How much does your company's invisibility in AI answers cost, and how do you calculate it?

42.51% of search results pages contained an AI Overview in the fourth quarter of 2024, and the figure grew by 8.83 percentage points quarter over quarter, according to Advanced Web Ranking data. Since the share was growing at that pace, the gap is bigger today, not smaller, so every subsequent customer purchase decision now starts in the AI window, before any link to the company's website even appears.

Last week I heard a sentence from the founder of a Polish tech company that I hear every month: "I go to meetings from referrals myself, I don't have time to run the company." This person doesn't need another growth promise, because he knows perfectly well that a calendar full of referral meetings also means no window to decide on a system. A B2B SME owner needs a calculation: how much does the lack of presence cost, measured in lost queries, delayed revenue, and rising spend on paid channels.

At AGAPE, I calculated this for the first time in March and left out margin, which made the number come out several dozen percent too high. When I corrected the model, it turned out that the cost of invisibility is a risk item that can be entered into a decision spreadsheet next to salary cost and server cost. Below I show exactly how to build this calculation using only data from CRM, financial systems, and the founder's calendar.

The mechanism behind disappearing queries

Three layers of invisibility you measure

Before you calculate the amount, you need to define exactly what you're measuring, because invisibility in AI answers has three layers. The first layer is the absence of brand mentions in the answers of tools like ChatGPT, Perplexity, Claude, Google AI Overviews, or Copilot for queries describing your category, customer problems, and typical contract size. Checking this yourself takes 30 minutes: you enter 20 queries your customers use, and note whether your company appears in the answer. The second layer is the absence of citations of company pages as sources in these answers, even if the brand is mentioned in the content. The third layer is the absence of traffic from AI platforms in analytics, and the absence of queries in sales conversations that start with statements like "I found you on ChatGPT" or "Perplexity recommended you in this category."

These three layers add up to one question every SME CEO asks themselves: how much does absence from ChatGPT cost when a customer asks about my category. The answer is calculable.

The scale of the phenomenon in numbers

A Seer Interactive study conducted on more than 25 million organic impressions and 1.1 million paid impressions showed that for informational queries displaying an AI Overview, the organic click-through rate drops by 61%, and the paid one by 68%. Even without AI summaries, organic CTR dropped by 41%, indicating a broader shift in user behavior as they move part of their traffic to AI chats and social search.

Pew Research, monitoring 68,000 real queries, found that when an AI summary appears on the results page, users click on results only 8% of the time, versus 15% when there's no summary. That's a relative drop of 46.7%, based on real queries, not aggregated impressions. In extreme cases, such as DMG Media and Daily Mail, CTR dropped by as much as 89% for content where an AI Overview appeared.

These figures relate to publishers, but the mechanism works identically in B2B, where a customer is looking for a business process automation vendor. If your company doesn't appear in the model's answer, the customer doesn't visit your website, doesn't fill out a form, and doesn't call. Absence from the model's answer means automatic exclusion from any further conversation.

AI traffic quality is higher, so the loss is more expensive

Forrester quotes Kasada's CMO, who points out that users coming from tools like ChatGPT spend three times more time on a page than those from traditional search engines, and their queries average 15-23 words. This is typical of a more advanced stage of the buying process, when the customer already has organized knowledge and is looking for specific solutions. For a B2B SME CEO, this means that invisibility in the answers of such tools means losing queries with higher purchase intent, i.e., more valuable ones.

Forrester, in its Buyers' Journey Survey 2024, found that 89% of surveyed B2B buyers use generative AI in at least one stage of the buying process. A synthesis by Gracker AI of four independent studies with a sample of more than 6,000 buyers showed that 50% of B2B software buyers now start their buying process in an AI chat window, and in the technology sector that figure reaches 80%. 6sense research, cited in Gracker AI's analyses, shows that 95% of vendors who get chosen are already on the shortlist on day one.

When 95% of chosen vendors are on the shortlist on day one, and that shortlist is shaped by AI answers, absence from that answer means losing the right to even enter the conversation. This isn't a drop in traffic; it's exclusion from the game.

What's changing in user behavior

Advanced Web Ranking shows that CTR for "what, when, where, how" queries dropped on desktop by more than 7 percentage points for the top four positions. In plain terms: an increasing share of queries, especially informational ones, are answered by AI modules, and users click on classic results less often. Brands present in AI Overview get 35% more organic traffic and 91% more paid clicks than brands that are omitted. These differences prove that the gap is calculable in both directions.

A model for calculating the cost of invisibility

Five steps using CRM data

The cost of invisibility in AI answers is calculated in five steps, using only internal company data. Each step has one operational instruction and one data source.

1. Number of monthly queries from a year ago compared with today. Pull the number of leads from the last 12 months out of the CRM and compare it with the figure from the previous 12 months. Only this difference shows the scale of the phenomenon, because a single number says nothing about the trend. Check the median, not the average, because one large deal can distort the picture.
2. Percentage of queries that never reach the conversation stage. Here you need B2B funnel benchmarks. MarketJoy reports that average conversions are 20-25% from lead to MQL, 12-18% from MQL to SQL, 10-12% from SQL to opportunity, and 6-9% from opportunity to close. Without these numbers, a company estimates the cost of invisibility by feel and overstates or understates the result by tens of percent.
3. Average annual contract value (ACV). Pull the average gross revenue per customer in the first year of cooperation out of your financial system. In Polish conditions for B2B SMEs, we often talk about a range of PLN 30,000-300,000 in annual customer value. Work with your own numbers, not benchmarks.
4. Sales cycle length. ORM-Tech cites benchmarks showing that cycles for contracts below USD 25,000 ACV are shorter and have higher win rates (30-40%), while for contracts above USD 100,000 ACV, cycles are longer and win rates drop to 15-20%. For SMEs with ACV above PLN 50,000, this means that every month of invisibility means missing revenue in the budget for 3-9 months.
5. Cost of acquiring one meeting from paid channels. Customer Acquisition Cost (CAC) is the total sales and marketing cost divided by the number of new customers acquired in a given period. It includes not only ad spend, but also sales reps' salaries, commissions, tools, and agency fees. This step provides a benchmark for the alternative spend the company would have to incur to make up for lost queries.

The query value formula

The value of a single query is expressed by the formula: query value = probability of closing a deal × average contract value (ACV). The probability of closing a deal is the product of conversions at every stage of the funnel. If you assume conservative assumptions that 20% of leads move to MQL, 15% from MQL to SQL, 10% from SQL to opportunity, and 7% from opportunity to close, the total conversion from lead to customer is about 0.21%. That means about 2 customers per 1,000 queries.

Multiplying the value of a query by the number of lost monthly queries gives the monthly cost of invisibility. The same calculation at the margin level shows the amount the company's budget is actually fighting for. To convert revenue into margin, multiply the result by direct margin (the percentage of revenue left after the direct costs of delivering the service). A CEO should look at the margin version, because it shows how much cash actually stays in the company.

A numerical example

Let's build an explicitly labeled illustrative example for a company with 60 monthly queries that saw a drop to 45 queries, i.e., lost 15 queries per month.

These numbers look low because we assumed a very conservative conversion rate. In reality, B2B companies often have higher conversion at the opportunity stage, especially when the lead comes from AI, because it's better qualified. If you assume a 1% conversion rate (which is still conservative for high intent), the values grow fivefold. With an ACV of PLN 120,000 and 15 lost queries per month, the monthly cost of invisibility is PLN 18,000, and on an annual basis, PLN 216,000.

For a company with 100 monthly queries and an ACV of PLN 150,000, at a 1% conversion rate, the monthly cost of invisibility reaches PLN 150,000. This shows how the assumptions produce the lower and upper bounds of the PLN 50,000-150,000 monthly range.

Revenue vs. margin

Steps 1-4 calculate lost revenue, while step 5 calculates the alternative cost of recovery. These numbers shouldn't be added together uncritically, because they belong to different categories. If you want to calculate the real impact on the budget, multiply lost revenue by direct margin. For a service company with a 40% margin, PLN 150,000 of lost revenue per month means PLN 60,000 of lost margin. This is the amount a CEO should enter into a decision spreadsheet next to salary cost and server cost.

Cost of acquiring a meeting from paid channels

Step 5 requires a specific benchmark. According to SaaS Tweaks data for B2B SaaS in 2026, the average CAC in paid channels is USD 200 to 500 per lead, and the cost of acquiring one meeting (SQL) reaches USD 500-1,500. In Polish conditions for B2B SMEs, the cost of acquiring one meeting from paid channels is PLN 800 to 2,500. With 15 lost meetings per month, recovering them through paid channels would cost PLN 12,000 to 37,500 per month. That's a budget the company hasn't planned for.

Practical application of the data

What the calculation looks like in a typical Polish B2B company

In a typical Polish B2B company operating on contracts above PLN 50,000 a year and dozens of monthly queries, the cost of invisibility ranges between PLN 50,000 and PLN 150,000 of lost potential revenue per month. This amount results from a combination of CTR declines, the shift of purchase intent to AI chats, and rising CAC benchmarks in paid channels.

Let's build an explicitly labeled illustrative example for a company with 80 monthly queries that saw a drop to 60 queries, i.e., lost 20 queries per month. With an ACV of PLN 80,000 and a conversion rate of 0.5% (conservative), the value of one query is PLN 400, and the monthly cost of invisibility is PLN 8,000. At a conversion rate of 1.5% (realistic for high intent), the value of one query is PLN 1,200, and the monthly cost of invisibility is PLN 24,000. On an annual basis, that comes to between PLN 96,000 and PLN 288,000 of lost revenue.

If the company operates on higher contracts, for example an ACV of PLN 200,000, with 20 lost queries and a 1% conversion rate, the monthly cost of invisibility is PLN 40,000, and the annual cost is PLN 480,000. That's an amount that justifies any investment in AI visibility, as long as it's calculated and presented to business partners in the language of numbers.

Why this is a planning tool, not a reaction

A sales cycle of six months means that decisions made today about measuring AI visibility will only affect revenue in half a year. A pipeline without AI queries today means missing revenue in the budget two quarters from now. That's why this calculation is a planning tool, not a reaction.

A founder who postpones the decision to measure AI visibility loses control over the timing and flow of revenue. His calendar, full of referral meetings, doesn't allow him to work systematically on the pipeline. After reading this, he has a number, not an opinion.

How to communicate this to business partners

Instead of saying "we need to be visible in AI," say "we're losing X PLN a month because our company doesn't appear in model answers, and customers are starting their purchases in AI chat." Instead of "we need a budget for GEO," say "the cost of acquiring one meeting from paid channels is Y PLN, and we're losing Z meetings a month, so recovering them through paid channels would cost W PLN." That's a language business partners understand.

The risk of absence from models

Three layers of risk

The risk of absence from models has three layers. The first is lost sales conversations, i.e., queries that never make it into the CRM because the customer didn't know you existed. The second is delayed revenue, i.e., missing contracts that should be signed in 3-9 months but won't be, because there are no leads today. The third is rising spend on paid channels, i.e., the need to compensate for lost queries with more expensive channels.

TCO of brand invisibility

The TCO of brand invisibility is the sum of lost revenue, lost margin, and the alternative cost of recovery. This isn't a marketing cost; it's a risk item that a CEO should enter into a decision spreadsheet next to salary cost and server cost. If a company doesn't calculate the TCO of invisibility, it will make budget decisions based on salespeople's opinions rather than on numbers.

Pipeline without AI queries

A pipeline without AI queries is a pipeline that will shrink every quarter, because an increasing share of customers start their purchases in AI chat. If your company doesn't appear in those conversations, it won't appear in the pipeline in six months. That's simple math you can calculate today.

The question to calculate

How many queries per month came from search a year ago, and how many come today?