Three AI budgets, no shared reckoning
Benedikt Langer
Advertorial · in collaboration with the Evernine Group 4 min read Almost every company can quantify ...
Read Article
Advertorial · in collaboration with the Evernine Group
5 min read
Alexander Roth and Alec Chizhik lead the Evernine Group. Their observation from client conversations: decision-makers are weary of AI because every recommendation ends in a major project. Their answer is three checks in traffic-light format that demonstrate within one to three weeks where sales, visibility and marketing leave money on the table. A conversation about data quality, AI search engines and the question of who benefits most internally from a hard number.
Key points in brief
- Three checks, one principle. The AI Sales Check, AI Visibility Check and AI Efficiency Check each assess one area in the traffic-light system, within one to three weeks, with a prioritised action list.
- The AI question is first a data question. Almost every CRM holds data that is never evaluated. Before automation comes the check of what is viable.
- What leadership understands gets funded. Number-based short analyses speak the language of investment committees. Lengthy reports do not.
Related:Why technology accounts for only 30 per cent of success in AI transformation / The hyperscalers’ multi-billion bet and your cloud bill
The frustration has become measurable
Mr Roth, according to McKinsey more than 80 per cent of companies have so far seen no measurable impact from their AI use on results. You say no one needs another eight-week AI assessment. Where does this bluntness come from?
Alexander Roth: From conversations over the past months. Managing directors, CFOs and marketing directors almost all tell the same story: lost budgets, initiatives that fizzle out, too many service providers with oversized concepts. Everyone arrives with a big topic and makes it complex. We work extensively with private equity firms. They care about one thing: the number. What was invested, what came back, where the leverage lies. We apply this discipline to the mid-market.
And one to three weeks suffice for that?
Roth: For the decision-making basis, yes. We do not promise transformation in three weeks. We deliver the reality check that enables leadership to decide which transformation will pay off. The result is a traffic-light system with an action list. Green is on track, amber has headroom, red costs money every month.
The first lever lies in the CRM
What does an AI Sales Check involve in practice, Mr Chizhik?
Alec Chizhik: A compact analysis of CRM architecture, data quality and sales processes with a view to AI readiness. We examine how clean the data foundation is, which processes can be automated and where information gets lost in the team. The result is three to five prioritised use cases, each with a calculation behind it. The whole exercise takes two to three weeks.
What do you find in the CRM systems?
Chizhik: Almost always the same picture: the system is paid for, the AI licences are paid for, yet the two are barely connected. The holdings contain market data, customer histories and signals that no one ever analyses. That is why the AI question is first a data question. Whoever automates before the foundation is in place is automating their errors.
The system is paid for, the licences are paid for. Yet the two are barely connected.
– Alec Chizhik, Leadership Evernine Group (2026)
If AI cites you, you exist
Gartner expects that classic search volume will decline by around 25 per cent because AI assistants answer questions directly. The second check measures visibility in AI search engines. Why is the classic ranking no longer enough?
Chizhik: Because a growing share of the buying journey begins in ChatGPT, Perplexity or Gemini. These systems deliver an answer, not a list of results. Anyone absent from the answer cedes it to the competition. The AI Visibility Check measures exactly that: for which questions from the target audience does the company appear, where is the competition cited, and which gaps can be closed with what effort. The result is available after one to two weeks.
Mr Roth, many providers offer this right now. What makes your version suitable for management?
Roth: The translation. A visibility report for the SEO department helps the SEO department. We structure the result so that a board member can take it to their marketing lead or a marketing director can take it to their agencies. Traffic lights, priorities, effort per measure. Those who work with agencies work better with them afterwards, because it finally becomes measurable what is being discussed.
And how does a company then become present in AI answers?
Chizhik: AI systems preferentially cite sources from established specialist publications. The gap is closed through exactly such environments. One tool for this is the GEO programme from Evernine Media, which places specialist content there and makes its visibility in AI answers measurable. The check reveals the gap, the programme closes it.
The third check breaks down the silos
That leaves the AI Efficiency Check for marketing. What does it examine?
Roth: Whether the business-critical communications disciplines work in a connected way. Whether AI carries information between departments, whether workflows are lean and whether reporting is reliable. In most organisations, marketing, sales and communications operate in silos with their own stacks. The check shows in traffic-light format where budget is being lost. It also presents a 90-day roadmap. The savings potential is regularly significant because duplicate tools and manual handovers simply cost money.
One objection: anyone selling a check wants to sell the implementation afterwards.
Roth: The action list is deliberately structured so that internal teams or existing service providers can implement it. If a client wants us for implementation afterwards, we are pleased. The check creates no obligation. That is exactly why it works as an entry point: small ticket, clear endpoint, reliable result.
What the decision-maker personally gains from it
Final question: who benefits most internally from such a check?
Roth: The person who commissions it. Whoever first establishes the need for action leads the discussion. A CIO or CMO enters the management meeting with a traffic-light analysis and moves from defending a budget to driving the initiative. They are seen as the initiator of the AI process in the organisation, with documented results. No one says this out loud, but for many it is the real reason to call.
Chizhik: And the entry point is deliberately low. An introductory call, 30 minutes, after which everyone knows whether a check is the right format. No workshop marathon beforehand.
Next step
An introductory call with Alexander Roth and Alec Chizhik: 30 minutes, no presentation, one question. Which of the three traffic lights is most likely on red for you.
Frequently asked questions
Which three checks does the Evernine Group offer?
The AI Sales Check (CRM, data quality, AI readiness in sales), the AI Visibility Check (presence in answers from AI search systems) and the AI Efficiency Check (marketing stack, silos, reporting). All three conclude with a traffic-light rating and a prioritised action list.
How long does a check take?
The Visibility Check is available after one to two weeks, the sales and efficiency checks after two to three weeks. The format requires no internal resources beyond individual interviews and data access.
Does the check oblige you to implement with Evernine?
No. The action list is structured so that internal teams or existing service providers can implement it. Collaboration beyond the check is possible, but not a requirement.
Read more on Digital Chiefs
Digital ChiefsThree AI budgets, no shared reckoningDigital ChiefsWhy More Tech Budget Doesn’t Secure a Competitive EdgeDigital ChiefsThe IT department decides whether the spin-off pays offMore from the MBF Media network
cloudmagazinIceberg won the format war. Now the catalogue counts mybusinessfutureInvestment backlog: How AI unlocks hidden budgets securitytodayWhen the reporting deadline clock really starts tickingImage source: AI-generated (July 2026)