Which State Is Your Digital Project In?
A running digital job sits in one of four states, and each needs a different intervention. The figures show why that matters: consulting estimates put failure at around 70 percent.
This piece is a status map. First it locates which state your project is in, then it names the single intervention for that state. A correct intervention applied in the wrong state does not work either. Diagnosis comes before the move.
What Is on the Agenda?
BU BÖLÜMÜN ÖZETİ
- The failure rate runs high
- The picture in Türkiye is similar
- Ownership decides
- From opinion to evidence
- Integration, not access
Five findings on the table.
The failure rate runs high
Estimates sit around 70 percent. But the cause is not technical; the system works and nobody uses it. Installation ends, the work stays the same.
The picture in Türkiye is similar
A third of executives report high failure in their own projects. Since it is self-reported, that figure is a floor. The real picture may be heavier.
Ownership decides
The maturity report puts scaling a few initiatives with clear ownership first. Unowned work does not advance. Writing a name takes five minutes.
From opinion to evidence
The same report recommends basing decisions on a measure. The distance between “I think it’s going well” and “this number moved” sits here.
Integration, not access
Reaching technology is easy. Connecting it to how decisions get made is hard. The separation forms exactly there. The same tool lands differently in two businesses.
Why Is the Question Being Asked Now?
BU BÖLÜMÜN ÖZETİ
- Project counts rose
- Nobody asks about status
- Intervention replaces diagnosis
- Closure never gets declared
Four reasons.
Project counts rose
As entry costs fell, more jobs got started. But the capacity to finish them stayed the same. That gap accumulates as half-done work; a year later the list is full and the results are empty. Nobody knows what state each job is in.
Nobody asks about status
Interest runs high at the start and then the question stops getting asked. Work that goes unasked about does not advance. Months later it sits where it was, and restarting becomes harder.
Intervention replaces diagnosis
When a project stalls, something gets tried immediately. A new tool, new training, another meeting. An intervention made without knowing the state does not hold; energy gets spent and nothing changes. Three months on you stand in the same place.
Closure never gets declared
A finished project does not get closed and a stalled one does not get stopped. Both stay on the list. The list looks full while the inside is empty, and resources keep draining. Stopping a stopped job also requires a decision.
What Is Wrong?
BU BÖLÜMÜN ÖZETİ
- Applying the same fix to every blockage
- Counting a stalled project as running
- Counting installation as completion
- Determining the state by guesswork
Four diagnostic errors waste the intervention.
Applying the same fix to every blockage
Training does not help a project stalled by lack of ownership. Assigning an owner does not help one stalled by oversized scope. The state has to be known first, and the right fix fails in the wrong state.
Counting a stalled project as running
It sits on the list, gets mentioned occasionally, and nobody says it has ended. That project has actually stopped. But because the records show it running, it keeps consuming resources. Closure is a decision, and it never gets made.
Counting installation as completion
The project gets marked done once the system goes live. Yet an unused system is not a finished project. Without that distinction the success rate gets calculated wrongly and the real picture never appears.
Determining the state by guesswork
“How is it going” gets answered by impression, and impressions lean optimistic. Without looking at a measure, the blockage gets noticed late. An early blockage is easy to clear, which is why the success measure gets written at the start.
The Real Mechanism
BU BÖLÜMÜN ÖZETİ
- State 1: not started
- State 2: running
- State 3: stalled
- State 4: finished but unused
Every digital job sits in one of four states.
State 1: not started
It was discussed, decided, and nothing happened. The symptom: no name and no date. The intervention is single — assign an owner. Nothing else changes this state; not a new meeting, not a new tool.
State 2: running
There is an owner, a weekly check, visible progress. No intervention is needed here. The only thing to do is protect the rhythm and stay out of the way. Every new request added to a running project slows it, and protecting beats improving.
State 3: stalled
It started but is not advancing. The cause is usually one of three: the scope is too large, there is no owner, or the goal is unclear. The intervention follows the cause. Shrinking the scope is the one that works most often.
State 4: finished but unused
Installation complete, system up, nobody logging in. The most common and most expensive state. The intervention is not technical: close the old route and restart with a single job.
Who Is Affected, and How?
BU BÖLÜMÜN ÖZETİ
- The business with unstarted projects
- The business with many running projects
- The business with a stalled project
- The business with an unused system
The same map reads differently in four situations.
The business with unstarted projects
The most common state and the easiest to resolve. Usually a few ideas got discussed and none began. The trap here is adding new ideas; the list grows and nothing moves. Writing a name against one of them changes the picture, and the rest stay in the queue without disappearing.
The business with many running projects
It looks healthy from outside. But with more than three jobs running at once, all of them slow. The intervention here is subtraction rather than addition: put some into waiting. That is also the part people find hardest, since stopping a job feels worse than starting one.
The business with a stalled project
The state where most energy gets burned. Meetings happen, things get discussed, nothing advances. Every intervention made before finding the cause consumes more energy. Three questions usually surface the cause on the first attempt.
The business with an unused system
The most deceptive state, because the records show success. The project finished, the invoice was paid, nobody uses it. The intervention here is not new investment; it is bringing the existing system back with a single job.
Decision Order
BU BÖLÜMÜN ÖZETİ
- One: determine the state
- Two: apply the intervention for that state
- Three: make one intervention only
- Four: record the state
Four steps, in order.
One: determine the state
Three questions suffice. Is there a name? Has anything moved in the past two weeks? Is the system actually being used? The answers show the state directly.
Two: apply the intervention for that state
Not started, write a name. Stalled, shrink the scope. Unused, close the old route. Running, do not touch it — the hardest intervention of the four.
Three: make one intervention only
Do not try three things at once; you will not know which worked. Make one change, wait two weeks, then look again.
Four: record the state
Two lines suffice. Which state it was in and what was done. Three months later that note prevents the same discussion.
Where to Start?
BU BÖLÜMÜN ÖZETİ
- List the open jobs
- Put the three questions to each
- Pick the worst one
- Write the single intervention
The first session, half an hour.
List the open jobs
How many digital jobs exist? Write them with their names; there are usually more than expected.
Put the three questions to each
Owner, progress, usage. Combine the three answers and the state appears by itself.
Pick the worst one
An unused system usually turns up. Start there, since it is also the most expensive state.
Write the single intervention
What will be done and who will do it? One sentence is enough.
What Not to Do?
BU BÖLÜMÜN ÖZETİ
- Intervening without diagnosis
- Making several interventions at once
- Touching a running project
- Not declaring the state
Four application errors.
Intervening without diagnosis
A move made without knowing the state usually picks the wrong one. Training for an unowned project, a new tool for a scope problem. Both burn energy without producing a result, and they leave a weariness behind.
Making several interventions at once
Try three things together and you cannot tell which worked. The same uncertainty carries into the next project. Trying them one at a time makes learning possible.
Touching a running project
Adding a new request to work that is going well is the most frequent error. The scope grows, the timeline stretches, the team tires. A running project should be protected; improvement waits its turn as a separate job.
Not declaring the state
A stalled project that never gets stopped stays on the list and drains resources. Closure is a decision too. Unsaid, nobody notices, and the list keeps swelling.
What to Watch?
BU BÖLÜMÜN ÖZETİ
- Open job count
- Unowned job count
- Last progress date
- Usage rate
Four indicators suffice.
Open job count
How many projects run at once? Past three, all of them slow and none finishes.
Unowned job count
How many projects have a blank name field? That number is a direct stagnation indicator.
Last progress date
When did each project last move? Past two weeks it counts as stalled.
Usage rate
How many finished projects are genuinely used? That figure is your real success rate.
How Does This Period End?
BU BÖLÜMÜN ÖZETİ
- Half-done work will accumulate
- Those who diagnose will separate
- Usage rate will become the real measure
Three separations will show.
Half-done work will accumulate
Entry costs keep falling, so more projects will get started while finishing capacity stays flat. The half-done work will pile up and become unmanageable. A clear-out will eventually be forced, and hasty clear-outs produce errors of their own.
Those who diagnose will separate
A business that intervenes knowing the state will finish more work with less energy. One that acts on guesswork will keep hitting the same blockages. Learning will not accumulate either.
Usage rate will become the real measure
The conversation will move from how many projects got started to how many get used. A business adopting that measure sees its own picture. The others keep looking at sector averages, which describe someone else entirely.
A Solid Digital Foundation
BU BÖLÜMÜN ÖZETİ
- The open job list
- The status note
- The intervention log
- The usage list
Four documents, each one page.
The open job list
Which projects run, who owns them, when did they last move? Three columns, updated monthly.
The status note
Which state is each project in? One of the four gets written and the intervention follows from it.
The intervention log
What was done and what came of it? Two lines prevent the repeat.
The usage list
Which finished projects are genuinely used? Your real success rate shows here.
Frequently Asked Questions
Sık Sorulan Sorular
Three questions usually surface it. Does the team know how to do it? Is the scope small enough to finish in two weeks? Is a name written down? The first no shows the cause. If no answer comes, ask the team directly — they usually know, but nobody has asked. Put without blame, the real reason appears in the first five minutes.
One experiment is worth running first. Rather than trying to activate the whole system, move a single job onto it and close the old route for that job only. Look again after three weeks. If it works, the system was never the problem — the transition simply never happened. If it does not, the decision becomes clear and you can close it. The experiment costs three weeks and no new spending.
There is no fixed number, but a practical limit exists: how many can you follow weekly? Every running job wants half an hour a week. Three jobs make an hour and a half, which is the ceiling in most small businesses. More than that goes unfollowed, and unfollowed work stops. Set the number by your capacity rather than your enthusiasm.
The criterion is the end of the transition, not the end of the installation. Three questions: has the old method closed, does the team use it regularly, did the target number move? All three yes means the project is done. One no means it is still open with only the technical part complete. Without that distinction the success rate looks higher than it is and the real picture gets lost.