Last touch attribution gives 100% of the credit for a lead or booked job to the final touchpoint before conversion. For service businesses with 7 to 14 day sales cycles, that final click is usually a branded search or a return visit. Last touch hides the channels that found the customer, built trust and earned the call.
Picture a homeowner with a failing water heater. On Monday they see your Google ad and don't click. On Wednesday they read your blog post on tank versus tankless. On Friday they check your reviews. The following Tuesday they search your company name, click the top result and fill out your quote form. A last touch report says branded search won that job. In fact branded search answered a phone call that three other channels set up. If you cut budget based on that report, you cut the channels that started the job.
TruLata is an AI marketing platform for service businesses. It runs search, ads, content, outbound and email from one screen, and it counts leads from the business's own forms rather than from ad platform estimates. This piece explains where last touch attribution breaks for local service companies and what it costs you. It also covers how to start measuring attribution in a way you can act on.
What is last touch attribution, exactly?
Last touch attribution is a single touch marketing attribution model. It assigns the entire value of a conversion to the most recent interaction a customer had before they converted. If someone clicks a Facebook ad, then an email, then a paid search ad, and then submits your form, the paid search ad gets full credit. The ad and the email get nothing.
The model rests on one assumption: the last thing a customer did was the thing that convinced them. That assumption can hold for impulse purchases. It rarely holds for a roof replacement, a remodel, a recurring cleaning contract or an HVAC system. Homeowners compare quotes, read reviews, ask neighbors and take their time on those decisions.
Why is last touch the default almost everywhere?
Last touch is the easiest model to implement. It only needs one data point: where the visitor came from in the session in which they converted. Analytics tools, ad platforms and form plugins can all capture that with almost no setup. It also produces clean, confident numbers, and people trust clean numbers more than they should.
How is it different from first touch attribution?
First touch attribution is the mirror image. It gives all the credit to the first known interaction. First touch overvalues discovery channels, and last touch overvalues closing channels. Both are single touch models, and both ignore everything in the middle. For a service business, the middle is where most of the decision happens.
Why does last touch attribution fail for service businesses with long sales cycles?
The longer and more considered the purchase, the less the final click explains. A 7 to 14 day decision window gives a homeowner time for several separate visits. Those visits come through different channels, on different devices, often days apart. Last touch looks at only the final visit.
The final click is usually a navigation step, not a decision
Look at how most service leads actually arrive. Many come through a branded search ("Smith Plumbing"), a direct visit or a click from a saved email. Each of these means the customer already knew who you were. Branded search did not convince anyone. It was the route they took to a decision they had already made. Last touch attribution still records it as the winning channel, month after month.
Awareness channels look like they produce nothing
Some channels introduce you to a homeowner who isn't ready to book. Local service ads, social posts, video, blog content, and AI answer engine citations all fall in this group. Under last touch, these channels show few or no conversions. On paper they look like waste. In reality they fill the pipeline that your branded search and retargeting later "close."
Offline and untracked touchpoints disappear entirely
Service businesses have touchpoints that never show up in a click path:
- A yard sign on a neighbor's lawn
- A truck wrap seen at a stoplight
- A referral from a past customer
- A review read on a map listing without clicking through
Last touch has no way to see any of these. It assigns their influence to whatever digital click happened last.
Attribution windows cut the journey short
Ad platforms only credit conversions that happen within a set window after a click or view. When a homeowner's decision runs past that window, the platform loses the connection. Shorter windows lose it sooner. Ad platforms also tend to claim credit generously for conversions inside their own window. You can end up with two or three platforms each claiming the same job, while your actual form count tells a different story.
What does relying on last touch attribution actually cost you?
The cost isn't a reporting inaccuracy you can shrug off. It shows up in budget decisions, and each bad decision compounds over the following months.
You defund the channels that create demand
Last touch makes upper funnel channels look unprofitable, so they're usually the first thing cut. The effect is delayed. Branded search volume and direct traffic hold steady for a few weeks because earlier awareness is still working through the pipeline. Then lead volume falls. By that point, the cause is several decisions back and hard to identify.
You overpay for credit that bottom funnel channels did not earn
Branded search and retargeting look like your most efficient channels under last touch, so they get more budget. But they mostly capture demand you already created. Spending more there often buys back customers who would have found you anyway, which makes cost per lead look better than the real return.
You optimize ads toward the wrong signal
Ad platforms optimize toward the conversions you report back to them. If the reported conversions come from a last touch view, platforms learn to chase people who are already about to convert. That narrows your reach exactly when you need to reach new homeowners in your service area.
You can't tell a good month from a lucky one
A single touch view can't separate a channel that is building next month's pipeline from one that is harvesting last month's. Without that distinction, a slow month looks like a channel problem when it may be a timing problem. A strong month may just be a delayed result of work you've since stopped funding.
Is last touch attribution ever the right marketing attribution model?
Yes, in narrow cases. Last touch is reasonable when the sales cycle is very short and there are few touchpoints. Emergency services are a good example. A homeowner with a burst pipe at 2 a.m. searches, clicks and calls within minutes. In that situation the last click and the only click are often the same thing.
Last touch is also a useful lens for one specific question: which channels are good at converting people who are already interested? The problem only starts when you treat that partial answer as the full picture of where leads come from.
Most service businesses have both kinds of jobs: fast emergency calls and slower, considered projects. If you lump them together under one last touch report, the considered jobs, which are usually the higher value ones, get misattributed.
How does multi touch attribution fix the problem?
Multi touch attribution spreads credit for a conversion across every recorded touchpoint, so channels that introduced and nurtured the customer get counted. The common approaches are:
- Linear: equal credit to every touchpoint. It's simple and fair, but it treats a passing glance the same as a deciding visit.
- Time decay: more credit to touchpoints closer to conversion. This is often a sensible fit for a 7 to 14 day cycle.
- Position based (U shaped): heavier credit to the first and last touches, with the remainder split across the middle. It recognizes both discovery and close.
- Data driven: credit weighted by observed patterns across many conversions. It needs enough volume to be reliable.
Multi touch has a trap of its own. It's easy to end up crediting every channel generously until the totals add up to more jobs than you actually booked. We cover that failure mode in multi touch attribution for service businesses. Whichever model you use, the total should reconcile to a real lead count.
How should a service business start measuring attribution properly?
You don't need an enterprise data team to get past last touch. You need a reliable lead count, consistent source capture and a habit of reading the whole path. Here is a practical sequence.
1. Count leads from your own forms, not the ad platform
Your source of truth should be the forms and calls that actually reached your business. Ad platform conversion counts are estimates, and they overlap across platforms. Start from lead tracking counted from your own forms, then attach sources to each real lead. That way every attribution model you test reconciles to the same number.
2. Capture first touch and last touch on every lead
Store the original landing source and referrer when someone first visits, and keep it through to form submission. With first and last touch side by side on each lead, you can immediately see how often the closing channel differs from the discovering channel. For most service businesses the two differ often.
3. Tag every link you control
Use consistent UTM parameters on emails, social posts, review requests, QR codes on yard signs and truck wraps, and outbound messages. Untagged links collapse into "direct" traffic, and direct traffic is where attribution goes to die.
4. Ask the customer
Add a required "How did you hear about us?" field to your quote form, and have your team ask again on the phone. Self reported answers are imperfect, but they catch referrals, signs, trucks and word of mouth that no click path will ever record. Compare those answers with your tracked sources to find the gaps.
5. Close the loop to booked jobs
A lead isn't revenue. Mark each lead's outcome (booked, lost, no response) so you can see which source paths produce jobs, not just form fills. Speed matters here too, because a lead that waits hours for a reply rarely books. Our breakdown of why your first reply determines your close rate explains why outcome data and response time belong in the same view.
6. Read channels together, on fresh data
Attribution analysis falls apart when search data, ad spend, form leads and job outcomes live in five tabs, each pulled on a different day. Put them on one screen, stamped with when the data was pulled, so you can see that a dip in branded search followed a cut in awareness spend. Live marketing data, refreshed when you open it makes that kind of comparison possible without exporting spreadsheets.
7. Judge channels over the full cycle, not the week
With a 7 to 14 day decision window, a one week report will always undercount awareness channels. Review channel performance over at least two to four weeks. Watch for lagged effects before you cut anything.
What should you look for in your attribution reporting?
Once you have real lead counts and full source paths, a few questions become answerable:
- What share of booked jobs started on a channel other than the one that closed them?
- Which channels appear most often as a first touch on jobs that eventually booked?
- How many days and how many visits does a typical booked job take, by service type?
- Do platform reported conversions add up to more than your actual form leads? By how much?
- When you paused a channel last quarter, what happened to branded search and direct leads two to three weeks later?
These answers are what let you defend an awareness channel with evidence, or cut it with confidence. A last touch report can't answer any of them.
How does TruLata handle attribution for service businesses?
TruLata brings search, ads, content, outbound, email and leads into one view. It counts every lead from your own forms and attaches its source and outcome. When you open your TruLata dashboard, it refreshes and shows when each data source was last pulled. You always know how fresh the numbers are. Ad spend and cost per lead sit beside the real lead count, so platform claims can be checked against what actually arrived.
The goal isn't a more complicated model. It's a more honest one: a lead count you trust, sources you can trace, and outcomes that show which channels actually closed jobs.
Ready to see what actually closed your last ten jobs?
If your reports say branded search wins every month, the report is probably telling you how customers arrived, not why they chose you. Open the TruLata live demo to see leads counted from real forms, with sources and outcomes side by side. You can also sign up at login.trulata.com/signup and connect your own accounts. Start making budget decisions on the whole journey, not just the final click.
