Performance Marketing
Your ads are running.
The customers are not arriving.
That is rarely a bidding problem. It is usually the offer, the creative, what you are telling the platform to count, or what happens after someone clicks. Those four are the work.
01 / The diagnosis
Why spend goes out and customers do not come back
Smart Bidding, Performance Max and Advantage+ now handle bid changes, placement choice and much of the audience finding. The platform runs those decisions on its own, thousands of times a day, faster than anyone can by hand.
What it cannot do is decide whether your offer is worth buying, produce creative worth testing, work out which of your leads were any good, or answer the phone. Those stay with the business. They are also where almost every underperforming account loses its money.
The platform decides
- Bid amounts, adjusted per auction
- Which placements your ads appear on
- Audience expansion beyond your targeting
- Budget pacing across the day
- Which creative gets shown to whom
- Keyword matching, often looser than you set
You still decide
- The offer, and whether anyone wants it
- How much creative you produce, and how different each piece is
- Which conversions you report back as valuable
- What happens after the click
- Which products get promoted at all
- Exclusions, budget caps and when to stop
The question is not who optimises the settings. It is who improves the inputs.
02 / Order of leverage
What usually moves the result most
Most weak accounts are not held back by a hidden setting. They are held back by one of four inputs, and the order matters.
The offer
An offer people do not want will not be rescued by better media buying. It is the first thing we check and the most common reason a campaign fails. It is also the cheapest thing to change. A week spent fixing the proposition can save three months of paying for clicks that were never likely to convert.
The creative
When bidding is automated, creative becomes one of the few major levers left in human hands. Five near-identical ads in different colours give the platform almost nothing to learn from. The machine needs genuinely different messages, hooks and formats to test. Knowing which variations are worth running, and which are just noise, is the part that still needs judgement.
The conversion signal
Platforms optimise towards whatever you tell them counts. If every form fill is marked as a win, the system will find you more form fills, including the weak ones. A lead recorded by the platform is often just a submitted form, not a saleable enquiry. The fix is to send back the leads that were qualified, answered, booked or closed. Conversion tracking and attribution are the plumbing that lets that feedback reach the account.
What happens after the click
A landing page mismatch, slow follow-up or weak sales handling can waste a decent campaign. Plenty of accounts described as an ads problem are really a response-time problem. If enquiries are not answered within an hour, performance usually drops long before media buying is the real issue.
03 / The sum that decides it
How much can I afford to pay for a customer?
This is one of the few sums worth doing before any campaign starts. If the numbers do not work on paper, they rarely work in the ad account.
Worked example, in any currency
At a 50 percent gross margin, a 2,000 sale leaves 1,000 before ad costs and sales costs. If 1 in 5 leads becomes a customer, you can spend up to 200 per lead before that gross profit is gone.
If your market is likely to produce leads at 350, paid acquisition is probably not the right move yet. A business that cannot state its gross profit per sale and its lead-to-customer rate is not ready to judge campaign performance.
04 / Where to start
Paid search, paid social, or neither yet
Start with the buyer's intent, not the channel preference.
Paid search suits you if
- People are already searching for the exact thing you sell
- Demand exists and you need to capture it before a competitor does
Paid social suits you if
- Almost nobody is searching, because the need is latent or easy to ignore
- Demand has to be created rather than harvested
If you do not yet have a clear offer, a workable margin or the ability to follow up fast, neither channel is the right first move. Roughly 30 conversions a month is a useful floor for most automated bidding setups. Below that, you are paying to teach the system without giving it enough repetition to improve. This is why two businesses on the same monthly budget can get very different outcomes.
05 / Selected engagements
What this looks like in practice
Figures are taken from the client ad accounts. Where a number was reported by the client rather than measured by us, it says so.
Video production studio
| Measure | Result |
|---|---|
| Media spend | ₹93,337 |
| Clicks | 2,354 |
| Click-through rate | 4.49% |
| Enquiries generated | c. 100 |
| Attainment against objective | 50% |
| Cost per enquiry | ₹933 |
| Projects won (client-reported) | c. 10 |
| Client project value | ₹2.5L to ₹5L |
| Cost per project won | ₹9,334 |
| Acquisition cost as share of project value | 1.9% to 3.7% |
Spend, clicks and click-through rate are taken from the client's Google Ads account. The enquiry-to-project conversion was reported by the client and has not been independently verified. The paid search engagement concluded after two months. Performance beyond that date is unknown and is not claimed here.
The campaign reached half its stated enquiry target and remained the client's most efficient acquisition channel over the period. At an acquisition cost below 4 percent of the smallest project value, the shortfall against target had no bearing on commercial viability. The enquiry target was the wrong measure to govern the engagement, and that was evident within two months rather than after a year of expenditure.
Travel operator
| Measure | Result |
|---|---|
| Media spend | ₹3,000 |
| Enquiries generated | 32 |
| Cost per enquiry | ₹94 |
| Bookings confirmed | 2 |
| Cost per booking | ₹1,500 |
| Further conversations commenced | Not attributed |
Enquiry volume is taken from the Meta lead form. Further conversations began during the campaign period with prospects who were served the advertisement, did not click it, reached the website independently and made contact there. These cannot be attributed to the campaign on the available evidence and are excluded from the figures above.
Platform reporting captures only the conversions it can claim. A proportion of the demand a campaign creates will never carry attributable identifiers. The excluded conversations are noted here because their omission is the point, not an oversight.
Click-through at that level comes from the offer and the creative, not from bid management.
06 / The first month
What actually happens if you hire us
In order, and in the order that matters rather than the order that looks busiest.
We work out whether this should run at all
We ask for your gross profit per sale and your lead-to-customer rate, and we read the existing account. From those two numbers we calculate what you can afford to pay for a lead. If the sums do not work, you hear it in the first week rather than the third month.
The conversion signal
We correct what the account is counting as a win. Qualified enquiries rather than every form submission, with the outcome fed back so the platform optimises towards business rather than volume. This is usually the single change with the largest effect, and it costs nothing in media.
Offer and creative into test
Genuinely different messages and formats, not colour variants of the same ad. Where the offer itself is the weak point, we say so and work on it with you before spending more on distributing it.
The path after the click
Landing page against ad promise, enquiry routing, and how fast enquiries get answered. Response time inside the first hour is checked early, because it frequently explains more than anything inside the ad account.
The numbers you get back
Spend, enquiries, qualified enquiries, what each one cost, and where people are dropping out. Platform metrics are in the report, but they are not the headline. The headline is what a customer cost you and whether that number works against your margin.
07 / Engagement
Four ways to work with us
Lead Hustle
You pay per qualified lead rather than for management time. We carry the delivery risk, so we only agree to it where the offer, margin and market can support it.
We decline this model more often than we accept it. If we say no to it, that answer is useful in itself, because it usually means something upstream would make paid acquisition poor value.
Managed campaigns
Ongoing strategy, creative, measurement and optimisation. Our fee is separate from your media spend, which goes directly to the platforms on your own billing.
UberBrains is a boutique consultancy with a small senior team, so the people you meet are the people who do the work.
Audit and fix
A fixed-scope review of an existing account, with the changes made rather than just listed. Most useful when spend is steady but results have drifted.
An audit only matters if it changes the account. The point is to remove waste, tighten signals and correct what is already costing money.
Second opinion
We review what your current agency is doing and tell you plainly whether it stands up. No handover, no forced switch, no pitch to replace them.
Sometimes the verdict is that the agency is doing a decent job and the problem sits with the offer, the site or the sales follow-up. That is still worth knowing before you change suppliers.
Your media spend and your agency fee should never be the same invoice.
Media spend is yours. It goes to Google or Meta on your own billing account, at cost, with no markup in between. You keep ownership of the ad accounts, the pixels and the conversion history from day one, and if you leave, those assets leave with you. Industry fees typically run at 10 to 20 percent of ad spend, which pays the agency more when you spend more. Ours does not move with your budget.
08 / Due diligence
Six questions worth asking us, or anyone else
Whoever you end up hiring, these answers tell you most of what you need to know. Ask for them in writing before a contract starts.
- Who owns the ad account, billing profile, pixels and conversion data from day one?
- Is your fee tied to media spend, or is it separate?
- What exactly will you report each month: spend, leads, qualified leads, sales, response times, or just platform numbers?
- What happens when results are poor: do you change the offer, the creative, the signal, the landing page, or only the bids?
- Will you say no if the campaign is a poor fit, or will you run it anyway?
- If we stop, what do we keep access to on the same day?
If the answers stay vague until after kickoff, expect the same pattern once money is in motion.
09 / Questions
Frequently Asked Questions
Why are my Google Ads not converting?
In many accounts the ads are not the main problem. The usual causes are loose conversion reporting, a weak offer, a landing page that answers a different question from the ad, or follow-up that is too slow.
Reply time within the first hour is one of the clearest numbers to check in an underperforming account. That single condition explains more wasted spend than most bid settings ever will.
How do I work out what I can afford to pay for a customer?
Start from gross profit per sale and your lead-to-customer rate. If an average sale is 2,000 at a 50 percent gross margin, that leaves 1,000 before ad costs and sales costs. If 1 in 5 leads becomes a customer, your ceiling is 200 per lead before that gross profit is gone.
If your market is likely to produce leads above that ceiling, paid acquisition is probably not the right move yet.
Should I start with paid search or paid social?
Start with buyer intent rather than channel preference. Paid search suits you when customers are already searching for the exact thing you sell and the demand just needs capturing. Paid social suits you when almost nobody is searching because the need is latent or easy to ignore, so demand has to be created.
If you do not yet have a clear offer, a workable margin or fast follow-up, neither is the right first move.
What is the minimum budget for paid ads to be worth running?
Small budgets do not fail because they are small. They fail because they do not generate enough signal for the platform to learn. Roughly 30 conversions a month is a useful floor for most automated bidding setups.
Work backwards from that. If your likely cost per click and conversion rate cannot produce around 30 conversions on your budget, the money is better spent elsewhere until the offer or the margin changes.
How long before I know whether paid campaigns are working?
Automated bidding usually needs 2 to 4 weeks of steady conversion volume before performance settles. Frequent changes to budgets, bids or conversion actions during that period often reset the learning.
Your sales cycle matters just as much. A business with same-day decisions can judge far sooner than one with a three-month buying cycle, and reading the account too early leads to bad calls.
Can you tell me what my cost per lead will be before we start?
Not reliably, and nobody else can either without seeing the offer, the market and your current data. Anyone giving you a neat number before that is guessing.
What we can do early is calculate what your cost per lead needs to stay under for the campaign to be worth running. That number comes from your margin, and it is the more useful one.
How do you charge, and is ad spend separate?
Our fee is separate from your media spend. Spend goes to Google or Meta on your own billing account, at cost, with no markup in between.
Industry rates commonly run at 10 to 20 percent of monthly ad spend. That structure pays an agency more when you spend more, which is the wrong incentive when part of the job is telling you when to stop.
Who owns the ad account, the pixels and the data if we stop working together?
You do. The ad accounts, the pixels and the conversion history stay yours throughout, and they leave with you.
If any agency insists on running media through its own account, ask what you lose when you leave. Accumulated account history and audience data are often worth more than the campaign currently running.
Can a pay-per-lead model work for my business?
Sometimes. Through Lead Hustle you pay per qualified lead rather than for management time, and we take the delivery risk.
We only agree to it where the offer, margin and market give it a realistic chance of working. We decline this model more often than we accept it, and if we decline, the reason is usually more valuable than a forced yes.
Is an agency still worth it now that platforms automate most of the old work?
Only if the work is focused on the parts the platform cannot own: the offer, the creative, the conversion signal and the post-click experience.
Platforms now automate bid management, placement choice and audience finding. Human judgement still matters in the inputs, in the commercial call on whether the campaign should run at all, and in spotting when the real problem sits outside the ad account.
Our Clients
+ Few Solopreneurs & Startups doing cool stuff in Food, Lifestyle, & Technology.
Some of them are providing hyperlocal services while some aiming the world. Some are close to securing their funding.
Chat with us on WhatsApp
Share your details and we'll follow up with you.
Cancel
Chat with us on WhatsApp
Share your details and we'll follow up with you.
Cancel