numbers make the account look worse than it is; the bad picture leads to new campaigns; the new campaigns dilute the budget and lead to stopping. The way out is the reverse order — reconciled numbers first, then one campaign with enough budget, then patience with unchanged settings.
When is it better NOT to advertise yet?
When the margin does not allow you to pay for a customer, when the site has an obvious defect in checkout, and when there is nobody to process the orders. In these cases the same money, put into the product or the site, comes back faster.
The margin is checked before the start, not after the first week. Revenue minus cost of goods, delivery, payment fees and returns shows what part of the order remains, and from that comes the threshold ROAS below which every order is a loss. If the threshold turns out to be unreachable for your niche, the work is on the price and the average order value, not on the advertising.
A checkout defect shows up in minutes from a phone: a delivery price that appears only at the last step, mandatory registration, a single payment method, a field that does not accept a Bulgarian address. A lack of capacity is the same brake — delayed shipments and unanswered enquiries eat what the advertising has brought in. Advertise when the path after the click holds up.
FAQ — Frequently asked questions about the first budget
What is the minimum sensible budget?
Enough for the campaign to collect enough conversions per week. For an expensive product and a long path to purchase, the threshold is higher.
Should I hire a specialist right away?
If the budget is small, first fix the site and the tracking — the return from the same money is greater there.
Is it worth testing several products?
Yes, but not all at once. Two or three with a clear difference between them give more readable data than the whole catalogue.
When should I add a second channel?
When the first is stably above your threshold and has scaled to its ceiling.
What about organic traffic?
It is slower, but cheaper in the long run. They do not exclude each other — advertising buys time while organic grows.
Conclusion
The first months are planned as research with a budget, not as an investment with an expected return. Stores that accept this usually reach profit faster than those chasing a return from the first week.
If you are starting from scratch and want to avoid paying for the obvious mistakes, start with an independent review.
Request a quote for a digital audit →
Related articles: ROAS and margin thresholds · product page · Google Shopping
