It sounds like a simple question. But it gets right to the heart of how most advertisers on Bol work. And honestly: most of them do it at the wrong time, for the wrong reasons.
You see a high ACOS. First reaction: lower bids. You see that a keyword has not generated an order after three days. First reaction: turn it off. But in both cases you are reacting to incomplete data. And making decisions based on incomplete data leads to unstable campaigns, missed opportunities, and unnecessary losses.
In this article, we explain how to optimize in a structured way. What do you do daily, weekly and monthly? And just as importantly: when do you do nothing?
It doesn’t start with frequency. It starts with your product.
Before you think about how often you optimize, you need to know what to expect from your product. What is your conversion rate? What are your margins? How many clicks do you need for one order?
Only once you know that can you determine how much data you need to make a responsible decision.
A rule of thumb we use: give a keyword at least twice the number of clicks you need for one conversion before intervening. Do you have a product with a conversion rate of 5%? Then wait until you have at least 40 clicks. With a conversion rate of 2%, wait until you have 100 clicks.
Do it earlier, and you are reacting to noise. Not reality.
When can you intervene more quickly?
There are situations in which you can make a decision based on less data. If, intuitively, you can already see that a keyword is simply not relevant for your product, then you do not have to wait for 40 clicks. If you are advertising a two-person mattress on the keyword "mattress", you know you are in an exploratory, generic market. After 20 to 25 clicks without an order, it is defensible to make an adjustment.
But if you are convinced that a keyword matches your product and you expect high relevance, give it room. Twice the required number of clicks for one conversion is your guideline.
Daily: control, not optimization
On a daily basis, you do not adjust bids based on performance. In most cases, today’s data is still incomplete. Only the next day, and sometimes even a few days later, is yesterday’s data fully available in the dashboard.
What you do daily is check, not optimize.
What you check daily | Why |
|---|---|
Are campaigns being served? | No impressions means no traffic and no sales |
Are daily budgets running out too early? | 70% of conversions happen after 5:00 PM |
Have CPCs risen unexpectedly? | After bid changes, this can escalate quickly |
Are the right products running? | Prevent the wrong products from taking budget |
This is especially important after a major optimization round. Did you raise or lower bids yesterday on an important product? Then check the next day whether the campaign is still running well. Not to adjust performance, but to prevent something from having gone structurally wrong.
So reflect daily. But do not intervene daily.
Weekly: this is where the real optimization happens
Based on the past 7 days, in most cases you have enough data to make adjustments. This is the time for the real optimization round.
Adjust bids based on ACOS
Look per product and per keyword to see whether performance deviates from your target ACOS. Are products structurally above your desired ACOS? Lower the bid. Are products performing well within your target with enough volume? Then there is room to scale up.
Always do this with extra context. Also zoom out to the past 30 days before intervening. A keyword that performed poorly last week but ran consistently well for the three months before that should not be scaled down blindly. That is exactly when many advertisers misjudge shorter datasets.
Exclude bad keywords
On a weekly basis, you also filter out the loss-making placements. Keywords that consistently generate clicks but never convert, with enough data to justify that conclusion, can be excluded or reduced.
Track organic positions
Advertising is not only for direct revenue. It contributes to your discoverability on the platform. Bol re-indexes organic positions every day. Each week, check whether your ad spend contributes to a rising or stable organic position.
If you spend a lot but still lose organic position, something is fundamentally wrong. You want to detect that on a weekly basis, not only after a month.
What you optimize weekly | Based on |
|---|---|
Bids per keyword and product | Past 7 days, checked against 30 days |
Excluding bad keywords | At least 2x the clicks needed for conversion |
Organic position per product | Week-over-week comparison |
Inventory levels | Scale campaigns down in time when stock is low |
Monthly: the health check
Monthly, you step away from the daily and weekly numbers and look at the bigger picture. This is not an optimization round. This is a strategic check.
Profitability at product level
What is your TACoS over the past month? What has advertising contributed to your company’s total revenue? Are there products that cost money month after month without a structural contribution to growth or position?
This is the moment not to panic if your TACoS is high. Are you in the process of launching new products? Then higher spend is logical and temporary. As long as you have mapped that out in advance, you continue making data-driven decisions instead of reacting emotionally.
Inventory forecasts
On a monthly basis, you also look at your inventory over the medium term. Will you still have enough stock in six weeks? Then you want to align your ad budget accordingly. Pushing too hard while a new batch is on the way leads to out-of-stock products, ranking loss, and missed revenue.
Pricing strategy
A longer time horizon also gives you insight into whether your price is still right in the market. Have your margins changed enough that there is room for a larger ad budget? Or do you need to adjust the price to remain profitable while advertising?
What you evaluate monthly | Why |
|---|---|
TACoS per product | Is advertising structurally profitable? |
Inventory forecasts | Scale campaigns up or down in time |
Pricing strategy | Margins determine how much room you have |
Product launches | Higher spend temporarily acceptable, consciously tracked |
Organic growth | Does advertising contribute to long-term position? |
Why does my account feel so unstable?
Many sellers recognize this feeling. Results go all over the place. ACOS shoots up and down. There seems to be no pattern.
Often the answer is simple: too little volume. With little data and few products, you see large fluctuations. That is not a reason to intervene harder, but rather a reason to be more patient. The more you scale and the more products you have running, the more stable the whole picture becomes.
In other cases, it comes down to the time of day. Friday and Saturday perform very differently from Monday and Tuesday. Conversion rates vary strongly by day of the week. If you decide based on one day’s data, you are comparing apples and oranges.
And sometimes the solution is dayparting: automatically turning campaigns on and off based on which hours of the day perform for your product. That smooths out the fluctuations and concentrates your budget on the moments when it really works.
The optimization cycle at a glance
Frequency | What you do | What you do not do |
|---|---|---|
Daily | Check delivery, budget, and outliers | Adjust bids based on performance |
Weekly | Adjust bids, exclude bad keywords, check organic position | Make decisions on less data than 2x clicks per conversion |
Monthly | Evaluate TACoS, inventory forecasts, assess pricing strategy | Panic over temporarily higher spend during launches |
The core of good advertising
Structure. Data-driven work. And knowing when to intervene and when to wait.
Not every movement in your dashboard requires action. But recognizing the right signals at the right time, that is what makes the difference between advertising that costs money and advertising that makes money.
The advertisers who perform best are not the ones who optimize most often. They are the ones who know when to wait and when to act.
Want to partially automate the optimization process without losing control? See how Adyard uses bid automation and dayparting to achieve structurally better results, or start a free trial.
