What executive content actually looks like when it’s done well.

CAC is the most overrated metric in DTC and I will die on this hill.

Let me tell you why.

Year two of the business, we got obsessed with bringing our blended CAC down. Pulled it from $47 to $31. I remember showing the board the slide. Felt like we had actually figured something out.

Revenue was flat.

Took me months to work out why. The cheaper customers were one and done. They bought once at a discount, we never saw them again. The $47 customers, the ones we had basically stopped acquiring because they looked expensive, had a 3.2x repeat rate. They were funding the whole business.

We optimized ourselves into a worse company.

The real question is never what your CAC is. Its what the LTV to CAC ratio looks like cohort by cohort, channel by channel. Harder to screenshot for a board deck. Actually tells you if youre building something.

Curious if anyone else has fallen into this one. Feels like a rite of passage.

Last year we returned $340K in inventory to our manufacturer. Not defective. Not a quality issue. I had just made a stupid decision and we had to eat it.

Here is what happened. We had a product line doing really well. Our merch lead came to me and said we should do colorways. I said yes without thinking it through. We ordered 8 new SKUs in 6 colors each. That is 48 new variants added to the catalog overnight.

What I did not do was run the math. On warehouse slots. On forecast accuracy at the SKU level. On whether any of these variants would sell at a rate that made the inventory cost worth it.

Six months in, we were sitting on dead stock. Sage green throw pillows in the oversized size nobody bought. Warehouse fees eating us alive. Cash locked in boxes I could not move.

Things I wish I had known before I said yes to the colorways.

Every new SKU is a bet. Treat it like capital allocation, not merchandising.

Forecast accuracy gets unreliable fast past your top 20 SKUs. Below that you are basically guessing.

We are just expanding the line is an expensive sentence in DTC. Nobody in the room will question it because it sounds like growth.

If a new SKU is not selling through at 60% of the parent rate within 90 days, kill it. Dont wait to see if it picks up. It wont.

We run a SKU profitability review every quarter now. Most expensive lesson I have paid for. Sharing in case it saves someone else a call to their manufacturer.

On Tuesday I sat in my car in the warehouse parking lot and cried for about ten minutes.

Nothing bad happened. That is the weird part.

We had just shipped our 100,000th order. The team was inside celebrating. There was a cake with a really sweet misspelling on it. I had given a little speech, everyone clapped, it was lovely. I walked out to grab my phone charger from the car and just lost it.

Four years ago I was packing the first 50 orders on my kitchen floor. My husband was there with packing tape and a label printer that kept jamming. I was pretty sure the business was going to fail and we were going to lose the savings we put in to start it. I remember being scared I had talked him into something that was going to hurt us.

And now there is a warehouse. There are 40 people who depend on this place for their paychecks. And a 100,000th order that shipped while I was making coffee this morning.

Nobody really tells you about this part of building a company. That you will grieve the earlier versions of it even while you are proud of what it has become. The kitchen floor was brutal. It was also probably the most creative I will ever get to be.

I do not miss it exactly. I miss the person I was when I was doing it. The one who did not yet know if any of this was going to work.

If you are in your kitchen floor phase right now, I know it is terrifying. I promise you will miss it more than you expect to.

Going back inside now. There is cake.

RISE with SAP is not a migration strategy. It is a commercial repackaging. These are different things and the industry keeps pretending they are the same.

I have sat through probably 40 SAP customer briefings in the last 18 months where the sales team walks in with a RISE deck and the CIO walks out thinking they now have a plan for S/4HANA. They do not. They have a licensing model.

Here is what RISE actually solves. It simplifies the commercial relationship. One contract instead of six. One throat to choke. Predictable OpEx. These are real benefits for procurement and finance. I will not pretend they are not.

Here is what RISE does not solve. Your custom code. Your integrations. Your data model. Your business process gaps. Your change management. In other words, the actual work of a migration. SAP is happy to sell you RISE. SAP is not going to do the migration for you.

I have watched three clients in the last two years sign RISE agreements thinking the hard part was over. In each case they called us 9 months later asking why their Brownfield conversion was stuck. The answer was always the same. You bought a contract, not a roadmap.

If you are a CIO evaluating RISE right now, the question is not whether to sign. It might be the right commercial structure for you. The question is what your actual migration plan is independent of the licensing decision. Those are two separate conversations and you need to have them in that order.

Happy to be told I am wrong about this. But I have been in the room for enough of these to know what I am seeing.

In 2019 we walked away from a $4.2M S/4HANA implementation six weeks before go-live. It is still the best business decision I have made in 13 years of running this firm.

The client was a mid-market manufacturer. Good people. Smart CFO. We had been on the project for 11 months and by any reasonable measure things were going well. Functional testing was largely complete. Data migration scripts were built. Training decks were drafted.

But something was off. The business process workshops that were supposed to happen in month 2 had never really happened. The client kept rescheduling them. We kept building against the original scoping assumptions. And in month 10 I started hearing from our functional leads that the numbers in the test environment did not look like the numbers the CFO was quoting in board meetings.

We ran a reconciliation. The gap was $6M in revenue recognition that the new system handled correctly and their legacy system had been handling wrong for years. If we went live, the first quarterly report would flag it and the audit would be ugly.

I flew out and met with the CEO and CFO in person. Told them what we had found and what I recommended. Stop the implementation. Fix the data lineage issue in the legacy system first. Re-scope the project. Restart in 60 days.

The CFO was not happy. My CFO was less happy. We refunded roughly $800K in fees that had not yet been earned in a way that reflected the delay. My partners thought I was insane.

What I took away from it.

Your most valuable currency in enterprise consulting is your reputation for walking away. Every partner can tell you why their methodology is great. Very few can show you a project they killed when it needed to be killed.

Clients forget the cost overruns. They do not forget the partner who told them the truth when it was expensive to do so.

That manufacturer came back to us 14 months later, ran the project properly, and has been a client for six years. Three of their board members have hired us for their companies since.

The economics of honesty in this business are better than anyone thinks. It just takes a while to see them show up on the P&L.

My father worked at the same auto parts plant in Dearborn for 38 years. He was a foreman. Ran the overnight shift for the last 12 of those years. I used to bring him dinner in a tupperware when my mother was working late.

He did not understand what I did for a living. He understood that I traveled a lot. He understood that I wore a suit. He was proud of it in the way parents are proud of things they cannot picture.

The last real conversation we had was about three weeks before he passed. I was home for a weekend and we were sitting on his porch. He asked me what my company actually does. I had tried to explain it before and never found a way to make it land. This time I said we help big companies replace the old software that runs their factories and their back offices with newer software that does the same thing better.

He thought about it for a minute. Then he said, so you fix the stuff that breaks. I said yeah pretty much. He nodded and said thats a good job. Someone has to do that.

I did not realize until after he was gone that he had given me the best description of the work I had ever heard.

We talk about digital transformation and enterprise architecture and business process reengineering. We build slides with capability maps. We charge a lot of money to say things that sound important.

But mostly what we do is fix the stuff that breaks. Large companies run on systems that were designed in the 90s by people who have retired. Those systems are breaking in slow motion and someone has to go in and replace them without the business falling over in the process. Thats the job.

My father spent 38 years making sure an auto parts plant ran through the night. I spend my time making sure companies run through their transformations. When I put it like that, what we do is not that different.

I think about him a lot on the hard projects.

I am going to say something that will not make me popular at vendor dinners. The AI strategy your CIO is being asked to present to the board next quarter is mostly theater.

Not because AI does not matter. It does. But because what boards are asking for right now is not a strategy. It is a document that looks like a strategy. There is a meaningful difference.

A real AI strategy for a mid-market industrial company in 2026 would answer questions like these. Which three business processes generate enough structured data that we could actually train a model on them. What is our data governance posture and can it survive a regulatory audit if we start feeding customer information into an LLM. Who owns model risk when something goes wrong in production. How do we staff for this when we cannot compete with FAANG salaries.

The documents I am seeing CIOs present to boards do not answer those questions. They describe a vendor landscape. They list use cases from McKinsey reports. They include a slide titled our AI transformation journey with a curved arrow on it.

The board accepts the document. The CIO gets a budget line. Six months later something gets deployed that looks like AI and the budget line becomes a sunk cost nobody wants to revisit.

This is not the CIOs fault. The honest answer to the board is often we do not know yet, we need 90 days to figure out what is real here. Boards do not reward that answer and CIOs who give it get replaced by CIOs who do not.

But we should at least be honest with each other about what is happening. A lot of what is being called AI strategy is actually procurement theater. The real AI work, when it gets done well, looks boring. It is data hygiene. It is governance. It is waiting to adopt until the tooling stabilizes. It is not a curved arrow.

Curious if other CIOs are seeing this. Happy to be wrong.

In 2022 we had a ransomware incident that cost us roughly $3.4M before we were fully recovered. I have not written about it publicly until now. Our legal team preferred I not. I am going to share what I learned because I think it will help other CIOs who have not been through this.

The attack came in through a compromised vendor account. Not a phishing email. Not an unpatched server. A service account belonging to a third party logistics provider who had VPN access to our warehouse management system for the last seven years. Their credentials were leaked in an unrelated breach and used against us roughly 72 hours later.

We had reasonable security controls by any standard framework. Multi factor authentication on employee accounts. EDR on endpoints. Segmented networks. Tabletop exercises every six months. What we did not have was any meaningful program for managing third party access.

Here is what I wish every CIO knew before it happens to them.

The first 48 hours are not a technical problem, they are a communication problem. Our IT team knew what to do. What we were not prepared for was the volume of decisions that had to be made simultaneously by the CEO, the CFO, legal, the board, insurance, and external counsel. Every one of them had different information needs. Nobody was coordinating them. That is where incidents actually go off the rails.

Your cyber insurance policy is not a plan. I had read ours. I thought I understood it. Under pressure I learned that a significant portion of our loss was not covered because of exclusions we had never meaningfully reviewed. Read yours with your broker this month. Ask them specifically about business interruption, ransom payment restrictions, and regulatory fines. Do not wait.

Third party access is where most mid-market breaches are going to originate for the next several years. Everyone is looking at their own perimeter. Almost nobody has a real inventory of which vendors have active credentials in their environment, what those credentials can do, and when they were last rotated. Build that inventory.

The CEO relationship matters more than the technical response. Our CEO at the time trusted me. That trust was built over three years before the incident. It is what allowed us to make the right calls fast. If you are a new CIO, the worst time to introduce yourself to your CEO is in the middle of a breach.

We rebuilt our third party access program from scratch over the following 18 months. It is the work I am most proud of from that period. Happy to share more of the structural details if it would be useful.

My first week as a CIO I sat in a budget meeting where the CFO asked me, in front of seven other executives, whether I understood how a P&L works.

I was 42. I had been in technology leadership for 18 years. I had an MBA. I had run IT at two prior companies. I had prepared for that meeting for three weeks.

He was not being cruel. He genuinely was not sure. In his experience CIOs were technical people who did not speak his language and who needed to have things explained to them. That was the CIO he had worked with before me and he was not wrong about that one.

I was so angry in that moment that I almost responded in a way that would have ended the meeting. Instead I took a breath and said, yes, and I have a question about the way we are allocating depreciation on the new facility. Then I asked the actual question I had come prepared to ask.

He answered it. We moved on. I did not get the budget I wanted that day but I got something more important. He treated me differently in the next meeting. And the one after that. By month four I had enough trust with him to push back on the CEO on an infrastructure spend decision, and he backed me up.

I have thought about that meeting a lot over the years. What I learned from it was not that I had to prove I belonged. I already belonged. What I learned was that in every room you walk into as a new leader, someone has a model of what your role is and that model was built by the person who had the job before you. You are either going to accept that model or replace it. Both take work. Replacing it takes more.

I am not the first woman in a CIO seat to have a story like this. I will not be the last. What I tell the women I mentor who are stepping into these roles is that the anger is useful information but it is not a strategy. The strategy is answering the next question with so much substance that the premise of the first one becomes irrelevant.

Three years into the role now. Different CFO. Different conversation. But I still think about that first meeting when I walk into rooms where I am the only person who looks like me.