Whoever you hire to advertise your bakery will ask you this in the first call: "What is a new customer worth to you, and how will we know when an ad brought one in?" Many owners have no answer ready. That is normal, and it is fixable in an afternoon. This page starts from that question and works backwards, so you can answer it, decide what to do first, and judge whether outside help makes sense.
Start with the answer they want: what one new customer is worth
A bakery customer is not a single sale. Someone who buys one birthday cake may come back for a graduation order, a holiday box and a few weekend loaves. Look at your register or order notebook and estimate what a typical first-time buyer spends in the following months. Use rough, honest figures, and use gross profit rather than revenue, since flour, butter and labor come out of every sale. That number sets the ceiling on what you can pay to win a customer.
Decide what you are actually selling through ads
An agency cannot advertise "a bakery." It can advertise a Saturday pre-order box, custom cakes, wholesale trays for offices, or morning pastries for people within walking distance. Each has a different buyer, a different margin and a different lead time. Pick the one with the best margin that you can fill without overloading the oven. If you are unsure, the pre-order approach in our [bakery marketing plan](/guides/bakery-marketing-plan) is a good place to start.
Define a "customer" in terms you can count
The second half of the question is about proof. A person who likes your post is not a customer. A phone call, a pre-order form, a direction request or a coupon redeemed at the counter is. Choose one or two of these as your counted actions before spending anything. Then ask yourself whether you can see them at all today: is there a form, a tracked phone number, a code staff can enter at the register? If the honest answer is no, fixing that comes before buying any ads.
| Question an agency will ask | What a useful answer sounds like |
|---|---|
| What is a new customer worth? | Gross profit from a typical first buyer plus likely repeat orders, roughly estimated |
| What are you selling? | One product or service, such as custom cakes or weekend pre-orders |
| Where do your customers live? | The neighborhoods or towns you can realistically deliver to or be visited from |
| How do orders come in? | A form, a phone call, a walk-in code, or a message, each one countable |
| What can you lose while testing? | A fixed amount that would not hurt payroll or ingredient orders |
Keep your expectations small and local
When we merged the phrasings people use for this topic, six distinct queries added up to roughly 50 searches a month. That is a small pool, and most of those searchers are looking for a first step. The same logic applies to your own customers. You only need the few thousand people near your oven who might want what you bake, which is why local reach beats broad reach.
Do the free groundwork before paying for clicks
A complete Google Business Profile with current hours, real photos and a working order link costs nothing and catches people already searching nearby. Ask happy customers for reviews at the counter. Make sure your menu and ordering page work on a phone. Paid ads send people to these places, so weak ones waste the money. This groundwork is also what any agency will check first.
Which ads fit a bakery, and where each one lives
Search ads on Google catch people typing something like "custom birthday cake near me" when they are ready to order, so they suit cakes and special orders. Social ads on Facebook and Instagram work better for visual, impulse-friendly items such as seasonal boxes, and for reminding past buyers that a holiday is coming. If you want to see how each is run, our pages on [Google Ads](/services/google-ads) and [Meta ads](/services/meta-ads) explain the mechanics without any commitment.
Run a small test and read it honestly
Run one offer, one area and one counted action for a few weeks. At the end, divide what you spent by the customers you can actually point to, then compare that with the worth you estimated at the start. If you paid less than a customer is worth, hold steady or nudge up. If you paid more, change the offer or the page before raising the budget. Keep notes in a simple spreadsheet so the result survives a busy season.
When an outside pair of hands makes sense
Doing it yourself is a perfectly good choice while the budget is small. Outside help starts to earn its place when you have no time for a weekly check, when you cannot tell whether orders came from ads, or when tracking is the stuck part. If you hire, bring your worth-of-a-customer figure, your counted action and your loss limit. An agency that works from those three things is likely treating your money carefully. One that skips them is a warning sign.