Looks like you're in . Want the local site?
Performance Max gives Google enormous control over where your budget goes — and without proper conversion signals, audience inputs, and brand exclusions, it defaults to the easiest spend: your own branded searches and remarketing lists. That is a problem you can fix, and it starts with understanding exactly what Google is doing with your money.
E-commerce store owners running PMax for the first time — You activated Performance Max because Google recommended it, and the dashboard shows a decent ROAS. But you have not separated it from your standard Shopping campaigns, and you are not sure whether the conversions being reported are new customers or repeat buyers clicking a branded ad.
Lead-gen businesses in Nepal seeing high volume but low quality — PMax is generating form fills, but the enquiries coming through are either low-intent or already familiar with your business. Google is optimising for conversion volume rather than conversion value, because no one has told it which conversions matter most.
Business owners who inherited a PMax campaign from a previous setup — The campaign was built by someone else, and you have never been inside the asset groups or audience signal settings. The conversion tracking may be double-counting, and the asset quality scores have never been reviewed.
Google needs signals to understand who your best customers are. Without a customer list, a high-value visitor segment, or structured first-party data, PMax guesses using general interest categories. In Nepal, where CRM adoption is lower, most campaigns launch with no signals whatsoever, forcing Google into a cold-start period it never actually exits.
PMax is bidding on searches like "[your brand name] Nepal" and "[your brand name] price" — people who were going to find you anyway. This inflates your reported ROAS because the conversion cost for branded clicks is low, but it masks the real cost of acquiring genuinely new customers. Your growth metrics look better than your business is actually performing.
A single asset group with generic headlines, one hero image, and no video means Google has very little to work with across six different placements. The algorithm deprioritises low-quality assets, which means your ads surface less often and in worse positions. A clothing business selling three different product lines from one asset group is leaving significant impression share on the table.
If your conversion event fires on the thank-you page load rather than on an actual form submission, or if the same purchase event is tracked in two places, PMax is optimising toward a signal that does not reflect real business outcomes. Value-based bidding is impossible without accurate purchase revenue data feeding into the campaign.
We verify that your primary conversion events are firing correctly and are not duplicated across Google Ads and Google Analytics imports. For e-commerce businesses, we confirm that purchase revenue values are passing through accurately so Google has the data it needs to run value-based bidding rather than optimising for volume alone.
We restructure your campaign into separate asset groups aligned to product category, service type, or customer intent. Each group gets its own headlines, descriptions, images, and where possible a short video. This gives the algorithm more relevant material per context and improves asset quality scores across placements.
For Nepal businesses without a large CRM database, we build audience signals from your website visitor lists — segmenting by pages visited, time on site, and prior purchasers. We configure custom segments based on search behaviour relevant to your category so Google has a directional signal from day one rather than starting blind.
We build a brand exclusion list at the campaign level to prevent PMax from bidding on your own brand name and domain. We also separate PMax from any existing standard Search campaigns targeting branded terms, so each campaign is doing a distinct job and the budget allocation reflects actual new-customer acquisition goals.
We set up placement reports to show you where your ads are running across the PMax network and flag any low-quality placements. We review search term transparency reports weekly in the first month, track asset performance ratings, and monitor for any signal drift that would indicate the algorithm is reverting to low-resistance spend patterns.
Performance Max and standard Search campaigns serve different functions and should run alongside each other, not replace one another. PMax covers the full Google network including Display, YouTube, and Gmail, while standard Search gives you keyword-level control and exact match precision for your highest-intent queries. Running them together without proper segmentation causes cannibalization, but removing Search entirely leaves you without the targeting control you need for bottom-funnel keywords.
High ROAS on PMax frequently indicates that a large share of conversions are coming from branded search terms — people who already intended to buy from you. The campaign appears efficient because branded clicks convert cheaply, but it is not generating new demand. A proper segmentation analysis will show you the split between branded and non-branded conversion value so you can see what the campaign is actually producing.
Yes, though the initial learning period is longer without CRM data to use as audience signals. We build signals from website visitor behaviour — segmenting by product pages visited, time on site, and return visits — which gives Google a directional input even before a customer list is large enough to use. As the campaign matures and conversion data accumulates, these signals are supplemented with lookalike-style expansion.
Most rebuilt PMax campaigns go through a learning period of two to four weeks after major changes. During this time the algorithm re-evaluates audiences, placements, and asset combinations against your conversion signals. Bid strategy performance during this window should not be used as the basis for decisions. Meaningful performance data typically becomes available from week five onward.
Google's Smart Bidding requires a minimum of around 30 to 50 conversions per month to exit the learning phase and optimise effectively. Below that threshold, the algorithm does not have enough signal to make reliable bid decisions. If your current conversion volume is below this, we will discuss whether PMax is the right campaign type at your current scale or whether a restructured standard campaign would serve you better in the short term.
The audit takes five days. We review your conversion setup, asset groups, audience signals, and brand exclusion configuration, then give you a written report with specific, prioritised changes. You do not need to commit to ongoing management to get the audit — it stands alone as a useful diagnostic.
5-day turnaround · PMax audit + signal review · No retainer required