WEBVTT

NOTE Timings are aligned to the minute markers in the source transcript.

00:00:00.000 --> 00:00:08.161
<v Jörn "Joe" Menninger>What happens to SaaS when custom software becomes cheap enough for small and medium-sized companies?</v>

00:00:08.231 --> 00:00:14.215
<v Jörn "Joe" Menninger>In part one, we talked with Delia Hallberg of Amadeni AI</v>

00:00:14.285 --> 00:00:18.637
<v Jörn "Joe" Menninger>about the operational reality inside Mittelstand SME companies.</v>

00:00:18.707 --> 00:00:24.692
<v Jörn "Joe" Menninger>Broken processes, fragmented tools, manual workarounds, and AI projects that only</v>

00:00:24.762 --> 00:00:29.658
<v Jörn "Joe" Menninger>create value when the underlying work is understood first.</v>

00:00:29.728 --> 00:00:35.168
<v Jörn "Joe" Menninger>In part two, we go deeper into the market tensions.</v>

00:00:35.238 --> 00:00:40.135
<v Jörn "Joe" Menninger>If an AI lowers the cost of building software,</v>

00:00:40.205 --> 00:00:46.189
<v Jörn "Joe" Menninger>does the future belong to better SaaS platforms or to custom</v>

00:00:46.259 --> 00:00:50.611
<v Jörn "Joe" Menninger>systems built around the way companies actually work?</v>

00:00:50.681 --> 00:00:52.858
<v Jörn "Joe" Menninger>Welcome back to Startuprad.io.</v>

00:00:52.928 --> 00:01:00.000
<v Jörn "Joe" Menninger>This is part two of our conversation with Delia Hallberg, co-founder and CEO</v>

00:01:00.000 --> 00:01:08.826
<v Jörn "Joe" Menninger>of Amadeni AI, a company building AI-supported custom software for small and medium-sized enterprises—the German Mittelstand.</v>

00:01:08.896 --> 00:01:16.067
<v Jörn "Joe" Menninger>In part one, we looked at the practical side, how Amadeni analyzes workflows,</v>

00:01:16.137 --> 00:01:22.205
<v Jörn "Joe" Menninger>where AI automation creates measurable relief, and why bad processes plus</v>

00:01:22.275 --> 00:01:26.137
<v Jörn "Joe" Menninger>AI can become bad processes on steroids.</v>

00:01:26.207 --> 00:01:30.068
<v Jörn "Joe" Menninger>Now we turn to the strategic question.</v>

00:01:30.138 --> 00:01:36.206
<v Jörn "Joe" Menninger>Delia argues that SaaS, as we know it is under pressure</v>

00:01:36.276 --> 00:01:40.137
<v Jörn "Joe" Menninger>because the cost of code has collapsed.</v>

00:01:40.207 --> 00:01:46.275
<v Jörn "Joe" Menninger>If custom software can be built faster, cheaper, and closer to</v>

00:01:46.345 --> 00:01:52.965
<v Jörn "Joe" Menninger>the actual operating model of an SME, then the old trade-off changes.</v>

00:01:53.035 --> 00:02:01.309
<v Jörn "Joe" Menninger>Companies may no longer have to choose between rigid, off-the-shelf tools and unaffordable custom development.</v>

00:02:01.379 --> 00:02:07.447
<v Jörn "Joe" Menninger>In this second part, we ask which SaaS products are vulnerable,</v>

00:02:07.517 --> 00:02:14.688
<v Jörn "Joe" Menninger>when SaaS still wins, how German SMEs think about data ownership and GDPR,</v>

00:02:14.758 --> 00:02:21.378
<v Jörn "Joe" Menninger>and why model-agnostic AI architecture may become a serious European business requirement.</v>

00:02:21.448 --> 00:02:27.516
<v Jörn "Joe" Menninger>Delia, before the break, we left listeners with one unresolved tension:</v>

00:02:27.586 --> 00:02:35.860
<v Jörn "Joe" Menninger>experienced developers still matter, but AI agents now produce more and more of the software.</v>

00:02:35.930 --> 00:02:41.998
<v Jörn "Joe" Menninger>Where does engineering craft end and agentic software in production begin?</v>

00:02:42.068 --> 00:02:43.171
<v Delia Hallberg>Hi, Jörn.</v>

00:02:43.241 --> 00:02:44.344
<v Delia Hallberg>Thank you.</v>

00:02:44.414 --> 00:02:47.724
<v Delia Hallberg>It is great to be back.</v>

00:02:47.794 --> 00:02:53.862
<v Delia Hallberg>I would say, the developer's role has changed into that of</v>

00:02:53.932 --> 00:03:00.000
<v Delia Hallberg>an IT department head, and your agents are your team of,</v>

00:03:00.000 --> 00:03:02.058
<v Delia Hallberg>programmers that you orchestrate</v>

00:03:02.128 --> 00:03:05.729
<v Jörn "Joe" Menninger>And they will never complain to HR.</v>

00:03:05.799 --> 00:03:06.828
<v Jörn "Joe" Menninger>I see.</v>

00:03:06.898 --> 00:03:12.556
<v Jörn "Joe" Menninger>Maybe there will be a department for agents in the future.</v>

00:03:12.626 --> 00:03:18.285
<v Jörn "Joe" Menninger>Delia, when you say SaaS as we know it will no</v>

00:03:18.355 --> 00:03:22.985
<v Jörn "Joe" Menninger>longer exist in its current form within a year,</v>

00:03:23.055 --> 00:03:27.171
<v Jörn "Joe" Menninger>what specifically do you expect to break first?</v>

00:03:27.241 --> 00:03:30.327
<v Jörn "Joe" Menninger>Pricing, implementation, lock-in, customization, or distribution?</v>

00:03:30.397 --> 00:03:35.027
<v Delia Hallberg>I think the SaaS market is very rapidly changing,</v>

00:03:35.097 --> 00:03:42.814
<v Delia Hallberg>and that's why we are becoming more relevant for a lot of Mittelstand and SMBs.</v>

00:03:42.884 --> 00:03:48.542
<v Delia Hallberg>Something that would have been completely beyond Mittelstand budgets five years</v>

00:03:48.612 --> 00:03:52.213
<v Delia Hallberg>ago is now a very plausible alternative.</v>

00:03:52.283 --> 00:04:00.000
<v Delia Hallberg>What I observe among my clients is that SaaS rigidity creates a lot of frustration.</v>

00:04:00.000 --> 00:04:05.163
<v Delia Hallberg>SaaS forces you to adapt your workflows to the software.</v>

00:04:05.233 --> 00:04:10.911
<v Delia Hallberg>Especially, for example, for niche SMBs, that trade-off has become very</v>

00:04:10.981 --> 00:04:15.628
<v Delia Hallberg>painful and has a structural impact on day-to-day operations.</v>

00:04:15.698 --> 00:04:22.925
<v Delia Hallberg>In the past, they had to live with it because there were no alternatives.</v>

00:04:22.995 --> 00:04:28.674
<v Delia Hallberg>CEOs and managing directors are also reading and following what is</v>

00:04:28.744 --> 00:04:34.423
<v Delia Hallberg>happening in the AI and IT world, and they are understanding</v>

00:04:34.493 --> 00:04:40.688
<v Delia Hallberg>that they are no longer locked into living with these process inefficiencies.</v>

00:04:40.758 --> 00:04:46.437
<v Delia Hallberg>I mentioned earlier the example of the engineering firm that had</v>

00:04:46.507 --> 00:04:52.186
<v Delia Hallberg>been asking its SaaS provider for over a year to implement</v>

00:04:52.256 --> 00:05:00.000
<v Delia Hallberg>one controlling feature which didn't get implemented and that was really the breaking point where</v>

00:05:00.000 --> 00:05:06.956
<v Delia Hallberg>they said, "Okay, we're going custom." A second thing that I'm</v>

00:05:07.026 --> 00:05:11.452
<v Delia Hallberg>observing with my clients is the pricing.</v>

00:05:11.522 --> 00:05:21.007
<v Delia Hallberg>Because let's be frank about it code costs have dropped, but SaaS licenses haven't moved.</v>

00:05:21.077 --> 00:05:28.033
<v Delia Hallberg>That is a mental divide that a lot of CEOs and</v>

00:05:28.103 --> 00:05:31.897
<v Delia Hallberg>managing directors are beginning to recognize.</v>

00:05:31.967 --> 00:05:39.555
<v Delia Hallberg>At the same time, the lifespan of tools has gotten incredibly short.</v>

00:05:39.625 --> 00:05:46.581
<v Delia Hallberg>SMBs are sometimes locked into contracts of 12 months or longer.</v>

00:05:46.651 --> 00:05:52.974
<v Delia Hallberg>They pay a high price while living with inefficient processes.</v>

00:05:53.044 --> 00:06:00.000
<v Delia Hallberg>If, on top of that you have no access to your</v>

00:06:00.000 --> 00:06:05.665
<v Delia Hallberg>data because of a non-existing API, that is a very clear</v>

00:06:05.735 --> 00:06:11.401
<v Delia Hallberg>case where our clients have usually said farewell to their SaaS.</v>

00:06:11.471 --> 00:06:18.167
<v Delia Hallberg>And I think that's putting a lot of pressure on the SaaS model.</v>

00:06:18.237 --> 00:06:22.357
<v Delia Hallberg>So just to add, horizontal tools will stay.</v>

00:06:22.427 --> 00:06:27.062
<v Delia Hallberg>Everybody needs an email service, everybody needs a browser.</v>

00:06:27.132 --> 00:06:28.677
<v Delia Hallberg>Those will remain.</v>

00:06:28.747 --> 00:06:34.928
<v Delia Hallberg>But the vertical, like highly industry-specific SaaS solutions are under serious pressure.</v>

00:06:34.998 --> 00:06:37.058
<v Delia Hallberg>That's what I observe</v>

00:06:37.128 --> 00:06:43.309
<v Jörn "Joe" Menninger>Yeah, there there are some areas of SaaS really under huge pressure,</v>

00:06:43.379 --> 00:06:50.074
<v Jörn "Joe" Menninger>and I was just talking to an entrepreneur who works in regulated SaaS.</v>

00:06:50.144 --> 00:06:55.810
<v Jörn "Joe" Menninger>He said he was not too concerned, but he would be</v>

00:06:55.880 --> 00:07:00.000
<v Jörn "Joe" Menninger>very concerned if he were in a non-regulated</v>

00:07:00.000 --> 00:07:00.644
<v Jörn "Joe" Menninger>area.</v>

00:07:00.714 --> 00:07:06.508
<v Jörn "Joe" Menninger>But let us return to the SaaS products here.</v>

00:07:06.578 --> 00:07:14.948
<v Jörn "Joe" Menninger>What kind of SaaS product is actually most vulnerable to AI-built custom software?</v>

00:07:15.018 --> 00:07:16.949
<v Jörn "Joe" Menninger>Is it something,</v>

00:07:17.019 --> 00:07:18.307
<v Jörn "Joe" Menninger>is old?</v>

00:07:18.377 --> 00:07:26.746
<v Jörn "Joe" Menninger>Is it something that is made by a vibe coder in a weekend?</v>

00:07:26.816 --> 00:07:29.391
<v Jörn "Joe" Menninger>Or what is it?</v>

00:07:29.461 --> 00:07:38.474
<v Delia Hallberg>So something built by a vibe coder over a weekend is not really SaaS.</v>

00:07:38.544 --> 00:07:40.476
<v Delia Hallberg>That's a prototype.</v>

00:07:40.546 --> 00:07:47.628
<v Delia Hallberg>I genuinely hope that nobody is using that for critical processes.</v>

00:07:47.698 --> 00:07:54.136
<v Delia Hallberg>If you do, we are always open to fixing that.</v>

00:07:54.206 --> 00:08:00.000
<v Delia Hallberg>SaaS solutions in heavily regulated industries are relatively safe.</v>

00:08:00.000 --> 00:08:06.016
<v Delia Hallberg>SaaS solutions that try to target a very broad group of customers,</v>

00:08:06.086 --> 00:08:11.601
<v Delia Hallberg>I would say these are the most vulnerable because they fit</v>

00:08:11.671 --> 00:08:15.181
<v Delia Hallberg>everyone and nobody at the same time.</v>

00:08:15.251 --> 00:08:19.763
<v Delia Hallberg>But then again, it comes to product management, right?</v>

00:08:19.833 --> 00:08:26.351
<v Delia Hallberg>And I would say the least vulnerable are the ones that are adaptable.</v>

00:08:26.421 --> 00:08:31.935
<v Delia Hallberg>That is not connected to age, but to how much the</v>

00:08:32.005 --> 00:08:35.014
<v Delia Hallberg>SaaS has been evolving with time.</v>

00:08:35.084 --> 00:08:40.598
<v Delia Hallberg>We have clients, for example, that have used their solution for</v>

00:08:40.668 --> 00:08:46.183
<v Delia Hallberg>eight years but the provider has gone through waves of transformation</v>

00:08:46.253 --> 00:08:49.262
<v Delia Hallberg>and remains close to its clients.</v>

00:08:49.332 --> 00:08:52.340
<v Delia Hallberg>They listen, evolve, have open interfaces.</v>

00:08:52.410 --> 00:08:55.919
<v Delia Hallberg>I would say these are generally fine.</v>

00:08:55.989 --> 00:09:00.000
<v Delia Hallberg>We have other clients whose prices increase monthly,</v>

00:09:00.000 --> 00:09:07.008
<v Delia Hallberg>and they have not received a new value-based feature in the past years.</v>

00:09:07.078 --> 00:09:10.852
<v Delia Hallberg>I would say that is more vulnerable.</v>

00:09:10.922 --> 00:09:16.852
<v Delia Hallberg>And I would say the ones that are least vulnerable are</v>

00:09:16.922 --> 00:09:23.930
<v Delia Hallberg>those that serve niches, that understand their clients and their problems very well,</v>

00:09:24.000 --> 00:09:26.695
<v Delia Hallberg>that serve that specific problem.</v>

00:09:26.765 --> 00:09:34.313
<v Delia Hallberg>So I would say the future is really the more you serve a niche,</v>

00:09:34.383 --> 00:09:40.313
<v Delia Hallberg>the safer you are, and the more adaptable and the more</v>

00:09:40.383 --> 00:09:47.391
<v Delia Hallberg>attentive you are to your market, to your customers, the safer you are.</v>

00:09:47.461 --> 00:09:49.078
<v Delia Hallberg>And obviously pricing.</v>

00:09:49.148 --> 00:09:52.922
<v Delia Hallberg>Acceptance of legacy pricing is declining significantly.</v>

00:09:52.992 --> 00:10:00.000
<v Delia Hallberg>But there are other ways where you can offer a fair solution to</v>

00:10:00.000 --> 00:10:01.291
<v Delia Hallberg>your clients.</v>

00:10:01.361 --> 00:10:02.006
<v Delia Hallberg>Yeah</v>

00:10:02.076 --> 00:10:10.465
<v Jörn "Joe" Menninger>We have talked about SaaS in general, but let us get more specific.</v>

00:10:10.535 --> 00:10:18.279
<v Jörn "Joe" Menninger>Where is it structurally superior to custom software, even with AI-driven development?</v>

00:10:18.349 --> 00:10:25.448
<v Delia Hallberg>I can tell you our carpenter clients generally do not have</v>

00:10:25.518 --> 00:10:29.390
<v Delia Hallberg>anything unusual about their business model.</v>

00:10:29.460 --> 00:10:35.913
<v Delia Hallberg>They are well served by solutions built for their trade.</v>

00:10:35.983 --> 00:10:43.727
<v Delia Hallberg>So I would say that wherever companies have a standard business model,</v>

00:10:43.797 --> 00:10:50.895
<v Delia Hallberg>where they have a good solution for their industry or for</v>

00:10:50.965 --> 00:11:00.000
<v Delia Hallberg>their specific business model, that is where it is important to go for the</v>

00:11:00.000 --> 00:11:01.030
<v Delia Hallberg>standard solution.</v>

00:11:01.100 --> 00:11:06.766
<v Delia Hallberg>I would definitely not develop a custom solution because obviously the</v>

00:11:06.836 --> 00:11:12.501
<v Delia Hallberg>provider has spent years working in a very specific industry and</v>

00:11:12.571 --> 00:11:18.752
<v Delia Hallberg>understands the overall problems of the clients, not only that specific one.</v>

00:11:18.822 --> 00:11:24.487
<v Delia Hallberg>They may have considered problems that we would not if we</v>

00:11:24.557 --> 00:11:27.647
<v Delia Hallberg>design for a very specific customer.</v>

00:11:27.717 --> 00:11:34.413
<v Delia Hallberg>It is also important to stress that custom software is an investment project,</v>

00:11:34.483 --> 00:11:42.209
<v Delia Hallberg>so it takes time and resources, but it is also a co-creation process over weeks.</v>

00:11:42.279 --> 00:11:47.944
<v Delia Hallberg>If I'm supposed or our team is supposed to create something</v>

00:11:48.014 --> 00:11:54.710
<v Delia Hallberg>that is really custom for you I need to know how you work,</v>

00:11:54.780 --> 00:11:59.415
<v Delia Hallberg>what is special about the way you're doing things.</v>

00:11:59.485 --> 00:12:00.000
<v Delia Hallberg>Why</v>

00:12:00.000 --> 00:12:03.731
<v Delia Hallberg>is an industry solution not working for you?</v>

00:12:03.801 --> 00:12:07.066
<v Delia Hallberg>And let's be frank, that takes time.</v>

00:12:07.136 --> 00:12:12.266
<v Delia Hallberg>That's why we always talk in ROI because we need a</v>

00:12:12.336 --> 00:12:15.600
<v Delia Hallberg>strong business case to undertake that process.</v>

00:12:15.670 --> 00:12:20.801
<v Delia Hallberg>And I would generally say, or the way I think about</v>

00:12:20.871 --> 00:12:27.866
<v Delia Hallberg>it is that SaaS and custom aren't competitors and in a good case, they're complements.</v>

00:12:27.936 --> 00:12:33.066
<v Delia Hallberg>My ideal situation is really where the SaaS solution, for example,</v>

00:12:33.136 --> 00:12:39.199
<v Delia Hallberg>does 80% or maybe 70% of the most important processes of the client,</v>

00:12:39.269 --> 00:12:44.866
<v Delia Hallberg>and then the client may have one specific process they handle differently.</v>

00:12:44.936 --> 00:12:50.066
<v Delia Hallberg>And if then the SaaS has an API or open interfaces,</v>

00:12:50.136 --> 00:12:55.733
<v Delia Hallberg>then we can create a small add-on to cover the last 30%.</v>

00:12:55.803 --> 00:13:00.000
<v Delia Hallberg>And that's really my favorite way of addressing projects.</v>

00:13:00.000 --> 00:13:06.018
<v Jörn "Joe" Menninger>you already hinted at my next question a bit.</v>

00:13:06.088 --> 00:13:14.780
<v Jörn "Joe" Menninger>If a Mittelstand company has a closed SaaS stack from many years ago,</v>

00:13:14.850 --> 00:13:24.211
<v Jörn "Joe" Menninger>what is the switching threshold where replacing it becomes less risky than keeping it?</v>

00:13:24.281 --> 00:13:31.637
<v Delia Hallberg>First, if workarounds have become load-bearing, if your office manager leaves</v>

00:13:31.707 --> 00:13:40.399
<v Delia Hallberg>and nobody knows how reporting works, well, that's a very big business risk.</v>

00:13:40.469 --> 00:13:48.493
<v Delia Hallberg>I also sometimes see that the system is blocking your next move.</v>

00:13:48.563 --> 00:13:53.243
<v Delia Hallberg>Let's talk about a very simple case.</v>

00:13:53.313 --> 00:14:00.000
<v Delia Hallberg>Langdock is, a GDPR compliant solution that has been growing</v>

00:14:00.000 --> 00:14:02.081
<v Delia Hallberg>quite strong in Germany.</v>

00:14:02.151 --> 00:14:08.393
<v Delia Hallberg>And I have a lot of clients coming to me saying, "Oh,</v>

00:14:08.463 --> 00:14:12.624
<v Delia Hallberg>I cannot integrate my custom software with Langdock.</v>

00:14:12.694 --> 00:14:18.936
<v Delia Hallberg>Can you help me build an API?" I have to say, "Sorry,</v>

00:14:19.006 --> 00:14:27.329
<v Delia Hallberg>I cannot build an API for your custom software." And they really want to use Langdock.</v>

00:14:27.399 --> 00:14:35.202
<v Delia Hallberg>They really want to create their own workflows, customize their agents, interact with their data.</v>

00:14:35.272 --> 00:14:40.474
<v Delia Hallberg>But as I already said previously, there is no API,</v>

00:14:40.544 --> 00:14:47.306
<v Delia Hallberg>there is no way to connect it to a solution such as Langdock.</v>

00:14:47.376 --> 00:14:50.497
<v Delia Hallberg>The pain and frustration are growing.</v>

00:14:50.567 --> 00:14:56.289
<v Delia Hallberg>What surprises me is when management continues paying for a product</v>

00:14:56.359 --> 00:15:00.000
<v Delia Hallberg>in which the vendor has stopped investing.</v>

00:15:00.000 --> 00:15:04.808
<v Delia Hallberg>Surprisingly, that is the reality for many industry solutions.</v>

00:15:04.878 --> 00:15:07.549
<v Delia Hallberg>Clients feel and see that.</v>

00:15:07.619 --> 00:15:10.290
<v Delia Hallberg>It is a simple equation.</v>

00:15:10.360 --> 00:15:17.840
<v Delia Hallberg>The switching threshold is when the cost of staying exceeds the cost of changing.</v>

00:15:17.910 --> 00:15:23.786
<v Delia Hallberg>By the time a company calls us, they have usually been</v>

00:15:23.856 --> 00:15:32.404
<v Delia Hallberg>past that point for a while, and they are very happy to finally have a solution</v>

00:15:32.474 --> 00:15:38.351
<v Jörn "Joe" Menninger>I was wondering, looking at your website, it promises software that</v>

00:15:38.421 --> 00:15:44.831
<v Jörn "Joe" Menninger>belongs to the client, data stored encrypted in Germany, and GDPR-compliant development.</v>

00:15:44.901 --> 00:15:49.710
<v Jörn "Joe" Menninger>How much of that is a real buying criterion,</v>

00:15:49.780 --> 00:15:56.190
<v Jörn "Joe" Menninger>and how much is still only verbal comfort for cautious German buyers?</v>

00:15:56.260 --> 00:16:00.000
<v Delia Hallberg>So if we would have had that</v>

00:16:00.000 --> 00:16:10.238
<v Delia Hallberg>conversation 2023 or would have said it's more a thing of comfort and maybe data-privacy-loving Germans.</v>

00:16:10.308 --> 00:16:17.347
<v Delia Hallberg>the world has really changed, and geopolitical developments are something that</v>

00:16:17.417 --> 00:16:22.536
<v Delia Hallberg>comes up in a lot of sales conversations.</v>

00:16:22.606 --> 00:16:32.845
<v Delia Hallberg>For example, now we have clients that say, "I want zero US services in my tooling.</v>

00:16:32.915 --> 00:16:39.953
<v Delia Hallberg>I want to be able from one day to the next,</v>

00:16:40.023 --> 00:16:47.062
<v Delia Hallberg>be completely independent of US solutions." That was something that came</v>

00:16:47.132 --> 00:16:54.811
<v Delia Hallberg>up a year ago, for example, when the Greenland conflict appeared imminent.</v>

00:16:54.881 --> 00:17:00.000
<v Delia Hallberg>I was saying, "Wow, that is good foresight."</v>

00:17:00.000 --> 00:17:04.765
<v Delia Hallberg>Then June happened one month ago, when, for example,</v>

00:17:04.835 --> 00:17:10.130
<v Delia Hallberg>the US government prevented new models from entering the market.</v>

00:17:10.200 --> 00:17:16.024
<v Delia Hallberg>That is an example of geopolitical decisions having an impact on</v>

00:17:16.094 --> 00:17:19.271
<v Delia Hallberg>the real tools that we use.</v>

00:17:19.341 --> 00:17:25.165
<v Delia Hallberg>Sovereignty is now a concern everywhere, and I would say that</v>

00:17:25.235 --> 00:17:30.529
<v Delia Hallberg>from the very beginning, were building our systems model agnostic,</v>

00:17:30.599 --> 00:17:36.424
<v Delia Hallberg>but given recent developments and because we do not know what</v>

00:17:36.494 --> 00:17:40.200
<v Delia Hallberg>the future will bring, it is essential.</v>

00:17:40.270 --> 00:17:47.153
<v Delia Hallberg>It is not only verbal comfort; in the end, it's about business continuity.</v>

00:17:47.223 --> 00:17:53.047
<v Jörn "Joe" Menninger>I have to tell you from all the conversations I have</v>

00:17:53.117 --> 00:18:00.000
<v Jörn "Joe" Menninger>had about the US government deciding not to deliver certain models, during a</v>

00:18:00.000 --> 00:18:05.235
<v Jörn "Joe" Menninger>the restrictions specifically affected non-US users,</v>

00:18:05.305 --> 00:18:13.682
<v Jörn "Joe" Menninger>that sent ripples through the software world that we will still feel a decade from now.</v>

00:18:13.752 --> 00:18:18.464
<v Jörn "Joe" Menninger>I think a lot of people start to rethink,</v>

00:18:18.534 --> 00:18:21.675
<v Jörn "Joe" Menninger>which models their systems rely on.</v>

00:18:21.745 --> 00:18:25.933
<v Jörn "Joe" Menninger>But leave the geopolitics a little bit aside.</v>

00:18:26.003 --> 00:18:29.145
<v Jörn "Joe" Menninger>You already talked about model-agnostic architectures.</v>

00:18:29.215 --> 00:18:34.974
<v Jörn "Joe" Menninger>What would have to happen, maybe even geopolitically or commercially for</v>

00:18:35.044 --> 00:18:39.755
<v Jörn "Joe" Menninger>a client to actually need to swap models quickly?</v>

00:18:39.825 --> 00:18:40.349
<v Delia Hallberg>June.</v>

00:18:40.419 --> 00:18:42.513
<v Delia Hallberg>That is my answer.</v>

00:18:42.583 --> 00:18:44.677
<v Delia Hallberg>June had to happen.</v>

00:18:44.747 --> 00:18:52.600
<v Delia Hallberg>Yeah, I think, one security report, one government letter, and a model can be gone.</v>

00:18:52.670 --> 00:19:00.000
<v Delia Hallberg>And I mean, in that case, were lucky that it was a new model,</v>

00:19:00.000 --> 00:19:04.977
<v Delia Hallberg>So operations could not yet have been running on it.</v>

00:19:05.047 --> 00:19:10.025
<v Delia Hallberg>But imagine a model being withdrawn that is already deployed.</v>

00:19:10.095 --> 00:19:10.592
<v Delia Hallberg>Right?</v>

00:19:10.662 --> 00:19:17.133
<v Delia Hallberg>So that's a geopolitical trigger, but then also commercial reasons, like price hikes.</v>

00:19:17.203 --> 00:19:22.678
<v Delia Hallberg>I don't know where the future of AI pricing is going,</v>

00:19:22.748 --> 00:19:29.218
<v Delia Hallberg>and I don't think that it's going to stay cheap as it is.</v>

00:19:29.288 --> 00:19:32.275
<v Delia Hallberg>So that's also a strong trigger.</v>

00:19:32.345 --> 00:19:37.820
<v Delia Hallberg>It is also very important that we try to use as</v>

00:19:37.890 --> 00:19:45.356
<v Delia Hallberg>many European solutions as we can, obviously, when it fits the purpose and the quality.</v>

00:19:45.426 --> 00:19:50.901
<v Delia Hallberg>but it is also important not to have a monopoly and</v>

00:19:50.971 --> 00:19:55.450
<v Delia Hallberg>to always have alternatives to which you can switch,</v>

00:19:55.520 --> 00:20:00.000
<v Delia Hallberg>and obviously also then to have the architecture to</v>

00:20:00.000 --> 00:20:06.741
<v Delia Hallberg>be able to switch models in a model-agnostic way within hours</v>

00:20:06.811 --> 00:20:15.390
<v Jörn "Joe" Menninger>One of the interesting things is that you have partnered with organizational development experts,</v>

00:20:15.460 --> 00:20:22.200
<v Jörn "Joe" Menninger>and your pitch says some AI projects shake role definitions so</v>

00:20:22.270 --> 00:20:25.947
<v Jörn "Joe" Menninger>deeply that management conflicts can surface.</v>

00:20:26.017 --> 00:20:34.596
<v Jörn "Joe" Menninger>What is the operating rule for when you bring organizational development into the room?</v>

00:20:34.666 --> 00:20:41.406
<v Delia Hallberg>I like to think of these large transformation projects where we</v>

00:20:41.476 --> 00:20:48.217
<v Delia Hallberg>aim at an efficiency increase of perhaps 50 or 60 percent,</v>

00:20:48.287 --> 00:20:53.189
<v Delia Hallberg>not as IT projects, but as organizational projects.</v>

00:20:53.259 --> 00:21:00.000
<v Delia Hallberg>If perhaps half of the team suddenly has</v>

00:21:00.000 --> 00:21:06.918
<v Delia Hallberg>time for new then obviously you're affecting the very root of how people work,</v>

00:21:06.988 --> 00:21:12.918
<v Delia Hallberg>how they define their work, what they're doing on a daily basis.</v>

00:21:12.988 --> 00:21:18.424
<v Delia Hallberg>And usually we bring an organizational-development expert into every project or</v>

00:21:18.494 --> 00:21:22.942
<v Delia Hallberg>a change expert just in case something comes up.</v>

00:21:23.012 --> 00:21:29.930
<v Delia Hallberg>We have had projects where we've introduced them at the beginning of the project.</v>

00:21:30.000 --> 00:21:33.459
<v Delia Hallberg>The team was very clear, very structured.</v>

00:21:33.529 --> 00:21:41.436
<v Delia Hallberg>It had good leadership, and then we never needed the change experts, which is totally fine.</v>

00:21:41.506 --> 00:21:45.953
<v Delia Hallberg>But we also had projects where after two months,</v>

00:21:46.023 --> 00:21:50.965
<v Delia Hallberg>we received a call from the managing director saying, "Yeah,</v>

00:21:51.035 --> 00:21:56.471
<v Delia Hallberg>we're thinking about splitting up." In that case, tensions among three</v>

00:21:56.541 --> 00:22:00.000
<v Delia Hallberg>managing directors had exploded because through that</v>

00:22:00.000 --> 00:22:04.261
<v Delia Hallberg>really transformation project, it became clear who was doing what,</v>

00:22:04.331 --> 00:22:09.018
<v Delia Hallberg>who was responsible for what, and then people had different understandings</v>

00:22:09.088 --> 00:22:13.776
<v Delia Hallberg>of that where I was very happy to have already introduced</v>

00:22:13.846 --> 00:22:18.533
<v Delia Hallberg>a change expert who already knew the customers, who had done</v>

00:22:18.603 --> 00:22:24.569
<v Delia Hallberg>a little bit of work, who could immediately step in and mediate that conflict.</v>

00:22:24.639 --> 00:22:29.752
<v Delia Hallberg>Because management teams splitting up is a huge risk for our project,</v>

00:22:29.822 --> 00:22:33.657
<v Delia Hallberg>but were able to continue it within one week.</v>

00:22:33.727 --> 00:22:39.266
<v Delia Hallberg>So to answer your question, we always include it from the start and,</v>

00:22:39.336 --> 00:22:46.154
<v Delia Hallberg>if we don't make use of it it's fine, but it's good, to have them there.</v>

00:22:46.224 --> 00:22:50.912
<v Jörn "Joe" Menninger>I would also argue if you have three managing directors at</v>

00:22:50.982 --> 00:22:55.669
<v Jörn "Joe" Menninger>a customer and they split up, it is also a huge</v>

00:22:55.739 --> 00:23:00.000
<v Jörn "Joe" Menninger>risk for the company itself, not only for the project.</v>

00:23:00.000 --> 00:23:09.670
<v Jörn "Joe" Menninger>But let's talk a little bit about leadership ownership of AI transformation projects inside an SME.</v>

00:23:09.740 --> 00:23:18.806
<v Jörn "Joe" Menninger>Who should ultimately own and be responsible for this AI transformation project inside an SME?</v>

00:23:18.876 --> 00:23:25.524
<v Jörn "Joe" Menninger>Should it be the CEO, operations, IT, finance, or the frontline</v>

00:23:25.594 --> 00:23:29.221
<v Jörn "Joe" Menninger>team that actually feels the pain?</v>

00:23:29.291 --> 00:23:30.499
<v Delia Hallberg>The CEO.</v>

00:23:30.569 --> 00:23:32.987
<v Delia Hallberg>A very straightforward answer.</v>

00:23:33.057 --> 00:23:39.705
<v Delia Hallberg>No, really, it is the CEO because we're talking about SMEs.</v>

00:23:39.775 --> 00:23:44.610
<v Delia Hallberg>These are organizations with 20 to 250 employees.</v>

00:23:44.680 --> 00:23:47.098
<v Delia Hallberg>That's definitely the CEO.</v>

00:23:47.168 --> 00:23:53.816
<v Delia Hallberg>For major AI transformation processes, I am not talking about some</v>

00:23:53.886 --> 00:23:57.512
<v Delia Hallberg>minor side process that we're automating.</v>

00:23:57.582 --> 00:24:00.000
<v Delia Hallberg>I'm talking about the</v>

00:24:00.000 --> 00:24:01.709
<v Delia Hallberg>big transformation process.</v>

00:24:01.779 --> 00:24:06.904
<v Delia Hallberg>It touches processes, roles, budgets, and sometimes even identity.</v>

00:24:06.974 --> 00:24:09.822
<v Delia Hallberg>It can be very fundamental.</v>

00:24:09.892 --> 00:24:15.017
<v Delia Hallberg>So this has to be owned by the CEO,</v>

00:24:15.087 --> 00:24:21.352
<v Delia Hallberg>and really our best projects have strong ownership among C-level managers.</v>

00:24:21.422 --> 00:24:27.687
<v Delia Hallberg>But the frontline team obviously has to be in the room,</v>

00:24:27.757 --> 00:24:31.744
<v Delia Hallberg>and they have to participate in it.</v>

00:24:31.814 --> 00:24:38.078
<v Delia Hallberg>They are the ones who know how the work actually happens.</v>

00:24:38.148 --> 00:24:44.413
<v Delia Hallberg>So we always bring the team together, and as I said,</v>

00:24:44.483 --> 00:24:52.456
<v Delia Hallberg>the best projects combine strong CEO ownership of decisions with frontline-team ownership of feedback.</v>

00:24:52.526 --> 00:24:57.652
<v Delia Hallberg>That's the combination that we need for successful projects</v>

00:24:57.722 --> 00:25:00.000
<v Jörn "Joe" Menninger>The CEO should own</v>

00:25:00.000 --> 00:25:06.297
<v Jörn "Joe" Menninger>the project, but how centralized should AI decision-making be in a company,</v>

00:25:06.367 --> 00:25:12.665
<v Jörn "Joe" Menninger>let's say a 200-person Mittelstand startup company before it becomes a bottleneck?</v>

00:25:12.735 --> 00:25:16.408
<v Delia Hallberg>So I think you have two types.</v>

00:25:16.478 --> 00:25:19.627
<v Delia Hallberg>You have these major transformation projects.</v>

00:25:19.697 --> 00:25:25.469
<v Delia Hallberg>Obviously that's the one that I just elaborated, but you also</v>

00:25:25.539 --> 00:25:32.362
<v Delia Hallberg>have a different type of AI adoption that is, for example, using agents,</v>

00:25:32.432 --> 00:25:37.679
<v Delia Hallberg>using models to automate small parts of your daily work.</v>

00:25:37.749 --> 00:25:42.472
<v Delia Hallberg>Here, I would say strategy has to be centralized,</v>

00:25:42.542 --> 00:25:48.315
<v Delia Hallberg>but usage should be decentralized because the C-level decides where to</v>

00:25:48.385 --> 00:25:52.583
<v Delia Hallberg>invest and sets the data policy and governance.</v>

00:25:52.653 --> 00:26:00.000
<v Delia Hallberg>It is very important to have guidelines, but I would leave the daily AI</v>

00:26:00.000 --> 00:26:02.113
<v Delia Hallberg>usage within these guardrails.</v>

00:26:02.183 --> 00:26:05.880
<v Delia Hallberg>This should be as decentralized as possible.</v>

00:26:05.950 --> 00:26:10.176
<v Delia Hallberg>Set the guardrails, then let the people run</v>

00:26:10.246 --> 00:26:16.056
<v Jörn "Joe" Menninger>Hopefully you did not forget to set a very important guardrail</v>

00:26:16.126 --> 00:26:21.937
<v Jörn "Joe" Menninger>because it's always easy to say, "Ah, let's do this and</v>

00:26:22.007 --> 00:26:28.874
<v Jörn "Joe" Menninger>this and this." But asking whether you have thought about every possible scenario,</v>

00:26:28.944 --> 00:26:33.169
<v Jörn "Joe" Menninger>is challenging, especially when you have creative interns.</v>

00:26:33.239 --> 00:26:39.050
<v Jörn "Joe" Menninger>What is one operational belief that you had at the start</v>

00:26:39.120 --> 00:26:44.402
<v Jörn "Joe" Menninger>of Amadeni that a real client project has since disproven?</v>

00:26:44.472 --> 00:26:46.585
<v Delia Hallberg>That's a good question.</v>

00:26:46.655 --> 00:26:52.465
<v Delia Hallberg>So I think we started with a belief that there would</v>

00:26:52.535 --> 00:26:55.704
<v Delia Hallberg>be a lot of nice-to-have projects.</v>

00:26:55.774 --> 00:27:00.000
<v Delia Hallberg>What I just mentioned, small side process that's</v>

00:27:00.000 --> 00:27:07.043
<v Delia Hallberg>been inconvenient, let's put it that way, and that automating them would make good projects.</v>

00:27:07.113 --> 00:27:11.339
<v Delia Hallberg>That's something that we've learned is not the case.</v>

00:27:11.409 --> 00:27:17.043
<v Delia Hallberg>The best projects are the ones where people genuinely feel the pain,</v>

00:27:17.113 --> 00:27:23.217
<v Delia Hallberg>where we do not have to convince a team that digitalization is important.</v>

00:27:23.287 --> 00:27:28.451
<v Delia Hallberg>If I have to start this conversation, it's already a red</v>

00:27:28.521 --> 00:27:35.564
<v Delia Hallberg>flag and I totally accept that they have to go through a few learning curves.</v>

00:27:35.634 --> 00:27:40.799
<v Delia Hallberg>The best projects are the ones where the pain is real.</v>

00:27:40.869 --> 00:27:43.217
<v Delia Hallberg>The team has felt it.</v>

00:27:43.287 --> 00:27:48.451
<v Delia Hallberg>They have already looked for alternatives, haven't been able to find</v>

00:27:48.521 --> 00:27:54.156
<v Delia Hallberg>it and when they find us, it's a match made in heaven.</v>

00:27:54.226 --> 00:27:56.573
<v Delia Hallberg>That is something we learned.</v>

00:27:56.643 --> 00:27:57.113
<v Jörn "Joe" Menninger>Great.</v>

00:27:57.183 --> 00:28:00.000
<v Jörn "Joe" Menninger>Now let us close with the</v>

00:28:00.000 --> 00:28:08.940
<v Jörn "Joe" Menninger>two typical questions here at Startuprad.io to learn a little bit about you and your company.</v>

00:28:09.010 --> 00:28:16.274
<v Jörn "Joe" Menninger>Question number one, of course, are you open to talking to new investors?</v>

00:28:16.344 --> 00:28:20.815
<v Delia Hallberg>For Amadeni AI itself, we are not raising.</v>

00:28:20.885 --> 00:28:23.120
<v Delia Hallberg>We are intentionally bootstrapped.</v>

00:28:23.190 --> 00:28:29.336
<v Delia Hallberg>What happens in our projects is that solutions sometimes emerge that</v>

00:28:29.406 --> 00:28:33.318
<v Delia Hallberg>have the potential to become standalone products.</v>

00:28:33.388 --> 00:28:41.210
<v Delia Hallberg>If it is a promising market, a promising product, then we spin those out.</v>

00:28:41.280 --> 00:28:46.868
<v Delia Hallberg>And for those, we are open to talking to investors.</v>

00:28:46.938 --> 00:28:53.085
<v Delia Hallberg>If somebody finds our philosophy interesting, it's definitely worth reaching out</v>

00:28:53.155 --> 00:28:56.507
<v Delia Hallberg>and becoming part of our network</v>

00:28:56.577 --> 00:28:59.371
<v Jörn "Joe" Menninger>Speaking of networks and people.</v>

00:28:59.441 --> 00:29:00.000
<v Jörn "Joe" Menninger>Are</v>

00:29:00.000 --> 00:29:03.601
<v Jörn "Joe" Menninger>you looking for talented people right now?</v>

00:29:03.671 --> 00:29:09.844
<v Delia Hallberg>With a large set of AI agents doing a lot of work,</v>

00:29:09.914 --> 00:29:14.544
<v Delia Hallberg>we are reaching a natural bottleneck, I would say.</v>

00:29:14.614 --> 00:29:18.730
<v Delia Hallberg>We are currently in a selective growth phase.</v>

00:29:18.800 --> 00:29:24.458
<v Delia Hallberg>We are looking for consultants with great personalities, a deep understanding</v>

00:29:24.528 --> 00:29:31.731
<v Delia Hallberg>of the SMB landscape, sound project and product-management skills, and a solid technical understanding.</v>

00:29:31.801 --> 00:29:32.315
<v Delia Hallberg>Yes.</v>

00:29:32.385 --> 00:29:34.957
<v Jörn "Joe" Menninger>Thank you very much, Delia.</v>

00:29:35.027 --> 00:29:40.171
<v Jörn "Joe" Menninger>It was a pleasure talking to you in two interviews.</v>

00:29:40.241 --> 00:29:45.900
<v Jörn "Joe" Menninger>Actually, as we can tell our audience, we recorded four pieces.</v>

00:29:45.970 --> 00:29:53.172
<v Jörn "Joe" Menninger>Last time, you had a bad cough, and we decided to completely redo everything.</v>

00:29:53.242 --> 00:29:56.843
<v Jörn "Joe" Menninger>We have now effectively made four episodes.</v>

00:29:56.913 --> 00:30:00.000
<v Jörn "Joe" Menninger>So thank you very much for</v>

00:30:00.000 --> 00:30:01.643
<v Jörn "Joe" Menninger>sticking with me.</v>

00:30:01.713 --> 00:30:07.739
<v Jörn "Joe" Menninger>To everyone watching on video, I am sorry, it's over 32</v>

00:30:07.809 --> 00:30:13.834
<v Jörn "Joe" Menninger>degrees Celsius here in my room and the lights don't help.</v>

00:30:13.904 --> 00:30:19.930
<v Jörn "Joe" Menninger>But for our audience, if an SME owner remembers one decision</v>

00:30:20.000 --> 00:30:27.121
<v Jörn "Joe" Menninger>rule from this conversation, what should they ask before spending money on AI?</v>

00:30:27.191 --> 00:30:33.217
<v Jörn "Joe" Menninger>Should they ask, can we automate this, or do we understand</v>

00:30:33.287 --> 00:30:36.573
<v Jörn "Joe" Menninger>how this work should actually function?</v>

00:30:36.643 --> 00:30:40.478
<v Delia Hallberg>Well, I would always choose the second.</v>

00:30:40.548 --> 00:30:44.930
<v Delia Hallberg>"Can we automate this?" is a technology question.</v>

00:30:45.000 --> 00:30:48.834
<v Delia Hallberg>And the answer is almost always yes.</v>

00:30:48.904 --> 00:30:52.191
<v Delia Hallberg>You can automate almost anything today.</v>

00:30:52.261 --> 00:30:58.287
<v Delia Hallberg>But the real question is: do we understand how this works?</v>

00:30:58.357 --> 00:31:00.000
<v Delia Hallberg>It is process</v>

00:31:00.000 --> 00:31:01.929
<v Delia Hallberg>first.</v>

00:31:01.999 --> 00:31:07.787
<v Delia Hallberg>Do we understand how this work should actually function?</v>

00:31:07.857 --> 00:31:10.429
<v Delia Hallberg>That's the important question.</v>

00:31:10.499 --> 00:31:13.071
<v Delia Hallberg>Can we automate this?</v>

00:31:13.141 --> 00:31:18.285
<v Delia Hallberg>Well, it's almost always a yes, especially nowadays.</v>

00:31:18.355 --> 00:31:24.786
<v Delia Hallberg>But don't use process first as an excuse to wait.</v>

00:31:24.856 --> 00:31:26.142
<v Delia Hallberg>Start today.</v>

00:31:26.212 --> 00:31:32.642
<v Delia Hallberg>Look at where your team loses time every single day.</v>

00:31:32.712 --> 00:31:37.857
<v Delia Hallberg>That's the money you're leaving on the table.</v>

00:31:37.927 --> 00:31:46.929
<v Delia Hallberg>The companies that are looking at their processes and automating now are pulling ahead.</v>

00:31:46.999 --> 00:31:54.073
<v Delia Hallberg>In today's competitive landscape, most SMBs can't afford to stand still</v>

00:31:54.143 --> 00:31:54.786
<v Jörn "Joe" Menninger>Great.</v>

00:31:54.856 --> 00:32:00.000
<v Jörn "Joe" Menninger>Last word: Most SMEs, most startups, most companies</v>

00:32:00.000 --> 00:32:05.391
<v Jörn "Joe" Menninger>can't afford to stand still.</v>

00:32:05.461 --> 00:32:09.774
<v Jörn "Joe" Menninger>Thank you very much</v>

00:32:09.844 --> 00:32:12.000
<v Delia Hallberg>Thank you</v>
